Human Systems Engineering and Design PDF

This book focuses on novel design and systems engineering approaches, including theories and best practices, for promoting a better integration of people and engineering systems. It covers a range of hot topics related to: development of activity-centered and user-centered systems; interface design and human-computer interaction; usability and user experience; cooperative, participatory and contextual models; emergent properties of human behavior; innovative materials in manufacturing, and many more. Particular emphasis is placed on applications in sports, healthcare, and medicine. The book, which gathers selected papers presented at the 1st International Conference on Human Systems Engineering and Design: Future Trends and Applications (IHSED 2018), held on October 25-27, 2018, at CHU-Université de Reims Champagne-Ardenne, France, provides researchers, practitioners and program managers with a snapshot of the state-of-the-art and current challenges in the field of human systems engineering and design.

102 downloads 2K Views 122MB Size

Report

Download pdf

Recommend Stories

Empty story

Idea Transcript

Advances in Intelligent Systems and Computing 876

Tareq Ahram Waldemar Karwowski Redha Taiar Editors

Human Systems Engineering and Design Proceedings of the 1st International Conference on Human Systems Engineering and Design (IHSED2018): Future Trends and Applications, October 25–27, 2018, CHU-Université de Reims Champagne-Ardenne, France

Advances in Intelligent Systems and Computing Volume 876

Series editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: [email protected]

The series “Advances in Intelligent Systems and Computing” contains publications on theory, applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all disciplines such as engineering, natural sciences, computer and information science, ICT, economics, business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the areas of modern intelligent systems and computing such as: computational intelligence, soft computing including neural networks, fuzzy systems, evolutionary computing and the fusion of these paradigms, social intelligence, ambient intelligence, computational neuroscience, artiﬁcial life, virtual worlds and society, cognitive science and systems, Perception and Vision, DNA and immune based systems, self-organizing and adaptive systems, e-Learning and teaching, human-centered and human-centric computing, recommender systems, intelligent control, robotics and mechatronics including human-machine teaming, knowledge-based paradigms, learning paradigms, machine ethics, intelligent data analysis, knowledge management, intelligent agents, intelligent decision making and support, intelligent network security, trust management, interactive entertainment, Web intelligence and multimedia. The publications within “Advances in Intelligent Systems and Computing” are primarily proceedings of important conferences, symposia and congresses. They cover signiﬁcant recent developments in the ﬁeld, both of a foundational and applicable character. An important characteristic feature of the series is the short publication time and world-wide distribution. This permits a rapid and broad dissemination of research results.

Advisory Board Chairman Nikhil R. Pal, Indian Statistical Institute, Kolkata, India e-mail: [email protected] Members Rafael Bello Perez, Universidad Central “Marta Abreu” de Las Villas, Santa Clara, Cuba e-mail: [email protected] Emilio S. Corchado, University of Salamanca, Salamanca, Spain e-mail: [email protected] Hani Hagras, University of Essex, Colchester, UK e-mail: [email protected] László T. Kóczy, Széchenyi István University, Győr, Hungary e-mail: [email protected] Vladik Kreinovich, University of Texas at El Paso, El Paso, USA e-mail: [email protected] Chin-Teng Lin, National Chiao Tung University, Hsinchu, Taiwan e-mail: [email protected] Jie Lu, University of Technology, Sydney, Australia e-mail: [email protected] Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexico e-mail: [email protected] Nadia Nedjah, State University of Rio de Janeiro, Rio de Janeiro, Brazil e-mail: [email protected] Ngoc Thanh Nguyen, Wroclaw University of Technology, Wroclaw, Poland e-mail: [email protected] Jun Wang, The Chinese University of Hong Kong, Shatin, Hong Kong e-mail: [email protected]

More information about this series at http://www.springer.com/series/11156

Tareq Ahram Waldemar Karwowski Redha Taiar •

Editors

Human Systems Engineering and Design Proceedings of the 1st International Conference on Human Systems Engineering and Design (IHSED2018): Future Trends and Applications, October 25–27, 2018, CHU-Université de Reims Champagne-Ardenne, France

123

Editors Tareq Ahram Institute for Advanced Systems Engineering University of Central Florida Orlando, FL, USA

Redha Taiar Université de Reims Champagne-Ardenne Reims, France

Waldemar Karwowski University of Central Florida Orlando, FL, USA

ISSN 2194-5357 ISSN 2194-5365 (electronic) Advances in Intelligent Systems and Computing ISBN 978-3-030-02052-1 ISBN 978-3-030-02053-8 (eBook) https://doi.org/10.1007/978-3-030-02053-8 Library of Congress Control Number: 2018957480 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

This volume, entitled Human Systems Engineering and Design, aims to provide a global forum for presenting and discussing novel design and systems engineering approaches, tools, methodologies, techniques, and solutions for integrating people, concepts, trends and applications in all areas of human endeavor in industry, economy, government, and education. Such applications include, but are not limited to, energy, transportation, urbanization and infrastructure development, digital manufacturing, social development, human health, sustainability, a new generation of service systems, as well as safety, risk assessment, healthcare, and cybersecurity in both civilian and military contexts. Indeed, rapid progress in developments in cognitive computing, modeling, and simulation, as well as smart sensor technology, will have a profound effect on the principles of human systems engineering and design at both the individual and societal levels in the near future. This book focuses on advancing the theory and applications for integrating human requirements as part of an overall system and product solution, by adopting a human-centered design approach that utilizes and expands on the current knowledge of systems engineering supported by cognitive software and engineering, data analytics, simulation and modeling, and next-generation visualizations. This interdisciplinary approach will also expand the boundaries of the current state of the art by investigating the pervasive complexity that underlies the most profound design problems facing contemporary society today. This book also presents many innovative studies of systems engineering and design with a particular emphasis on the development of technology throughout the lifecycle development process, including the consideration of user experience in the design of human interfaces for virtual, augmented, and mixed reality applications. Reflecting on the above-outlined perspective, the papers contained in this volume are organized into eight unique research tracks with a total of eighteen sections, including: IHSED 1: Human-Centered Design Section 1: Human-Centered Design and User Experience Section 2: User Interface Design Applications and Human Systems Integration v

vi

Preface

Section 3: Virtual Reality and Usability Evaluation Applications Section 4: Human–Machine Collaboration Section 5: Design Evaluation, Learning, and Assessment IHSED 2: Innovative Materials for Product Development Section 6: Innovative Materials for Product Development IHSED 3: Systems Design and Human Diversity Section 7: Systems Design and Human Diversity Applications Section 8: Artiﬁcial Intelligence and Intelligent System Design IHSED 4: Safety Engineering and Systems Complexity Section Section Section Section Section

9: Safety Engineering and Complex Systems Design 10: Occupational Safety and Risk Assessment 11: Transportation System Design and Safety Engineering 12: Safety Evaluation and Ergonomic Risks 13: Organizational and Strategic Interventions

IHSED 5: Sports Design and Sports Medicine Section 14: Sports Design and Sports Medicine IHSED 6: Biomechanics, Health Disease, and Rehabilitation Section 15: Biomechanics, Evaluation, and Assessment Section 16: Health Management and Rehabilitation IHSED 7: Human Cyber-Physical Systems Interactions Section 17: Human Cyber-Physical Systems Interaction Applications IHSED 8: Business, Design, and Technology Section 18: Business, Design, and Technology We would like to extend our sincere thanks to Hervé Quinart, CHU-Université de Reims Champagne-Ardenne, France, for his support. Our appreciation also goes to the members of the Scientiﬁc Program Advisory Board who have reviewed the accepted papers that are presented in this volume. We hope that this book, which presents the current state of the art in human systems engineering and design, will be a valuable source of both theoretical and applied knowledge, enabling the human-centered design and applications of a variety of products, services, and systems for their safe, effective, and pleasurable use by people around the world. October 2018

Tareq Ahram Waldemar Karwowski Redha Taiar

Contents

IHSED 1: Human-Centered Design and User Experience Applying Human-Centered Design and Human-Machine Integration Techniques to Solve Key Healthcare Problems . . . . . . . . . . . . . . . . . . . . Neil Gomes and Viraj Patwardhan Subjective Evaluation of EV Sounds: A Human-Centered Approach . . . Verena Wagner-Hartl, Bernhard Graf, Markus Resch, and Paco Langjahr

3 10

Sequential Recognition Rate and Latency of Frequency-Based Tactons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ricardo Jimenez and Ana Maria Jimenez

16

Bringing It Together: Three Approaches to Combine Agile Software Development and Human-Centered Design . . . . . . . . . . . . . . . . . . . . . . Michael Minge and Antonia Föhl

21

User-Centered-Design Approach to Evaluate the User Acceptance of Seating Postures for Autonomous Driving Secondary Activities in a Passenger Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sibashis Parida, Sai Mallavarapu, Sylvester Abanteriba, Matthias Franz, and Wolfgang Gruener

28

User Evaluation of Industry 4.0 Concepts for Worker Engagement . . . . Susanna Aromaa, Marja Liinasuo, Eija Kaasinen, Michael Bojko, Franziska Schmalfuß, Konstantinos C. Apostolakis, Dimitrios Zarpalas, Petros Daras, Cemalettin Özturk, and Menouer Boubekeuer FatigueWatcher: Interactive Fatigue Detection for Personal Computer and Mobile Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ayumu Tanaka, Takashi Yokogawa, and Hiroaki Tobita

34

41

vii

viii

Contents

Introduction of Service Design in a Public Hospital’s Medical Oncology Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carlos Romero-Piqueras, Jorge Sierra-Pérez, and Eduardo Manchado-Pérez Early Validation of User Needs in Concept Development: A Case Study in an Innovation-Oriented Consultancy . . . . . . . . . . . . . . Marianne Kjørstad, Kristin Falk, Gerrit Muller, and José Pinto Introduction of User Experience into the Design of Academic Services at University Centre of Defence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jorge Sierra-Pérez, Carlos Romero-Piqueras, Myriam Cilla, Silvia Guillén-Lambea, and Marcos Pueo Research on Aesthetics Degree Evaluation Method of Product Form . . . Ming Li and Jie Zhang

48

54

61

68

Sensemaking on the Bridge: A Theoretical Approach to Maritime Information Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brit-Eli Danielsen

76

Examining Cloud Computing Applications from the Perspectives of Privacy and Uniﬁed Theory of Acceptance and Use of Technology . . Tihomir Orehovački, Darko Etinger, and Snježana Babić

82

Rob’Autism Project: Being Active in Social Interactions: The Robot-Extension Paradigm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rénald Gaboriau, Sophie Sakka, Didier Acier, and Dimitri Delacroix

88

Thermal Comfort Assessment: A Study Towards Workers’ Satisfaction in Metal Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Norma de Melo Pinto and Kazuo Hatakeyama

95

Abductive Thinking, Conceptualization, and Design Synthesis . . . . . . . . 101 Dingzhou Fei Is Truth Contextual? The Browsing Purpose, the Availability of Comparable Material, and the Web Content Credibility Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Katarzyna Abramczuk, Michał Kąkol, Radosław Nielek, and Cezary Biele All Doors Lead to the Kitchen – Sustainability and Wellbeing Challenges in a Shared Centrepiece of Living . . . . . . . . . . . . . . . . . . . . 111 Soﬁe Andersson and Ulrike Rahe User as Customer: Touchpoints and Journey Map . . . . . . . . . . . . . . . . 117 Camila Bascur, Cristian Rusu, and Daniela Quiñones Enhancing User Experience with Embodied Cognition . . . . . . . . . . . . . 123 SuKyoung Kim

Contents

ix

A Study on the Effect of Human Factor for Atypical Design in the Architectural Design Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Yungil Lee E-material Creating and Formatting Application . . . . . . . . . . . . . . . . . . 135 Kristine Mackare, Anita Jansone, and Maksims Žigunovs The User Experience of 3D Scanning Tangible Cultural Heritage Artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Chee Weng Khong and Muhammad Asyraf Mhd Pauzi International Museums and Transcultural Impact on Gulf States: The Louvre Abu Dhabi as a Case Study . . . . . . . . . . . . . . . . . . . . . . . . 148 Mohamed El Amrousi, Mohamed Elhakeem, and Evan Paleologos Live Action Carnavalia: A Case Study of a Process for User Engagement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Vladimir Barros and Breno Carvalho Incorporating Human Factors in In-Plant Milk Run System Planning Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Aleksandra Polak-Sopinska Design and Evaluation of an Innovative Assisting Device for Improving Blood Circulation in Osteoarthritis . . . . . . . . . . . . . . . . . 167 Yan-Chun Lin and Fong-Gong Wu Evaluation of Matching Degree Between Touch Gestures and User Mental Model Based on Event-Related Potential . . . . . . . . . . . . . . . . . . 173 Ningyue Peng, Jing Ma, and Chengqi Xue Co-creation Workshop Oriented to the Autonomous Elderlies in Chile: Unveiling the Experience of the Participants . . . . . . . . . . . . . . 185 Juan Carlos Briede-Westermeyer, Cristhian Pérez-Villalobos, Javiera Ortega-Bastidas, and Isabel Leal-Figueroa Do Design Outcomes Get Inﬂuenced by Type of User Data? An Experimental Study with Primary and Secondary User Research Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Abhishek Dahiya and Jyoti Kumar Medical Device Design Challenges Based on Users Hierarchy and Their Correlation with Illness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Fabiola Cortes-Chavez, Maria Giovanna-Trotta, Paulina Manzano-Hernandez, Alberto Rossa-Sierra, and Gabriela Duran-Aguilar

x

Contents

Putting Chatbots to the Test: Does the User Experience Score Higher with Chatbots Than Websites? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Amélie Beriault-Poirier, Sandrine Prom Tep, and Sylvain Sénécal IHSED 1: User Interface Design Applications and Human-Systems Integration Making Multi-team Systems More Adaptable by Enhancing Transactive Memory System Structures – The Case of CDM in APOC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Dirk Schulze Kissing, Carmen Bruder, Nils Carstengerdes, and Anne Papenfuss Deﬁning User Needs for a New Sepsis Risk Decision Support System in Neonatal ICU Settings Through Ethnography: User Interviews and Participatory Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Richard Harte, Leo R. Quinlan, Evismar Andrade, Enda Fallon, Martina Kelly, Paul O’Connor, Denis O’Hora, Patrick Pladys, Alain Beucheé, and Gearoid ÓLaighin FLOW: A Software Application Designed to Help Older Adults Build Distance Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Wonsil Jang, Stephen Gilbert, and Sunghyun Kang Learning from Human Behavior to Improve Preventative Health Information Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Remberto Martinez, Marcos Tong, Luis Diago, and Jaana Lindstrom Prototyping a User Interface for a New Sepsis Risk Decision Support System Using Participatory Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Richard Harte, Leo R. Quinlan, Evismar Andrade, Enda Fallon, Martina Kelly, Paul O’Connor, Denis O’Hora, Patrick Pladys, Alain Beucheé, and Gearoid ÓLaighin A Proposal for an Affective Design and User-Friendly Voice Agent . . . . 249 Heesung Park, Jeongpyo Lee, Sowoon Bae, Daehee Park, and Yenah Lee Methodologies for the Design of ATM Interfaces: A Systematic Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Joel Aguirre, Arturo Moquillaza, and Freddy Paz Design and Development of an Image-Based System to Facilitate Reading Comprehension of Chinese Classic Literature . . . . . . . . . . . . . 263 Tung-En Chien and Shelley Shu-Ching Young Human Factors Integrated System Validation in the Nuclear Power Plant Main Control Room . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Xiaoyang Ming, Guangwei Yu, and Shiguang Deng

Contents

xi

The Functional Requirement Analysis in Nuclear Power Plant Human System Interface Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Shiguang Deng, Xiaoyang Ming, and Guangwei Yu A Preliminary Study on Color and Grayscale Images Object Recognition and Scene Classiﬁcation Tasks on Amazon Mechanical Turk Crowdsourcing Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Aimee Yun-Fang Lin, Shelley Shwu-Ching Young, Harrison Pang-Sheng Lai, and Danna Gurari Design of an Integrative System for Conﬁgurable Exergames Targeting the Senior Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 Teresa Paulino, John Muñoz, Sergi Bermudez, and Mónica S. Cameirão Research Based on Product Design: The Example of Spimi Skin Detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Cao Ying and Cao Jing Development of a Prototype for Non-contact Keyboard . . . . . . . . . . . . . 299 Yasushi Kambayashi, Keita Ueda, Masanari Kasahara, Tatsumi Kusano, and Munehiro Takimoto IHSED 1: Virtual Reality and Usability Evaluation Applications Augmented Reality in the Context of Naval Operations . . . . . . . . . . . . . 307 Mário Marques, Filipe Elvas, Isabel L. Nunes, Victor Lobo, and Anacleto Correia Strategies and Metrics for Evaluating the Quality of Experience in Virtual Reality Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314 Xiangjie Kong and Yuqing Liu An Initial Design of the Mei Garden Augmented Reality Tour-Guide System Based on the Needs Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 Hung-Yeh Lin and Shelley Shwu-Ching Young Designing Virtual Reality to Enhance Spatiality . . . . . . . . . . . . . . . . . . . 326 Youngil Cho and Suehusa Mamoru UNICAP Virtual: User Experience for a VR Application in Brazilian University . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Christianne Soares Falcão and Breno Carvalho State of the Art and Future Trends in the Usability of Patient Monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 Evismar Andrade, Leo R. Quinlan, Richard Harte, Dara Byrne, Enda Fallon, Martina Kelly, Paul O’Connor, Denis O’Hora, Michael Scully, John Laffey, Patrick Pladys, Alain Beuchée, and Gearoid ÓLaighin

xii

Contents

Usability Evaluation of a Public Transport Mobile Ticketing Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Daniel Meireles de Amorim, Teresa Galvão Dias, and Marta Campos Ferreira Investigation of the Human Factors, Usability and User Experience of Patient Monitors used in a Hospital Setting . . . . . . . . . . . . . . . . . . . . 352 Evismar Andrade, Leo R. Quinlan, Richard Harte, Dara Byrne, Enda Fallon, Martina Kelly, Paul O’Connor, Denis O’Hora, Michael Scully, John Laffey, Patrick Pladys, Alain Beuchée, and Gearoid ÓLaighin Usability Assessment as a Guide to Improve the System Design of a Corneal Topographer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358 Carlos Aceves-Gonzalez, Carlos D. de Leon-Zuloaga, Zuli T. Galindo-Estupiñan, and Citlali Díaz-Gutierrez Investigation of Usability Issues Through Physiological Tools: An Experimental Study with Tourism Websites . . . . . . . . . . . . . . . . . . . 365 Jyotish Kumar and Jyoti Kumar IHSED 1: Human-Machine Collaboration The Human-Tech Matrix: A Socio-Technical Approach to Evaluation of Automated Transport Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 Jonas Andersson, Tor Skoglund, and Niklas Strand Designing and Management of Intelligent, Autonomous Environment (IAE): The Research Framework . . . . . . . . . . . . . . . . . . . 381 Edmund Pawlowski, Krystian Pawlowski, Jowita Trzcielinska, and Stefan Trzcielinski Impact of Machine’s Robotisation on the Activity of an Operator in Picking Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Adrian Couvent, Mathieu Dridi, Nicolas Tricot, Christophe Debain, Mahmoud Almasri, Gil De Sousa, Gerald Chaloub, Marie Izaute, and Fabien Coutarel On the Lack of Pragmatic Processing in Artiﬁcial Conversational Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 Baptiste Jacquet, Olivier Masson, Frank Jamet, and Jean Baratgin Optimal Design of a Robotic Assistant Based on the Structural Study Using Finite Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 Graciela Serpa-Andrade, Luis Serpa-Andrade, Vladimir Robles-Bykbaev, and Irene Serpa-Andrade

Contents

xiii

Human-Autonomous Technology Interaction: A Systemic-Structural Activity Theory Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 Julian P. Vince and Gregory Z. Bedny IHSED 1: Design Evaluation, Learning and Assessment Research Design to Access the Mental Workload of Air Trafﬁc Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 Thorsten Mühlhausen, Thea Radüntz, André Tews, Hejar Gürlük, and Norbert Fürstenau Measuring Collaborative Emergent Behavior in Multi-agent Reinforcement Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422 Sean L. Barton, Nicholas R. Waytowich, Erin Zaroukian, and Derrik E. Asher Products as Mass Media: Entertainment vs. Ediﬁcation . . . . . . . . . . . . . 428 Del Coates Graphic Design Analysis Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 Elisabete Rolo Education and Training of Road Safety Auditors on the Implementation of Human Factors Principles in Safe Road Design . . . . 439 Sophia Vardaki and Evangelos Bekiaris Ergonomic Analysis in the Welding Laboratory of the Federal Institute of Paraíba – IFPB, João Pessoa Campus . . . . . . . . . . . . . . . . . 445 Amanda Ramos de Amorim and Aarão Pereira de Araújo Junior Overview of Empathetic Approaches to Design Inclusive Products . . . . 451 Maria Giovanna Trotta Munno, Luis Alberto Rosa Sierra, and Fabiola Cortes Chavez Assistive Devices for Lower Limbs Under Mechanism of Neuromodulation and Blood Circulation . . . . . . . . . . . . . . . . . . . . . . 457 Hui-I Yin and Fong-Gong Wu Packaging Design to Support Small Business Enterprises in the Republic of El Salvador . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 Alberto Rossa-Sierra, Maria Giovanna Trotta, Fabiola Cortes-Chavez, and Francisco González-Madariaga The Importance of Industrial Design in Medical Devices in the 21st Century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 Fanny Guadalupe Valdivia-Márquez, Pilar Hernandez-Grageda, Gabriela Durán-Aguilar, and Alberto Rossa-Sierra

xiv

Contents

IHSED 2: Innovative Materials for Product Development Prevention of Work: Related Musculoskeletal Disorders Using Smart Workwear – The Smart Workwear Consortium . . . . . . . . . . . . . . . . . . 477 Carl Mikael Lind, Leif Sandsjö, Naﬁse Mahdavian, Dan Högberg, Lars Hanson, Jose Antonio Diaz Olivares, Liyun Yang, and Mikael Forsman Smart Textiles and Their Role in Monitoring the Body’s Fitness and Medical Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484 Mohamed Eldessouki, Redha Taiar, Tareq Ahram, and Stanislav Petrik Quantifying Sense of Depth Towards Visual Texture Using Optics Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 Masaki Shimomura, Ruriko Kamesaka, Mugi Nishihara, Kei Matsuoka, Takamasa Yoshimura, Takeo Kato, and Yoshiyuki Matsuoka Factor Analysis of Synesthetic Perceptual Dimensions Using Aluminum Alloy Material Textures Surface in Industrial Products . . . . 498 Jialun Huang, Xiaozhou Zhou, Chengqi Xue, Lei Zhou, and Yafeng Niu Multi-objective Optimization Applied to the Bioclimatic Design of Dwellings with Ecomaterials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506 Jesús Rafael Hechavarría Hernández, Robinson Vega Jaramillo, and Boris Forero Fuentes Assistance System (AS) for Vehicles on Indian Roads: A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512 Neha Soni, Enakshi Khular Sharma, Narotam Singh, and Amita Kapoor IHSED 3: Systems Design and Human Diversity Applications Comfort Design in Human Robot Cooperative Tasks . . . . . . . . . . . . . . . 521 Alireza Changizi, Morteza Dianatfar, and Minna Lanz PlayCube: Designing a Tangible Playware Module for Human-Robot Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527 Vinicius Silva, Filomena Soares, João Sena Esteves, and Ana Paula Pereira Design for Seniors: A Case Study Based on Human Centric Lighting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 Aldo Deli, Massimo Di Nicolantonio, and Emilio Rossi Human Diversity and Organizational Culture . . . . . . . . . . . . . . . . . . . . 540 Barbara Mazur

Contents

xv

Predictors of Preference for the Activity-based Flexible Ofﬁce . . . . . . . . 547 Linda Rolfö, Helena Jahncke, Lisbeth Slunga Järvholm, Maria Öhrn, and Maral Babapour Affective Design Approach to Mobile Security Authentication . . . . . . . . 554 Daehee Park, Jaeyong Lee, Yenah Lee, and Scott Song An Experimental Study on Relationship Between Intellectual Concentration and Personal Mental Characteristics . . . . . . . . . . . . . . . . 560 Wakako Takekawa, Kimi Ueda, Shogo Ogata, Hiroshi Shimoda, Hirotake Ishii, and Fumiaki Obayashi Product Packaging Evaluation Through the Eyes of Elderly People: Personas vs. Aging Suit vs. Virtual Reality Aging Simulation . . . . . . . . 567 Christina Zavlanou and Andreas Lanitis Design for Innovative Development in Construction Industry: Proposal Based on Comparative Analysis . . . . . . . . . . . . . . . . . . . . . . . 573 Kazuo Hatakeyama An Evaluation Method for Intellectual Concentration Based on Concentration Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 Kimi Ueda, Shota Shimonaka, Hiroshi Shimoda, Hirotake Ishii, and Fumiaki Obayashi Application of the Principle of Conformal Symmetry in the Structure of Human Internal Organs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585 Galina Spirina Design of a Framework to Promote Physical Activity for the Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589 Alexandre Calado, Pedro Leite, Filomena Soares, Paulo Novais, and Pedro Arezes Construction of Multi-purpose Japanese Sign Language Database . . . . 595 Yuji Nagashima Product Function Analysis: Reducing Cost of Production Service Line with Work Teams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600 Velia Castillo-Pérez, Liliana Carrasco-Arméndariz, Mario Corral-Chacón, and Ramon Elizondo-Rios Systemic Approach for Inclusive Design of Low-Income Dwellings in Popular Settlements at Guayaquil, Ecuador . . . . . . . . . . . . . . . . . . . . 606 Boris Forero, Jesus Rafael Hechavarría Hernández, Silvia Alcivar, and Virginia Ricaurte

xvi

Contents

IHSED 3: Artiﬁcial Intelligence and Intelligent System Design VibroTac S: An Electronic Assistive Device for Blind and Visually Impaired People to Avoid Collisions . . . . . . . . . . . . . . . . . . . . . . . . . . . 613 Simon Schätzle, Thomas Hulin, and Benedikt Pleintinger Real Time Trafﬁc Incident Detection by Using Twitter Stream Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 620 Maryam Afzaal, Nazifa Nazir, Khadija Akbar, Sidra Perveen, Umer Farooq, M. Khalid Ashraf, and Zonia Fayyaz On Inverse Problem of Artiﬁcial Intelligence in System-Informational Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627 Nicolay Vasilyev, Vladimir Gromyko, and Stanislav Anosov Analysis on IoT Activities in Japanese Companies: Toward Innovation or Cost Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . 634 Yuriko Sawatani Ethical Problems of Introducing Artiﬁcial Intelligence into the Contemporary Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 Olga Burukina, Svetlana Karpova, and Nikolas Koro Systemic Analysis of Bioclimatic Design of Low-Income State-Led Housing Program “Socio Vivienda” at Guayaquil, Ecuador . . . . . . . . . 647 Santiago Dick, Jesús Rafael Hechavarría Hernández, and Boris Forero Methodology of Shaping the Agility of the Intelligent Autonomous Environment Management System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 652 Stefan Trzcielinski, Jowita Trzcielinska, Edmund Pawlowski, and Krystian Pawlowski IHSED 4: Safety Engineering and Complex Systems Design High-Hazard Complex Systems Design: HF Integration in Practice . . . . 661 Karen Priestman Developing Safety Competence Process for Vocational Students . . . . . . 668 Sari Tappura, Sanna Nenonen, and Noora Nenonen Capturing the Ups and Downs of Accidents’ Figures – The Portuguese Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675 Celina P. Leão, Susana Costa, Nélson Costa, and Pedro Arezes Experimental Investigations and Finite Element Modelling of the Vibratory Comportment of a Manual Wheelchair . . . . . . . . . . . . 682 Nadir Skendraoui, Fabien Bogard, Sébastien Murer, Fabien Beaumont, Boussad Abbes, Guillaume Polidori, Jean-Baptiste Nolot, Damien Erre, Serge Odof, and Redha Taiar

Contents

xvii

Remaining Useful Life as Prognostic Approach: A Review . . . . . . . . . . 689 Beata Mrugalska IHSED 4: Occupational Safety and Risk Assessment Are Teammate Trust and Conﬁdence Dissociable in Risk Intensive Human Machine Teaming? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 John G. Blitch and Anna D. Skinner Predicting the Level of Safety Performance Using an Artiﬁcial Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 705 Emmanuel Bannor Boateng, Manikam Pillay, and Peter Davis Augmented Reality for Health and Safety Training Program Among Healthcare Workers: An Attempt at a Critical Review of the Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 711 Anna Rita Corvino, Elpidio Maria Garzillo, Paola Arena, Arcangelo Ciofﬁ, Maria Grazia Lourdes Monaco, and Monica Lamberti Evaluation of Occupational Safety and Ergonomics in the Reuse and Recycling of Solid Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716 Hebert R. Silva Kaizen Approach for the Systematic Review of Occupational Safety and Health Procedures in Food Industries . . . . . . . . . . . . . . . . . . . . . . . 722 Renan Zocca, Tânia M. Lima, Pedro D. Gaspar, and Fernando Charrua-Santos The Organizational Safety Culture Assessment . . . . . . . . . . . . . . . . . . . 728 Alin Gaureanu, Anca Draghici, Corina Dufour, and Hugo Weinschrott IHSED 4: Transportation System Design and Safety Engineering Driver Stress Response to Self-driving Vehicles and Takeover Request – An Expert Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 737 Paul März and Uwe Handmann Monitoring Driver Posture Through Sensorized Seat . . . . . . . . . . . . . . . 744 Alberto Vergnano and Francesco Leali Evaluation of Persona-Based User Scenarios in Vehicle Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750 Stefanie Beyer and Alexander Müller Participatory Design for Optimizing the Implementation of New Transport Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 757 Elise Crawford, Yvonne Toft, Ryan Kift, and Geoff Dell

xviii

Contents

Modularity in Seaport Management: Identiﬁcation of the Research Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763 Janusz Rymaniak and Bogdan Nogalski Intelligent Material Transportation System Design for Small and Medium-Sized Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 770 Rujun Gao A New Approach to Green Light Optimal Speed Advisory (GLOSA) Systems and Its Limitations in Trafﬁc Flows . . . . . . . . . . . . . . . . . . . . . 776 Hironori Suzuki and Yoshitaka Marumo Path Planning Based on A* Algorithm for Unmanned Surface Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 783 Kang Hou, Xiaojuan Lan, Yucheng Zhang, and Sumarga Kumar Sah Tyagi IHSED 4: Safety Evaluation and Ergonomic Risks Human Machine Interface Issues for Drone Fleet Management . . . . . . . 791 Salvatore Luongo, Marianna Di Gregorio, Giuliana Vitiello, and Angela Vozella A Comparative Analysis of AS/NZS 4801, ISO 45000 and OHSAS 18001 Safety Management Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 797 Manikam Pillay Workplace Violence in Finnish Emergency Departments . . . . . . . . . . . . 804 Johanna Pulkkinen Methodological Proposal for Ergonomic Risks Evaluation . . . . . . . . . . . 809 Cesar Corrales and Milagros Chambe IHSED 4: Organizational and Strategic Interventions ERGO@OFFICE: A Participatory Ergonomics Approach for Strategic Interventions and Prevention of Musculoskeletal Disorders in SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819 Tânia M. Lima and Denis A. Coelho Interdisciplinary Design Teams in Poland - Architecture as a Tool for Preventing Hospital-Acquired Infections . . . . . . . . . . . . . . . . . . . . . . 826 Rafal Janowicz Study of the School Furniture Adequacy to Students’ Anthropometric Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 832 Agostinho Fernandes, Paula Carneiro, Nelson Costa, and Ana C. Braga

Contents

xix

Formalization and Quantiﬁcation of Team Contexts for Meso-cognitive Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 838 Taro Kanno, Daichi Mitsuhashi, Satoru Inoue, Daisuke Karikawa, and Kohei Nonose IHSED 5: Sports Design and Sports Medicine DJ-Running: Wearables and Emotions for Improving Running Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 847 Pedro Álvarez, José Ramón Beltrán, and Sandra Baldassarri A Survey of Motion Capture Technology and Its Application in Sports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854 Tianyu He and Qi Luo Experimental Elucidation on Balance Mechanism in Golf Swing for Performance Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 860 Minoru Fukumoto, Kyoko Shibata, Yoshio Inoue, and Motomichi Sonobe Organization of Tennis Clubs to Eliminate Barriers Most Frequently Speciﬁed by People with Visual Impairments . . . . . . . . . . . . . . . . . . . . . 866 Aleksandra Polak-Sopinska and Ewa Nebelska Design and Realization of Catching and Grappling Course Multimedia CAI System Based on Web . . . . . . . . . . . . . . . . . . . . . . . . . 873 Xin Wang IHSED 6: Biomechanics, Evaluation and Assessment Full Body Three Dimensional Joint Angles Validation Using TEA Ergo Inertial Measurement Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 879 Thomas Peeters, Stijn Verwulgen, Raman Garimella, Koen Beyers, and Steven Truijen Biomechanical Digital Human Models: Chances and Challenges to Expand Ergonomic Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 885 Markus Peters, Eric Quadrat, Alexander Nolte, Alexander Wolf, Jörg Miehling, Sandro Wartzack, Wolfgang Leidholdt, Sebastian Bauer, Lars Fritzsche, and Sascha Wischniewski Variable Diagnostic and/or Strengthening Training Equipment for the Upper-Body Muscles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 891 Ľubomír Šooš and Alena Cepková Workers’ Body Constitution as a Risk Factor During Manual Materials Handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 898 Ana Colim, Pedro Arezes, Paulo Flores, and Ana Cristina Braga

xx

Contents

The Evaluation of the Interaction Between Human Buttocks Thighs and Wheelchair Seat Cushion to Prevent Pressure Ulcers Using Finite Element Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 904 He Thong Bui, Philippe Lestriez, The Nhan Pham Nguyen, Le Van Nguyen, Quang Bang Tao, Karl Debray, Thi Hai Van Nguyen, and Redha Taiar Design and Realization of Police Physical Fitness Assessment Management System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 911 Xin Wang EMG Comparison of Sport Manual Wheelchair Propelled by Lever Drive and Push Rims and Possible Consequences for Rehabilitation: A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 915 Krzysztof Fiok, Michalina Błażkiewicz, Ida Wiszomirska, Nadir Skendraoui, Fabien Bogard, Sébastien Murer, and Redha Taiar Validation of a Digital Interface for Assessment of Motor Function Based on MFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 921 Adriana Gomes L. de Souza, Dominique Vincent-Genod, Carole Vuillerot, Michel Dubois, and Guillaume Thomann Anthropometric Evaluation of the Classroom Desk for Middle Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 927 Ahamed Altaboli, Omar Elﬁturi, Ahmed Alturkey, Mohammed Dogman, and Mohammed Almagrhi Noninvasive Estimation of Lumbar Intervertebral Disk Load Using Multiple Regression Analysis to Consider the Pelvic Tilt . . . . . . . . . . . . 933 Kyoko Shibata, Yasuhito Tsuyoshi, Yoshio Inoue, Hironobu Satoh, and Motomichi Sonobe Development of Stride Estimation System for Improvement of Walking Efﬁciency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939 Gen Miyamoto, Kyoko Shibata, Motomichi Sonobe, and Yoshio Inoue Estimation of Hip Joint Moment by an Inertial Measurement Unit . . . . 946 Hiroki Kotani, Kyoko Shibata, Motomichi Sonobe, Yoshio Inoue, and Hironobu Satoh IHSED 6: Health Management and Rehabilitation Medical Devices Vigilance, «Matériovigilance», as an Actor of the Hospital Safety Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955 Marine Berrué and Dominique Thiveaud

Contents

xxi

Assessment of Therapeutic Progress After Acquired Brain Injury Employing Electroencephalography and Autoencoder Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 961 Adam Kurowski and Andrzej Czyżewski Development of a Biomechanical Bike with Assistive Technologies to Be Used for Rehabilitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 968 Anabela Gomes, Álvaro Santos, Carlos Alcobia, César Páris, Deolinda Rasteiro, Emília Bigotte, Fernando Moita, Filipe Carvalho, Gabriel Pires, Jorge Lains, Pedro Amaro, and Luís Roseiro National Patient Registry: A Web-based Technological Solution for Haemophilia in Portugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974 Leonor Teixeira, Vasco Saavedra, Beatriz Sousa Santos, and Carlos Ferreira A Classiﬁcation of Motor Imagery Brain Signals Using Least Square Support Vector Machine and Chaotic Particles Swarm Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 981 Arwa N. Al-Edaily Robotic System for Active-Passive Strength Therapy . . . . . . . . . . . . . . . 987 Eliseo Cortes Torres, Anibal Alexandre Campos, Daniel Martins, and Eduardo Bock Preliminary Study of Facial Soft Tissue Thickness in Indian Children from Mumbai City . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 994 Parth Shah, Yan Luximon, and Vividh Makwana The Inﬂuence of the Manner of Grasping a White Cane on the Ability of Visually Impaired People to Perceive the Texture of Objects . . . . . . . 1000 Kiyohiko Nunokawa, Manabu Chikai, Kouki Doi, and Shuichi Ino A Contactless Walking Stick Using a Depth Sensor and Vibrators . . . . 1007 Dai Kudo, Munehiro Takimoto, and Yasushi Kambayashi A Study for Adapting the Monitoring System in Order to Prevent Fall Down from a Bed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1013 Hironobu Satoh and Kyoko Shibata Therbligh Motions as a Basic of Movement Therapy for Stroke Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1018 Bernadus Kristyanto, Brilianta Budi Nugraha, Suyoto Suyoto, Anugrah Kusumo Pamosoaji, and Kristanto Agung Nugroho

xxii

Contents

IHSED 7: Human Cyber Physical Systems Interaction Applications A Human-in-the-Loop Approach for Energy Flexibility System Integration to Support Infrastructures . . . . . . . . . . . . . . . . . . . . . . . . . . 1027 Wim Zeiler and Timi Labeodan A Survey on Trust in Augmented Human Technologies . . . . . . . . . . . . . 1033 Jean-Marc Seigneur, Tareq Ahram, and Redha Taiar Are We Designing Cybersecurity to Protect People from Malicious Actors? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1038 Alex Cadzow Study on Interaction Modalities Between Humans and CPS in Sociotechnical Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1044 Stuart Chapman, Thomas Kirks, and Jana Jost Security Design from Ergonomic Perspective: From “Total Security” to “Acceptable Security” Design for a Better Real Security . . . . . . . . . . 1051 Ferdinand Monéger, Fabien Coutarel, Motak Ladislav, Patrick Chambres, Marie Izaute, and Michel Dhome Ethics as a Security Role . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1058 Scott Cadzow An Information Management Framework to Industry 4.0: A Lean Thinking Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1063 Leonor Teixeira, Carlos Ferreira, and Beatriz Sousa Santos Alternative Ensemble Classiﬁer Based on Penalty Strategy for Improving Prediction Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1070 Cindy-Pamela Lopez, Maritzol Tenemaza, and Edison Loza-Aguirre The Effect of Cognitive Load in 3D Virtual Environments . . . . . . . . . . 1077 Siao-Wei Huang and Yu-Chen Hsu Research on Visual Speech Recognition Based on Local Binary Pattern and Stacked Sparse Autoencoder . . . . . . . . . . . . . . . . . . . . . . . . 1082 Yuanyao Lu, Ke Gu, and Shan He A Software Tool for the Calculation of Time Standards by Means of Predetermined Motion Time Systems and Motion Sensing Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1088 Jaime León-Duarte, Luis Aguilar-Yocupicio, and Luis Romero-Dessens PythaPosi: Indoor Location Estimation with Physics Constraint and Recursive Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1094 Masaaki Ano and Hiroaki Tobita

Contents

xxiii

IHSED 8: Business, Design and Technology Becoming Digital – Instruments for SME . . . . . . . . . . . . . . . . . . . . . . . . 1103 Holger Heppner and Katharina Schlicher Extracting Customer-Related Information for Need Identiﬁcation . . . . . 1108 Antonia Fels, Kristof Briele, Max Ellerich, and Robert Schmitt Visualization System Design for the Process of Packaging Egg Powders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1113 Irena Barbara Jałmużna, Jan Królikowski, and Marcin Sadok Live Migration Control Method for Ensuring Sustainable Development of e-Learning Environment . . . . . . . . . . . . . . . . . . . . . . . . 1120 Satoshi Togawa and Kazuhide Kanenishi Analyzing Design Process - Reﬂection on 25 Years of Professional Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1126 Gonçalo Falcão A Research-Based Approach for Higher Education Systems: Nigeria in Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1133 Ifetayo Oluwafemi, Clinton Aigbavboa, Jan-Harm C. Pretorius, and Jesusetemi Oluwafemi Survey on the Inﬂuence of Managing Strikes, Industrial Manifestation for Industrial Harmony in Nigeria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1139 Ifetayo Oluwafemi, Clinton Aigbavboa, Jan-Harm C. Pretorius, and J. F. Oluwafemi Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147

IHSED 1: Human-Centered Design and User Experience

Applying Human-Centered Design and Human-Machine Integration Techniques to Solve Key Healthcare Problems Neil Gomes(&) and Viraj Patwardhan Thomas Jefferson University and Jefferson Health, 925 Chestnut Street, Philadelphia, PA 19107, USA [email protected]

Abstract. While healthcare has been behind in applying human-centered design (HCD) principles to enhance service, the use of human-machine systems integration techniques in concert with HCD principles is even rarer. This paper describes the value any organization or team can derive at the confluence of these two techniques, especially in an era of Artiﬁcial and Augmented Intelligence (AI) and Machine Learning (ML), and outlines the unique model used by Thomas Jefferson University and Jefferson Health’s DICE (Digital Innovation & Consumer Experience) Group, that applies HCD thinking and human-machine integration to solve critical healthcare problems that need solving for the industry to evolve into one that is more focused on the patient and the patient experience and one that also leverages modern digital technologies such as AI, ML and the Internet of Things (IoT). Keywords: Human-centered design HCD Human-systems integration HSI Human-machine integration HMI Artiﬁcial intelligence AI Machine learning ML Augmented intelligence Healthcare Digital innovation Consumer experience Internet of Things IoT Systems engineering

1 Introduction Stardate: 4729.4. Original Airdate: March 8, 1968. Star Trek: The Original Series [1] Captain’s Log, stardate 4729.4. The M-5 computer has been installed on board ship, and we have left the space station for test maneuvers. McCoy: I don’t like it, Jim. A vessel this size cannot be run by one computer. Spock: We are attempting to prove it can run this ship more efﬁciently than man. McCoy: Maybe you’re trying to prove that, Spock, but don’t count me in on it. Spock: The most unfortunate lack in current computer programming is that there is nothing available to immediately replace the starship surgeon. McCoy: Very funny. If it could, they wouldn’t have to replace me. I’d resign because everybody else aboard would be nothing but circuits and memory banks. You know the type, Spock. Jim, you haven’t had much to say about this.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 3–9, 2019. https://doi.org/10.1007/978-3-030-02053-8_1

4

N. Gomes and V. Patwardhan

Kirk: What do you want me to say? M-5 is an honor, they tell me. Well, I’m honored. Movies have always had a knack of predicting future technologies well before their realization. And this vision of the future by Laurence N. Wolfe, the writer of episode 53 of Star Trek: The Original Series of episodes in 1968 is one such example of television prophesy. And even in this example, we see a strong resistance and disbelief from Dr. Leonard “Bones” McCoy to being replaced by a machine. Recently, Sun Microsystems co-founder Vinod Khosla [2] was quoted in Wired UK magazine as saying that “machines will replace 80% of doctors in a healthcare future that will be driven by entrepreneurs, not medical professionals.” With advances in Artiﬁcial Intelligence (AI) and Machine Learning (ML), these prophesies, new and old, are close to being realized. We now have to deal with this future and its impact on our lives as humans and even more so as patients and clinicians accessing and delivering healthcare. This is the reason, we believe, that we have to develop a bridge between humans and machines that eases this transition into a future of real Augmented Intelligence where humans and machines work in tandem to advance our lives, improve our use of the limited resources our planet has to offer, forward science responsibly and ethically and, most importantly, help us lead healthy lives and save even more lives when called upon. This bridge, we believe, needs the application of human-centered design and humanmachine integration techniques as outlined in this article to be successful.

2 Digital Revolutions and the Rise of the Machines Over the past two decades technology has made signiﬁcant changes to how we live our day-to-day lives. From smart computers and mobile devices to online retail and cloud storage, technology has become an integral part of our lives. Many industries have taken a signiﬁcant leap in delivering their service offerings with technology, but in contrast, healthcare has been slow in adopting technology to improve care delivery. The ﬁrst real breakthrough in broad human-machine interaction for consumers was the introduction of the ATM machine (initially called cash machines) by Barclays in London in 1967. While other types of vending machines existed before these cash machines, they dispensed an actual product while the cash machines dispensed money from an account thereby requiring a strong trust relationship between human and machine. Sweet [3] in the Star Business Journal, while quoting Bernardo Batiz-Lazo, a business professor at Bangor University in Britain, says “early users of automated tellers were often checking their balances twice: once to see how much was in their account, then again after withdrawing money to see if it registered.” As the adjusted balances proved to be accurate, we as humans, began to trust these machines to transact a valued asset to us. In fact, we have now gone so far as to even begin to trust our children to robots such as iPal. Wong [4] quotes Avatar Mind founder Jiping Wang in The Guardian who states that the iPal robot “could keep children aged three to eight occupied for a ‘couple of hours’ without adult supervision” thereby “taking the debate over the automation of human jobs to the next level.”

Applying Human-Centered Design and Human-Machine Integration Techniques

5

Over time, our relationship with technology has evolved into one of trust and even complete dependence with the rise of the Internet, portable computing via laptops and smartphones, and the development of decision-making algorithms that even tell us what music or movies we might like before we know it. Streaming services like Pandora and Netflix have perfected algorithmic prediction to a science, not just for providing consumers with a better decision-making experience, but also for helping Netflix better predict when a customer might be heading towards a service cancellation or to better route investment in original content as reported by Zach Bulygo [5], a blogger at NeilPatel.com. However, healthcare as an industry has been slow in adopting technology. Even seemingly obvious concepts in other industries such as telehealth have taken an inordinate amount of time to make it into the regular delivery streams of care. Arndt [6], quotes Dr. Stephen K. Klasko, President and CEO of Thomas Jefferson University and Jefferson Health, in her blog post on Modern Healthcare’s Transformation Summit 2018, as he paraphrases what former Apple CEO John Sculley said about the rise of online banking, “‘we don’t talk about telebanking anymore’, so we should also stop talking about telehealth like it were somehow separated from healthcare.” While some of this resistance is with good reason as the cost of making a mistake in healthcare is high, a lot of the resistance is also due to the absence of good human-centered design and human-machine integration techniques when integrating and acculturating technologies into environments of care.

3 Human-Centered Design and Human-Machine Interaction 3.1

Human-Centered Design

Human-centered design (HCD) is a design methodology where problems are solved by keeping consumers at the center and using empathy as a tool to deﬁne and resolve problems. According to Wikipedia [7]: “Human-centered design (HCD) is a design and management framework that develops solutions to problems by involving the human perspective in all steps of the problem-solving process. Human involvement typically takes place in observing the problem within context, brainstorming, conceptualizing, developing, and implementing the solution.”

Over the past 20 years many industries have embraced HCD as a foundational tool to tackle issues they face. HCD is inclusive as it involves the users in the problemsolving process. This helps with the adoption and sustainability of implemented solutions. The furniture company, Steelcase [8], has used HCD for over two decades and credit much of their success in the industry to their focus on the consumer and their pursuit of excellence in the application of HCD techniques. As a result, their furniture is not just beautifully designed but also very functional and more easily accepted by consumers. The HCD approach allows for constant ideation which then allows for design evolution and modiﬁcation throughout the process of solution development. This approach also brings businesses closer to their consumers.

6

3.2

N. Gomes and V. Patwardhan

Confluence of Human-Centered Design and Human-Machine Interaction

With the exponential growth of technology, it was inevitable that industries would start to merge the HCD approach with technological advances to improve their offerings. Companies like Domino’s Pizza have made great advances in not just the use of HCD principles, but also in merging these principles with focused human-machine interaction techniques. In doing this, they have not just addressed the needs of their customers and become exemplars of change and innovation, they have also cemented their leadership in their industries via back-end process enhancement spurred by the express desire to deliver to their customer needs without any friction. As per Yahoo Finance [9] historical stock prices, in 2010, Dominos’ share price was at a low $11.79. As reported by Taylor [10] the brand had become a laughing stock for many consumers who publicly shared the opinion that the pizza was the “worst pizza I ever had” and that “the crust tastes like cardboard”. Fast-forward eight years and in 2018, Domino’s stock value is approximately $279 with over 14,200 locations in more than 80 countries making it the world’s sixth largest restaurant chain. How did Domino’s achieve this? In the same article Taylor [10] writes that Patrick Doyle, CEO of Domino’s pizza said at a CEO conference in Detroit “Domino’s is not just in the pizza-making business, but in the pizza-delivery business, which means it has to be in the technology business. “We are as much a tech company as we are a pizza company.” Patrick’s statement highlights Domino’s’ shift to placing their consumer at the center and thereby realizing, from customer feedback, that pizza delivery is as important as pizza making. His statement also reveals the companies’ focus on leveraging technology to deliver a better consumer experience. What this renewed mindset did was focus Domino’s on literally delivering a frictionless experience to its customers and consequently becoming a leader in its space. Note that in order to do this, Domino’s had to design more efﬁcient, technology-enabled pizza making and delivery practices and ease its franchises into this new way of delivering an experience versus just making a better pizza. One such technology-enabled development is the Pizza Tracker in the Domino’s mobile app. Not only does this engage and inform the consumer, it also ensures that Domino’s can track a franchisee’s performance at making a pizza on time and delivering a consistent experience. Several other such technological advances developed by Domino’s to attract and engage customers and enhance customer loyalty also result in franchisee performance improvement and better overall process design in the supply, production, and delivery processes at Domino’s. However, these successes would not have been realized by Domino’s or its franchisees if Domino’s did not ensure that these changes, enabled by human-machine interaction, were supported by focused training that eased this human-machine interaction and created an environment of trust between human and machine.

Applying Human-Centered Design and Human-Machine Integration Techniques

7

4 DICE Group and the Rise of the Machines (in Healthcare) Since 2010, many healthcare institutions have signiﬁcantly invested in human-centered design (HCD), either by hiring consulting agencies or establishing their own teams to solve key problems. These departments and innovation centers have made a signiﬁcant impact improving the patient experience. However, while human-centered design is gaining ground in this rapidly evolving industry, its success is enhanced further in the new age of machine learning and robotic process design by, in tandem, developing frameworks around human-machine systems integration. Very few health systems with design and process engineering groups have developed this type of hybrid approach to solving process, service delivery, and experience problems in healthcare. One such example is the Digital Innovation and Consumer Experience (DICE) Group at Jefferson (http://dicegrp.org). What started as a small team with single-digit employee strength has now grown to become a department (a company-within-acompany) that enables Jefferson and its staff to deliver a superior patient experience. Operating at the confluence of HCD and human-machine interaction, just like leaders in other industries, and enveloping consequent solutions with focused training, support and clinical talent, the DICE Group is uniquely positioned to solve complicated healthcare problems as its counterparts have done with a similar ethos in other industries. The HCD-focused approach allows DICE to identify and deﬁne some of the most important problems from consumers themselves – who can be patients and/or clinical staff – and do this more systematically and thoroughly before attempting to solve them. The development teams then work closely with the design teams to create, test, and implement solutions using the most relevant technology for those solutions. Finally, and most importantly, the training, documentation, and support teams apply human-machine interaction principles to ease developed solutions into everyday workflows that ensure adoption and continued use of the solutions developed collaboratively by the DICE Group. Over the past three years, leveraging these principles, the DICE group has developed several key healthcare solutions that have enhanced Emergency Room (ER) workflows to deliver a faster and more satisfying ER experience; created IoT-, AI, and ML-enabled solutions using consumer technologies such as Alexa and Google Voice to speed patient transfers between hospitals and smart-enable hospital rooms via voice assistants; enabled rapid responses to patient issues while they are in the hospital via patient rounding solutions; etc. The constant in all these successful solutions is the method of discovery, development, and delivery that leverages HCD and humanmachine interaction techniques encapsulated in the simple list of principles below: 1. Discover: Approach every problem with an exploratory mindset; like the explorers of yore who opted for a compass rather than a map. With this mindset, you can discover and notice problems and then start to devise solutions. Even if approached with a solution, we have to insist that we fully understand the problem before devising a solution. Discovery via ethnographic research and direct consumer interaction is highly advised. 2. Develop and Measure: Develop only solutions that address consumer friction and enhance back-end processes in a sustainable manner. Measure through the

8

N. Gomes and V. Patwardhan

development process to ensure success and enable pivots if measurements reveal the solution as not addressing the problem. Develop using agile principles for rapid and informed development and to reduce failure. 3. Inform, Engage, and Support: Inform, engage, and support process owners, process contributors, and consumers through the development, measurement, and delivery cycles to ensure that the solution has a life beyond your involvement. This also ensures that the constituents of the value chain own the solution, understand its global importance, and want to use it and deliver it. The DICE Group’s ER Dashboards initiative, among others, used the above process and delivered a real-world and digital solution that is regularly used by clinical staff to speed up our ERs (Left-Without-Being-Seen (LWBS) rate reduced by 3%, wait time to physician reduced by 24 min, average length-of-stay for discharged patients reduced by 42 min), contribute to a better patient experience, differentiate us from our peers, and most importantly, save lives.

5 Summary In summary, it is our ﬁrm belief that technology can help us solve many problems in healthcare if we design and build technology-enabled solutions and processes using human-centered design principles and implement them using human-machine interaction techniques that ease implementation and ensure continued success. Acknowledgments. We would like to acknowledge all of our patients who inspire us to develop digital and process solutions that improve lives; every member of our DICE Group who inspire us with their resolve; and our leadership, clinical and administrative staff, and board who actively support us in our endeavor to save lives and change the world.

References 1. Wolfe, L.N., Fontana, D.C., Lucas, J.M.: The Ultimate Computer [Episode 53]. In: Roddenberry, G. (eds.) Star Trek: The Original Series. National Broadcasting Company, New York (1968) 2. Clark, L.: Vinod Khosla: Machines will replace 80 percent of doctors. Wired UK, 4 September 2012. http://www.wired.co.uk/article/doctors-replaced-with-machines 3. Sweet, K: How the ATM changed the consumer-machine relationship. Star Bus. J., 21 September 2017. https://www.thestar.com/business/2017/09/21/how-the-atm-changed-thehuman-machine-relationship.html 4. Wong, J.C.: ‘This is Awful’: Robot can keep children occupied for hours without supervision. The Guardian, 29 September 2016. https://www.theguardian.com/technology/ 2016/sep/29/ipal-robot-childcare-robobusiness-san-jose 5. Bulygo, Z.: How Netflix Uses Analytics to Select Movies, Create Content, and Make Multimillion Dollar Decisions. NeilPatel.com (2018). https://neilpatel.com/blog/how-netﬂixuses-analytics/

Applying Human-Centered Design and Human-Machine Integration Techniques

9

6. Arndt, R.Z.: Transformation Summit 2018: Healthcare leaders say it’s time for innovative change, 13 June 2018. http://www.modernhealthcare.com/article/20180613/TRANSFORMATION02/180619969 7. Wikipedia: Human-centered Design (2018). https://en.wikipedia.org/wiki/Human-centered_ design 8. Arieff, A.: Steelcase: Designing for the Human/Technology Tension (2016). https://www. steelcase.com/research/articles/topics/technology/designing-for-the-humantechnologytension/ 9. Yahoo Finance: Domino’s Historical Stock Prices (2018). https://ﬁnance.yahoo.com/quote/ DPZ/history?p=DPZ 10. Taylor, B.: How Domino’s Pizza Reinvented Itself, 28 November 2016. https://hbr.org/ 2016/11/how-dominos-pizza-reinvented-itself

Subjective Evaluation of EV Sounds: A Human-Centered Approach Verena Wagner-Hartl1(&), Bernhard Graf2, Markus Resch2, and Paco Langjahr2 1

Campus Tuttlingen, Faculty Industrial Technologies, Furtwangen University, Kronenstraße 16, 78532 Tuttlingen, Germany [email protected] 2 AVL LIST GMBH, Vehicle and Powertrain Acoustics, Engineering and Technology Powertrain Systems, Hans-List-Platz 1, 8020 Graz, Austria {bernhard.graf,markus.resch,paco.langjahr}@avl.com

Abstract. In the future, an increase of electric vehicles (EVs) is expected. There is still uncertainty, however, how to design the sound of this new vehicle types in order to meet customers’ expectations and requirements. A study was conducted to get more insight in this important future ﬁeld. Overall, 48 participants evaluated different acceleration and recuperation sounds on the test track and in real trafﬁc. The results of the study show that the ﬁrst impression of the best assessed EV sounds within this study were that they give an appropriate and realistic feedback and that they are perceived as pleasant, unobtrusive and gentle. Furthermore, the assessments of the different sounds show that EV sounds with perceived high-quality are characterized by the fact that they are rather quiet, soft and reserved, rich and a bit dark, ﬁt well to the driving conditions and give the driver an important feedback about the driving behavior. Keywords: Sound design

Electric vehicles Human-system interaction

1 Introduction In the future, an increase of electric vehicles (EVs) is expected. In order to maintain future competitiveness, many OEMs are working on expanding their product portfolio regarding new drive technologies. Therefore, research and development have to face new challenges. One of these is the development of appropriate vehicle sounds or rather the sound design for electric vehicles. Vehicle sounds have a strong impact on the quality impression and the perceived comfort of drivers [1–6]. In addition, vehicle sounds are beside other quality aspects in a vehicle like e.g. design or materials, a feature that allows customers to distinguish between different brands [cf. 3, 7, 8]. Following Zeller’s [9] hierarchy of the “immediacy of experiencing essential vehicle characteristics”, vibroacoustics and sound are classiﬁed immediately behind design. This emphasizes the importance of well-designed vehicle sounds. Furthermore, previous research has established that a “quiet vehicle” is not recommendable, due to a customers’ perception of vehicles without soul [9]. In addition, absolute silence in a vehicle is perceived as threatening [10]. Therefore, sound © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 10–15, 2019. https://doi.org/10.1007/978-3-030-02053-8_2

Subjective Evaluation of EV Sounds: A Human-Centered Approach

11

design can be used to create pleasant sounds [11]. Furthermore, a sound should ﬁt well to the product to achieve customer acceptance. Electric vehicles are often described as “quiet vehicles” which needed a special sound design including two different aspects: safety for people outside the vehicle and ergonomics or feedback for people inside the vehicle [12]. The ﬁrst aspect – outside noise – is not part of the presented paper but it should be mentioned that it will be regulated by law in the near future [13]. The presented paper deals with the second aspect, the sound inside the vehicle. The lack of combustion engine noise leads to the new situation that other noises could get dominant and are not masked by the combustion engine noise. These noises could come from the electric drive unit itself or electric auxiliaries (e.g. AC compressor, steering pump) and do not give a feedback to the driver about the driving situation or condition. In addition, the electric drive unit could produce noises caused by electromagnetic excitations or by the reducer gears that could be perceived as disturbing by the customers. To sum it up, there is still uncertainty, however, how to design the sound of this new vehicle types in order to meet customers’ expectations and requirements. A study was conducted to get more insight in this important future ﬁeld. The aim of the study was to evaluate ten different new designed acceleration and recuperation sounds on the test track and in real trafﬁc. The research questions presented in the current paper are: 1. First impression: How are positively assessed EV sounds characterized by the participants? 2. Vehicle sound quality: How should EVs sound that are perceived as high-quality be characterized?

2 Materials and Method 2.1

Participants

Overall, 48 assessors (23 women and 25 men) participated in the presented study. They were between 20 and 62 years old (M = 34.44, SD = 10.85). All of them owned a valid driver license and had normal hearing abilities. Informed written consent was obtained from all participants. 2.2

Materials and Procedure

The participants evaluated ﬁve different acceleration and ﬁve different recuperation sounds in an electric vehicle (test vehicle: Tesla Model S) on the test track and in real trafﬁc. The sounds were selected from moderate order based sounds to futuristic texture based sounds to show the variety and possibilities to the assessors. All sounds were set to a similar loudness level to avoid any influence in the rating by having different interior noise levels. In general, the sounds played in the interior must fulﬁl two major tasks, one is the information and feedback to the driver and passengers that the vehicle is acceleration or deceleration (conscious), the second aspect is the emotional touch in conjunction with the driving style of the person (sub-conscious).

12

V. Wagner-Hartl et al.

First, the participants were picked up at the meeting point and were driven to the test track (see Fig. 1).

Fig. 1. Driving route.

Due to actuarial restrictions, the participants were not allowed to drive the vehicle by themselves. They sat in the passenger seat throughout the study. A test driver drove the vehicle and ensured that the study occurs under standardizes conditions. During the drive to the test track, the participants ﬁlled out a short questionnaire (soziodemographic data, questions regarding their attitude towards EVs and their attitude towards vehicle sounds in general). At the test track, the experimenter explained the study procedure. Afterwards, the drive at the test track started. During this part of the study, the participants experienced each auf the ﬁve different acceleration (0 to 90 kph) and ﬁve different recuperation sounds (recuperation started at 90 kph) in a randomized order. Each sound was experienced during two successive drives and assessed immediately afterwards (ﬁrst impression, assessments based on semantic differentials, EV-sound-speciﬁc items, assessments regarding how the sounds ﬁt to the driving conditions, feedback about the driving behavior etc.). The used items were based on assessments that were well-proven for different vehicle sounds (dimensions of vehicle sounds perception – see [6, 14] for more details). When the drives at the test track were ﬁnished, the participants had to rank the ﬁve different acceleration and the ﬁve different recuperation sounds (one ranking for each group). A short sound ﬁle including all ten sounds of both sound groups was played before the participants had to rank the sounds. This sound ﬁle should support the participants in remembering the characteristics of the different sounds they experienced at the test track. In addition, they were asked to reason their decision and answer questions regarding the sounds in a short interview. The best-assessed acceleration sound of each participant was used for the second part of the study, the drive in real trafﬁc. After the drive in real trafﬁc, the participants assessed the sound and participated in a post-interview. Each participant needed approximately 1 h and 45 min to complete the study.

Subjective Evaluation of EV Sounds: A Human-Centered Approach

2.3

13

Analysis

The ﬁrst impressions of the participants (qualitative data) were analyzed using the method of qualitative content analyses by Mayring [15]. The statistical analyses of the data were conducted using the software SPSS for Windows. The analyses were based on a signiﬁcance level of 5%.

3 Results 3.1

First Impression

To answer the ﬁrst research question, the participants were asked to verbalize their ﬁrst impression of each sound. Figure 2 shows the results for the sounds (acceleration and recuperation) that were best assessed and ranked on the three top places for each group (see method section).

Fig. 2. First impression - characterization of positively rated sounds (multiple answers possible).

3.2

Vehicle Sound Quality

A linear regression analysis was conducted to answer the second research question. The results show that the two dimensions of vehicle sounds perception [cf. 6, 14] “timbre” and “loudness” are negatively related, and the variables “Fit to the driving conditions” and “Feedback about the driving behavior” are positively related to the quality rating score (mean value, semantic differentials not at all/very high-quality, comfortable/ pleasant). The values are shown in Table 1.

14

V. Wagner-Hartl et al.

Table 1. Summary of linear regression analysis for variables predicting the quality rating score. Variable Timbre Loudness Roughness/Sharpness Fit to the driving conditions Feedback about the driving behavior Note. Adj. R2 = .612, **p < .01

B −.194 −.587 .009 .277 .099

SE B .038 .035 .041 .036 .035

b −.153** −.546** .007 .326** .114**

4 Discussion The results of the study show that the ﬁrst impression of the best assessed EV sounds within this study were that they give an appropriate and realistic feedback and that they are perceived as unobtrusive, pleasant and gentle (research question 1). Regarding research question two, the assessments of the ﬁve different acceleration and recuperation sounds show that EV sounds with perceived high-quality are characterized by the fact that they are rather quiet, soft and reserved, rich and a bit dark, ﬁt well to the driving conditions and give the driver an important feedback about the driving behavior. The result is in line with previous research that shows the importance of sound design regarding ergonomics for people inside the vehicle [12]. Furthermore, the results emphasize the importance of well-designed sounds for EV sounds that give the driver a feedback about the driving behavior and the driving conditions. Based on the ﬁrst results, the next step is to combine the subjective assessments with objective acoustic parameters. The connection of subjective and objective data can help to gain deeper insight in this relatively new ﬁeld of vehicle sound design. Following a multidimensional approach, the inclusion of psychophysiological measures to assess activation and emotional reactions in addition to subjective customer’s assessments could also be taken into account for future studies [cf. 16]. Our study has one limitation: Due to actuarial restrictions, the participants were not allowed to drive the vehicle by themselves. They sat in the passenger seat throughout the study. Especially regarding the importance of the sounds for an appropriate feedback of the driving behavior, it would be interesting for future studies to let the participants drive by themselves. To sum it up, the results of the presented study can contribute considerably to sound optimization of the product automobile and especially of the sound of future vehicle types.

References 1. Miśkiewicz, A., Letoweski, T.: Psychoacoustics in the automotive industry. Acustica Acta Acust. 85, 646–649 (1999) 2. Schulte-Fortkamp, B., Genuit, K., Fiebig, A.: A new approach for developing vehicle target sounds. Sound Vib., 2–5 (2007)

Subjective Evaluation of EV Sounds: A Human-Centered Approach

15

3. Nor, M.J.M., Fouladi, M.H., Nahvi, H., Arifﬁn, A.K.: Index for vehicle acoustical comfort inside a passenger car. Appl. Acoust. 69, 343–353 (2008) 4. Genuit, K.: Product Sound Quality of Vehicle Noise – A Permanent Challenge for NVH Measurement Technologies. SAE International, 2008-36-0517, pp. 1–17 (2008) 5. Beitz, T., Wagner, V., Enigk, H.: Importance of operational sounds for vehicle quality. In: Pischinger, S., Biermann, J.-W., Genuit, K., Vorländer, M. (eds.) Proceedings Aachener Acoustics Colloquium, pp. 33–36. Aachen, Germany (2010) 6. Wagner, V.: Hochwertigkeit von Geräuschen im Fahrzeuginnenraum [High-Qualityness of In-vehicle Sounds]. In: Sachse, P., Ulich, E. (eds.) Beiträge zur Arbeitspsychologie [Contributions to Work Psychology]. Pabst Science Publishers, Lengerich (2014) 7. Fastl, H.: Psycho-acoustic and sound quality. In: Blauert, J. (ed.) Communication Acoustics, pp. 139–162. Springer, Heidelberg (2005) 8. Schifferstein, H.N.J.: The perceived importance of sensory modalities in product usage: a study of self-reports. Acta Psychol. 121, 41–64 (2006) 9. Zeller, P.: Handbuch Fahrzeugakustik. Grundlagen, Auslegung, Berechnung, Versuch [Manual for Vehicle Acoustics. Basics, Design, Calculation, Testing]. Vieweg + Teubner, Wiesbaden (2009) 10. Steinberg, K.F.: Charakterisierung von Störgeräuschen [Characterization of noise]. In: Genuit, K. (ed.) Sound-Engineering im Automobilbereich [Sound Engineering in the Automotive Sector], pp. 183–204. Springer, Heidelberg (2010) 11. Büchling, J.: Sound-Design – zentrale Aufgabe bei der Automobilentwicklung [Sound design - central task in automotive development]. Automot. Eng. Partn. 5, 44–47 (2001) 12. Misdariis, N., Cera, A.: Sound signature of quiet vehicles: state of the art and experience feedbacks. In: Proceedings of Internoise 2013, Innsbruck, Austria, pp. 1–10 (2013) 13. European Union: Regulation (EU) No 540/2014 of the European parliament and of the council of 16 April 2014 on the sound level of motor vehicles and of replacement silencing systems, and amending Directive 2007/46/EC and repealing Directive 70/157/EEC. Ofﬁcial Journal of the European Union, L 158/131 (2014) 14. Wagner, V., Kallus, K.W., Foehl, U.: Dimensions of vehicle sounds perception. Appl. Ergon. 64, 41–46 (2017) 15. Mayring, P.: Qualitative Inhaltsanalyse: Grundlagen und Techniken [Qualitative content analysis: basics and techniques]. Beltz, Weinheim and Basel (2010) 16. Wagner, V., Kallus, K.W.: Sound quality of turn indicator sounds – use of a multidimensional approach in the automotive product development. J. Trafﬁc Transp. Eng. 3(3), 158–165 (2015)

Sequential Recognition Rate and Latency of Frequency-Based Tactons Ricardo Jimenez(&) and Ana Maria Jimenez(&) Barry University, 11300 NE 2nd Avenue Miami Shores, Florida, USA {rjimenez,amjimenez}@barry.edu

Abstract. Tactons are tactile cues that allow the sense of touch to be used in Human Computer Interaction (HCI). This experiment determines a computer user’s ability and time needed to associate, in sequence, two distinct sets of four tactons encoded with four different sinusoidal waveforms of different frequencies and same amplitude, compared to baseline recognition rates and latency using audible cues instead of tactons. To date 18 participants have been tested. Preliminary results show 55% ± 2.8 recognition rate of the ﬁrst tacton in the sequence, 46% ± 3.4 recognition rate of second tacton in sequence and 29% ± 3.3 rate of both tactons in sequence recognition. Data indicates that a signiﬁcantly higher rate of detection of at least one object is feasible. It is expected that with additional training, recognition rates would increase while latency time is reduced. This is an ongoing project, in which data continues to be collected. Keywords: Tactons Skin sensation

Human Computer Interaction Haptics

1 Introduction In an effort to increase human-computer interactions (HCI), to add both redundancy to existing modalities, and to possibly replace diminishing ones, tactons are being explored with increasing interest. Tactons are tactile icons that interrelate the cutaneous sense of touch on human skin as a haptic form of output which can then be used as an additional form of computer-based communication [1]. The tacton functions in conjunction with a vibrotactile device, which is placed on the skin surface and mechanically vibrates to produce sensations which can then be associated with computer functions, such as icons on a screen. Tactons allow HCI to occur with minimal need for an audio or visual interface [1], adding redundancy in low-light or high-noise environments. In addition, tactons can enhance diminishing sensory functions particularly in the elderly population. In the literature, tactons have been shown to be effective means of communicating complex information [2]; therefore, deﬁning tacton stimuli parameters and data structures; as well as determining tacton recognition rates and detection thresholds, is an ongoing pursuit [3–6] in the research community. Tactons and vibrotactile devices have been demonstrated to have a wide-range of potential uses. Possible uses include supplementation or even replacement of visual displays, particularly useful when display size is limited or unavailable [1]. This naturally lends itself © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 16–20, 2019. https://doi.org/10.1007/978-3-030-02053-8_3

Sequential Recognition Rate and Latency of Frequency-Based Tactons

17

to enhancing HCI for the blind and visually impaired computer users, adding an additional layer of communication for this population. Several studies have looked at tacton design to enhance blind-users’ computer interaction, as well as to provide navigational aid [7–10]. Speciﬁcally, McDaniel et al. [7] found that tactile rhythm-cues can be used to aid in gauging interpersonal distance by blind users. While, Ghiani et al. [8] used tactons as a navigational aid for blind users and demonstrated how they can be incorporated into mobile museum guides. Furthermore, Fraser and Gutwin [9] proposed a framework where assistive technology was used in mouse pointers speciﬁcally to address the needs of low vision users Ertan et al. [10] showed the feasibility of applying vibrotactile devices in a wearable haptic navigation guidance system. What’s more, tactons can be used to enhance desktop interaction by adding an additional modality, which can increase information exchange, by providing an additional sensory channel of information [1]. Currently, there are only a few commercially available vibrotactile devices, which have been speciﬁcally designed for safe use on the skin surface. These devices are capable of faithfully reproducing complex waveforms and can be more precisely controlled using linear actuators. Actuators are equipped with moving contacts that can be placed directly on the skin to provide more localized and precise transfer of vibrations. The best example of such advanced, commercially available devices is the EAI C2 Tactor. This paper presents ongoing investigations that aim to provide empirical data that can subsequently be used to deﬁne tacton stimuli-parameters and data structures, i.e., determining the recognition rates of sinusoidal tactons of varying frequencies, and latency of human response when using the C2 Tactor from EAI systems. Tactons of varying frequencies are associated with visual functions on a GUI interface; recorded user data includes recognition rate as well as response latency. Initial ﬁndings of this work were presented as a poster at AHFE 2018 conference [11] to illicit response from the academic community.

2 Materials and Methods The study consists of two identical protocols, where the variable is the type of cue given to the subject. Both protocols include a total of eight objects represented as boxes on a computer screen. Speciﬁcally, four identically colored boxes numbered 1 through 4 located on the center of the screen, and four additional color-coded boxes located on the four corners of the computer screen. Each of these eight objects is associated with a speciﬁc vibration frequency felt on the skin (tacton-protocol), as well as, with an audible cue, i.e., either the color or number identiﬁed on the box (auditory-cue), see Table 1. At the beginning of each protocol, the participant is allowed to train for up to 30 min, matching the cue with its corresponding object on the computer screen. When the participant reports being able to recognize the objects reflected on the screen after being prompted by the paired tacton-cues or the paired auditory-cues, the testing protocol is initiated and the user-responses recorded. The subject is presented with the tacton protocol ﬁrst, by delivering two vibrotactile cues, each one associated with a speciﬁc frequency generated by the C2 tactor and

18

R. Jimenez and A. M. Jimenez Table 1. Stimulation cues and linked response according to object on screen.

Object on screen expected response Red object Blue object Green object Yellow object Object 1 Object 2 Object 3 Object 4

Tacton stimulation frequency (Hz) 150 100 250 200 150 200 250 100

Audible stimulation cue Red Blue Green Yellow One Two Three Four

associated with either a color (tacton 1) or a number (tacton 2), see Table 1. The participant then proceeds to match the tacton cue to the corresponding object on the screen by clicking on the appropriate colored object (tacton 1), and on the appropriate numbered object (tacton 2). The cues were presented as pairs with tacton 1 followed by tacton 2. In order to test for recognition rather than memorization of object position on the screen, the objects randomly reappear on the screen at different locations after each trial, with a total of 16 trials. At the automatic completion of this protocol, the participant can continue to the auditory cue protocol. Association of Tacton and Audio Cues with Requested Response. Each tacton associated with a cutaneous vibration frequency used as a cue for the participant to respond to an object shown on the screen as a color (tacton 1) or a numbered box (tacton 2). The rate of user recognition and latency was compared to a baseline association of an audible cue for the name of the color (audio 1) or its number (audio 2). For the subsequent auditory-cue experimental protocol, the participant uses headphones to receive auditory cues, instead of tacton cues. As in the previous protocol, each cue is associated with an object on the screen. For example, an audible cue such as “yellow, one” prompts the subject to click on the matching object(s) on the screen, namely the yellow-colored box and the box labeled “1”. The cues are presented as pairs with the name of color ﬁrst, followed by the number. After each trial, the objects randomly reappear on the screen with a total of 16 trials. For both protocols, participant data is collected for the object chosen when reflected on the screen, as well as its corresponding latency from cue delivery to user-response. Recognition rate is based on the number of correct matches to its cue. In this report, the auditory-cue protocol served as a baseline comparison for the user’s recognition rate and recognition latency when using tactons. Equipment for training and testing includes a computer with the 21″ LCD monitor set at 1024 768 resolution and a mouse and keyboard within comfortable reach to the subject. Tactons are generated with a commercial vibrotactile device, the C2 tactor from EAI Systems. The tactor is placed on the non-hairy portion of the volar forearm with a strap. Each tacton represented an object number and an object color. Tacton encoding consisted of sinusoidal waveforms in the frequency range of 100–250 Hz in

Sequential Recognition Rate and Latency of Frequency-Based Tactons

19

increments of 50 Hz at 1-V amplitude and presented to the user in different sequences. Each encoded tacton is conﬁrmed for waveform, frequency, and amplitude compliance using a Techtronix TDS 310 Oscilloscope. Waveforms are generated using NCH Tone Generator software and is ampliﬁed through a Crown D-75A linear ampliﬁer and controlled via an in-house developed C# program. This program, generates testing interface, randomizes tactons projected on the computer screen, and captures userresponse, as well as latency time. Test subjects were screened through a questionnaire to determine age, gender, dominant hand, and any relevant medical history. Subjects who exhibit any medical traits, which may influence results, were excluded from the study such as color blindness or conditions skin abnormalities. The experimental protocol was submitted and approved by the Barry University Institutional Review Board.

3 Results Eighteen participants have been tested and responses presented here. The tacton data was compared to baseline recognition rates and times established using an identical protocol that uses audible cues instead of tactons for identifying the objects on the computer screen. Results show that audio recognition of all audio-cues individually, as well as in sequence, was 99.7% with an average response time of 2.28 s for each audiocue. The tacton results were as follows: 55% ± 2.8 recognition rate of the ﬁrst tacton (color) in the sequence, 46% ± 3.4 recognition rate of second tacton (number) in the sequence and 29% ± 3.3 rate of both tactons in sequence recognition. The average response time of tactons was 4.88 s, which is over double the response time for audio cues. A two-tailed t-test showed a signiﬁcant difference between recognition rates of the ﬁrst tacton in the sequence (color) and the second tacton in the sequence (number). When using tacton cues the most common recognized frequency was 250 Hz, with a recognition rate of 62.8%.

4 Discussion The results indicate that recognition of tactons in a sequence is quite feasible with a 29% recognition rate. More importantly with minimal training individual tacton recognition of 55% and 46% indicates that an effective interface using tactons can be achieved with additional training and by using the C2 optimal vibration frequency. The higher recognition rate of tactons encoded at 250 Hz is in line with other ﬁndings by our lab and in the literature. Since 250 Hz appears to be the optimal frequency type of mechanoreceptors particularly Pacinian receptors [12] on the skin, as well as the optimal vibration frequency for the C2 tactor. Nevertheless, the overall recognition rate using a frequency range of 100–250 Hz as a method of tacton encoding appears to be feasible and of great potential for further developing human-computer interactions based on haptic stimuli. Future studies will further explore the effect additional training on recognition rates and latency of response.

20

R. Jimenez and A. M. Jimenez

References 1. Brewster, S.A., Brown, L.M.: Tactons: structured tactile messages for non-visual information display. In: AUIC 2004: Proceedings of the Fifth Conference on Australasian User Interface, vol. 28, Dunedin, New Zealand, pp. 15–23. Australian Computer Society (2004) 2. Brown, L.M., Brewster, S.A., Purchase, H.C.: Multidimensional Tactons for non-visual information presentation in mobile devices. In: Proceedings of the 8th Conference on HCI with Mobile Devices and Services, Espoo, Finland, pp. 231–238. ACM (2006) 3. Craig, J.C., Rhodes, R.P., Busey, T.A., Kewley-Port, D., Humes, L.E.: Aging and tactile temporal order attention. Atten. Percept. Psychophys. 72(1), 226–235 (2010) 4. Hoggan, E., Brewster, S.: New parameters for Tacton design. In: Proceedings of the SIGCHI 2007 Conference on Human Factors in Computing Systems, San Jose, CA, pp. 2417–2422. ACM (2007) 5. Stuart, M., Turman, B.A., Shaw, J., Walsh, N., Nguyen, V.: Effects of aging on vibration detection thresholds at various body regions. BMC Geriatr. 3(1), 1–10 (2003) 6. Goble, A.K., Collins, A.A., Cholewiak, W.: Vibrotactile threshold in young and old observers: the effects of spatial summation and the presence of a rigid surround. J. Acoust. Soc. Am. 99(4), 2256–2269 (1996) 7. McDaniel, T., Krishna, S., Colbry, D., Panchanathan, S.: Using tactile rhythm to convey interpersonal distances to individuals who are blind. In: CHI 2009 Conference on Human Factors in Computing Systems, Boston, MA, pp. 4669–4674. ACM (2009) 8. Ghiani, G., Leporini, B., Paterno, F.: Vibrotactile feedback as an orientation aid for blind users of mobile guides. In: Proceedings of the 10th International Conference on Human Computer Interaction with Mobile Devices and Services, Bonn, Germany, pp. 431–434. ACM (2008) 9. Fraser, J., Gutwin, C.: A framework of assistive pointers for low vision users. In: Proceeding Assets 2000 Proceedings of the Fourth International ACM Conference on Assistive Technologies, pp. 9–16. ACM, New York (2000) 10. Eertan, S., Lee, C., Willets, A., Tan, H.: A wearable haptic navigation guidance system. In: 2nd International Symposium on Wearable Computers, Pittsburgh PA, pp. 164–165. IEEE (1998) 11. Jimenez, A.M., Jimenez, R.: Sequential recognition rate and latency of frequency-based Tactons. Poster presentation. In: 9th International Conference on Applied Human Factors and Ergonomics, Orlando Florida, 21–25 July 2018 12. Lederman, S.J., Klatzky, R.L.: Haptic Perception: A tutorial. Atten. Percept. Psychophys. 71(7), 1439–1459 (2009)

Bringing It Together: Three Approaches to Combine Agile Software Development and Human-Centered Design Michael Minge(&) and Antonia Föhl Chair of Cognitive Psychology and Cognitive Ergonomics, Insitute of Psychology and Ergonomics, Technische Universität Berlin, Marchstraße 23, Sekr. MAR 3-2, 10587 Berlin, Germany [email protected], [email protected]

Abstract. Nowadays, many software companies rely on a development process that includes both, an agile implementation of technical functionalities and a human-centered design approach for ﬁnding the right interaction concept. However, in practice, these two approaches do not easily ﬁt together: Humancentered activities need much time before and during implementation. Early user tests might yield “unfavorable” results, such as the need for revising a technically correct operating functionality due to usability problems. In consequence, this leads to ad hoc prioritizations, asynchronous iterations, and communication problems between team members. Therefore, in practice, a strong need exists to ﬁnd an appropriate way to combine agile development and human-centered design activities. In this contribution, three kinds of possible approaches are presented: One sprint ahead, dual task development, and integrative approaches. These approaches will be discussed with respect to their strengths and weaknesses. Task integration is considered to be particularly promising in order to improve communication through interdisciplinary cooperation and to develop innovative, user-friendly design solutions. Keywords: Usability User experience User-centered design Agile development Sprint zero One sprint ahead Lean UX

1 Introduction Agile development is an umbrella term for a number of different approaches which aim at implementing software in an iterative and collaborative way. Scrum, Kanban, Extreme Programming (XP), and Crystal are some of the best known agile frameworks [1]. By far the most frequently used agile framework is Scrum, or a company-speciﬁc adjustment of Scrum [2]. With roles, processes, meetings, and feedback loops, Scrum is a very comprehensive framework that provides a number of requirements to structure the software development process. Following Scrum, the development process is scheduled in short implementation cycles, the sprints. Each sprint aims at creating a new functional feature, a product increment. All requirements that are necessary for the completion of the product are © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 21–27, 2019. https://doi.org/10.1007/978-3-030-02053-8_4

22

M. Minge and A. Föhl

listed in the product backlog. Based on this backlog, before each sprint, it is decided which requirements will be realized (sprint backlog). This happens in the sprint planning, a workshop at which all team members participate. After each sprint, both the quality of the product (sprint review) and the collaboration and communication processes (sprint retrospective) are analyzed by all team members in order to identify opportunities for optimization in the future. Scrum requires different roles. The product owner establishes close collaboration with the customer and represents the perspective of all stakeholders. To that end, the product owner also maintains the product backlog. The scrum master is responsible for a trouble-free and productive cooperation. Team members are developers and designers who work autonomously, but in close coordination among each other. Sometimes, it is neglected that agile development does not generally cover the analysis neither of user needs, nor of requirements [3], and that the customer perspective is not necessarily the user perspective. Neither Scrum, nor any other agile framework, includes systematic activities in order to optimize the usability of a system or its capability to elicit a positive user experience (UX). However, early consideration of user needs and requirements is nowadays a key to success for many technical devices. The human-centered design (HCD) process is an iterative approach that aims at designing technology not only for but with end users [4]. Therefore, at all stages of product development, users should continuously participate, e.g., in order to understand the context of use, to specify all relevant requirements, to generate ideas, to validate the development process, as well as to ﬁnd and to verify the most appropriate design solution. Therefore, early and regular user tests play an important role in HCD. The creation of so-called minimum-viable products (MVP) to test a speciﬁc assumption and to provide a quick and valid proof of concept is a common practice in HCD [5]. Hence, the realization of mock ups and click prototypes in order to gain user feedback at early stages of ideation and development is a typical HCD activity as well. Today, many software companies rely on a development process that includes both, an agile implementation of technical functionalities and a human-centered design approach for ﬁnding the right interaction concept. However, in practice, these two approaches do not easily ﬁt together. The two approaches usually entail different work rhythms, which do not always directly converge. Furthermore, terminologies and the underlying objectives differ. For instance, agile development aims to implement functionalities following short development cycles and to do this in close cooperation with the customer. In contrast, the HCD process claims iterations to elaborate and to reﬁne solutions regarding usability and user experience in close cooperation with the end user. The different understanding of common terms (i.e., iteration, testing, and product quality) often leads to misunderstandings resulting in conflicts, performance losses, and unfavorable ad hoc prioritizations.

Bringing It Together: Three Approaches to Combine Agile Software Development

23

2 Agility and Human-Centricity: Three Different Approaches Based on previous literature, own considerations, and experiences that we made with small and medium-sized enterprises, we consistently found three distinctive approaches to describe how agile development and user-centered design activities can be combined in practice. These approaches are presented in the following. 2.1

One Sprint Ahead

The aim of this approach is to design and to test prototypical interaction concepts of product relevant functionalities before they will be implemented [6]. Typically, the start is a design sprint (“sprint zero”), in which user-centered activities take place, such as exploration of user needs, requirement analysis, participative design sessions, prototyping, layout planning, and usability evaluation (see Fig. 1). Pre-tested design solutions become part of the product backlog in order to anticipate and to visualize the ﬁnal implementation of the design solution. Once the product increment is realized, user tests can be conducted in a follow-up evaluation to reﬁne the interaction concept and to identify further opportunities for optimization.

Fig. 1. One sprint ahead.

This approach promises three advantages: Firstly, interaction concepts are planned from the user’s perspective, which might reduce the need for severe corrections at later stages. Secondly, design and implementation can contribute to the development in a very structured and equal way. Thirdly, developers and designers work parallel and independently from each other, e.g., while in a current sprint, functionality A is under

24

M. Minge and A. Föhl

implementation, functionality B is in the design sprint. In the following sprint, B is going into implementation, while A is follow-up tested with users [7]. However, one sprint ahead often fails in practice due to rapidly changing requirements and because user-centered design activities and implementation require different expenditure of time. Consequently, in many cases, this approach leads to delays or waiting times for different teams. In particular, the start of implementation might be inefﬁciently postponed, as the activities and the length of a sprint zero are not precisely deﬁned [8]. In the event that a user test reveals usability problems, it usually becomes a question of prioritization whether there is enough time for optimizations. 2.2

Dual Task Development

The dual task approach is characterized by the development team and the design team organizing two separated, but linked working activities [9]. The main activity is the iterative implementation of running functionality by working in development sprints. Before starting or after completing a sprint, the design team might become involved in user-centered design activities (see Fig. 2). For instance, designers can take the ﬁrst step and create prototypes based on the product backlog. Additionally, they are in demand to evaluate and potentially redesign the product increment. Compared to the one sprint ahead approach, the cooperation and communication in dual task development is less well structured and organized. Opportunities for transferring tasks and responsibilities arise on short notice and as far as available resources allow.

Fig. 2. Dual task development.

Bringing It Together: Three Approaches to Combine Agile Software Development

25

A major advantage of this approach is its flexibility. User-centered design activities can be integrated in the development process as needed. Both disciplines work independently from each other so that they manage their tasks in an empowered and selfdetermined way. The design team is not only involved in product development but might also pursue long-term objectives, e.g., generating ideas for innovative product solutions in the near or distant future. The biggest challenge in dual task development is probably the high level of communication and coordination between design and development. The risk exists that both teams are pursuing only their own agenda. Therefore, in many cases, the joint collaboration of all teams together will be considerably below expectations. One might get the impression that UX design is superfluous and causes distraction from the development process. Furthermore, complex problems, such as software development, are not solved in a truly effective, i.e., in a multidisciplinary, way. 2.3

Integration of UX in Agile Development

A third approach is to synchronize implementation and UX design closely in all activities and at all stages of software development, i.e., in planning meetings, in the sprint itself, and in the evaluation of a product increment. Following this approach, different disciplines are not working independently from each other, but the teams themselves are interdisciplinary. This means, each team consists of developers and a UX designer (see Fig. 3). The implementation of a functionality within a sprint is closely accompanied by user tests. Consequently, the product increment is optimized not only with regard to its technical quality but also its user experience and acceptance.

Fig. 3. Integrative approach.

26

M. Minge and A. Föhl

The integrative approach supports that user-centered design activities are considered from the early beginning and continuously during the development process. As all team members are involved at all stages, transparency will increase. Furthermore, complex problems can be analyzed and solved from different perspectives. A major advantage is that UX design becomes part of the sprint planning, i.e., the estimation of the team velocity, and review as well as retrospective meetings. As a result, humancentricity is no longer a nice extra, but an important objective in the implementation. There are, however, a couple of challenges that will need special attention, when integrative approaches are adopted. Firstly, UX methods have to be conducted in an effective and efﬁcient way yielding valid results [5]. Typically, there is no time for exhaustive need explorations or extensive test sessions with a large number of users. Implementation as well as UX design have to strictly follow the agile rhythm. Secondly, designers working in different teams have to coordinate their decisions and strategies on a meta level to ensure a consistent product design (“Scrum of Scrum”). Finally, trainings are necessary in which employees’ social (soft) skills are developed.

3 Conclusion In this contribution, three different approaches were presented, which describe how agile software development and human-centered design activities can be systematically combined. For all approaches, i.e., one sprint ahead, dual task development, and task integration, strengths and opportunities were discussed. However, at the same time, it became clear, that each approach also causes challenges that will need special attention when it comes to its adoption and implementation. Considering a company’s corporate culture and the requirements that arise from the speciﬁc product that is developed, a motivated decision for or against each approach should be taken. Due to the close cooperation between developers and designers when following an integrative approach and due to the fact, that only in this approach, UX design becomes an essential part in the sprint planning and evaluation, task integration can be considered as particularly useful and promising. Based on the presented ﬁndings and considerations, we will use these approaches in our future work, when analyzing agile practices in small and medium sized companies and when deriving best practices. Acknowledgments. This work has been supported by the German Federal Ministry for Economic Affairs and Energy (BMWi) as part of the project “Mittelstand 4.0-Kompetenzzentrum Usability” (Förderkennzeichen: 01MF17013C).

References 1. da Silva, T., Martin, A., Maurer, F., Silveira, M.: User-centered design and agile methods: a systematic review. In: Proceedings of the IEEE Agile Conference 2011, pp. 77–86 (2011). https://doi.org/10.1109/AGILE.2011.24 2. Version One: The 12th Annual State of Agile Report (2017). https://explore.versionone.com/ state-of-agile/versionone-12th-annual-state-of-agile-report. Accessed 14 Aug 2018

Bringing It Together: Three Approaches to Combine Agile Software Development

27

3. Beyer, H., Holtzblatt, K., Baker, L.: An agile customer-centered method: rapid contextual design. In: Zannier, C., et al. (eds.): XP/Agile Universe 2004. LNCS, vol. 3134, pp. 50–59. Springer, Berlin (2004) 4. ISO 9241-210: Ergonomics of human-system interaction – Part 210: Human-centred design for interactive systems. International standardization organization, Geneva (2010) 5. Gothelf, J., Seiden, J.: Lean UX: Applying Lean Principles to Improve User Experience. O’Reilly, Cambridge (2015) 6. Alt-Simmons, R.: Agile by Design: An Implementation Guide to Analytic Lifecycle Management. SAS Institute Inc., Cary (2015) 7. Sy, D.: Adapting usability investigations for agile user-centered design. J. Usability Stud. 2(3), 112–132 (2007) 8. Carlson, D., Soukop, E.: Why is Sprint Zero a Critical Activity? Cross Talk, pp. 35–37, January/February 2017 9. Graser, F.: User experience in der agilen Entwicklung (2016). www.dev-insider.de/userexperience-design-in-der-agilen-entwicklung-a-565788. Accessed 15 Apr 2018

User-Centered-Design Approach to Evaluate the User Acceptance of Seating Postures for Autonomous Driving Secondary Activities in a Passenger Vehicle Sibashis Parida1,3(&), Sai Mallavarapu1,2, Sylvester Abanteriba3, Matthias Franz1, and Wolfgang Gruener1 1

2

BMW Group, Knorrstr. 147, 80788 Munich, Germany {sibashis.parida,matthias.franz, wolfgang.gruener}@bmw.de, [email protected] Technische Hocschule Ingolstadt, Esplanade 10, 85049 Ingolstadt, Germany 3 RMIT University, 124 La Trobe Street, Melbourne 3000, Australia [email protected]

Abstract. With the development of autonomous driving technology, the passenger vehicles of the future, would allow drivers to participate in non-driving secondary activities or use-cases. In order to facilitate the non-driving use-cases, the interior and the seating arrangements of autonomous passenger vehicles needs to be designed accordingly, The aim of the study is to derive common seating positions and angles for the different autonomous driving secondary activities from their original environments and then experimentally evaluate the user acceptance of these postures for each secondary activity in a passenger vehicle. A user experience questionnaire (UEQ) approach is used to evaluate the user acceptance of the seating postures for the mentioned autonomous driving secondary activity or use-cases. The secondary activities considered in the study include; use of laptop, use of smartphone, general reading, window gazing and sleeping. The angles used for the seating postures include; seat-pan angle, back rest angle and neck-flexion angle. Keywords: Autonomous driving Autonomous driving secondary activities Vehicle seating Seating posture Autonomous driving vehicle interior User experience

1 Introduction Autonomous Driving (AD) may prove to be the most important innovation of the automotive industry due to their potential beneﬁts: enhanced safety, reduced accidents, reduced congestion and improved reliability, mobility for non-drivers, reduced driver stress, user convenience, improved fuel economy, improved parking and reduced congestion [1]. One of the biggest advantages of AD is the beneﬁt of freed driving time. In level 5 AD mode (SAE) [2], the driver/user need non concentrate on driving and could © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 28–33, 2019. https://doi.org/10.1007/978-3-030-02053-8_5

User-Centered-Design Approach to Evaluate the User Acceptance of Seating Postures

29

use the travel time otherwise. According to valuable literature, working on a laptop, reading for leisure, using a smartphone, window gazing and sleeping are the most common activities, individuals prefer to spend their travel time on a day to day basis [3– 8]. Hence the need to redesign the interior of fully autonomous vehicles (FAVs). The aim of the study is to derive common seating positions and angles for the different autonomous driving secondary activities from their original environments and then experimentally evaluate the user acceptance of these postures for each secondary activity in a passenger vehicle. The secondary activities considered in the study include; use of laptop, use of smartphone, general reading, window gazing and sleeping. The angles used for the seating postures include; seat-pan angle, back rest angle and neck-flexion angle.

2 Methodology Parida et al. [9] conducted a study to ﬁnd out the commonly used seating angles and postures for various activities including working on a laptop, using a smartphone, relaxing and sleeping. Listed in the table below are angles derived from the study (Table 1). Table 1. Common seating angles [9] Laptop use Smartphone use General reading Window gazing Sleeping

Backrest angle Seat pan angle Neck reflex angle 94º 9º 94º 100º 8º 94º 120° 10° 103° 120° 10° 103° 150° 30° 95°

These set of angles were used as inputs to simulate the seating positions for the corresponding secondary activity in the vehicle. A BMW in-house software EDIBAS is being used to position the seat to desired position for the given activity. The experiment was conducted in the driver seat of a current BMW 7 Series vehicle. The seat used for the experiment is an electric multi-function leather seat. Due to the technical limitations and legislation boundaries, the use of a level 5 Autonomous Vehicle (AV) was not possible for the experiment. Hence, the experiment was conducted in a stationary condition in the BMW research and development center. In order to simulate autonomous driving user scenario, the steering wheel of the vehicle was dismounted and the study was conducted in a driver seat of a left-hand-drive (LHD) vehicle. A total of 51 test subjects participated in the study ranging from 23 years to 60 years age, including 38 males and 13 females. The height of the participants ranged between 158 cm and 203 cm. The subjects were asked to use the seat in the given position and perform the activity for a maximum of ﬁve minutes. After each activity the subjects were asked to step out of the vehicle. This is a precautionary measure, to prevent one seating position influencing the other.

30

S. Parida et al.

Each given position was then evaluated using the UEQ emotional based questionnaire. The UEQ is a widely used method to measure the subjective impression of users towards the user experience of products. Using a 26 item questionnaire the following attributes are measured: attractiveness, perspicuity, efﬁciency, dependability, stimulation, and novelty. Attractiveness is a pure valence dimension (emotional reaction on a pure acceptance/rejection dimension). Perspicuity, efﬁciency, and dependability are pragmatic quality aspects, i.e. they describe interaction qualities that relate to the tasks or goals the user aims to reach when using the product. Stimulation and novelty are hedonic quality aspects, i.e. they do not relate to tasks and goals, but describe aspects related to pleasure or fun while using the product [10]. In order to evaluate if the user experience of the product is sufﬁciently high to fulﬁll the general expectations of users, Schrepp et al. [11] developed a benchmark which contains data with the UEQ. The evaluated products cover a wide range of applications. The following table shows the connection of these categories to the scale means for the UEQ scales. The benchmark contains data from 163 product evaluations with 4818 total participants. With the availability of a benchmark it is relatively easy to decide if the product has sufﬁcient user experience to be successful in the market (Table 2). Table 2. Benchmark intervals for the UEQ scales.

3 Results The UEQ mean scales for each secondary activity is listed in Table 3. The following ﬁgures show the mean scales of each activity with respect to the benchmark values (Figs. 1, 2, 3, 4 and 5)1. Table 3. UEQ Scales for the different activities.

1

All the angles are referenced to global co-ordinate system, which is vertical 90° and horizontal 0°.

User-Centered-Design Approach to Evaluate the User Acceptance of Seating Postures

Use of Laptop 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00

Excellent Good Above Average Below Average Bad Mean

Fig. 1. Mean values of scales for laptop use with respect to benchmark values.

Use of Smartphone 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00

Excellent Good Above Average Below Average Bad Mean

Fig. 2. Mean values of scales for smartphone use with respect to benchmark values.

General Reading 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00

Excellent

Good Above Average Below Average Bad Mean

Fig. 3. Mean values of scales for general reading with respect to benchmark values.

31

32

S. Parida et al.

Window Gazing 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00

Excellent Good Above Average Below Average Bad Mean

Fig. 4. Mean values of scales for window gazing with respect to benchmark values.

Sleeping 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00

Excellent Good Above Average Below Average Bad Mean

Fig. 5. Mean values of scales for sleeping with respect to benchmark values.

4 Discussion and Conclusion The seating positions used for the experiment are positions from their original environment. That is for laptop use, the seating position was referenced to an ofﬁce chair, for window gazing more a relaxed position and for sleeping the zero-gravity-seat. The goal is to simulate these seating postures in a passenger vehicle and evaluate these positions. The results clearly suggest that the proposed positions for the study is being negatively rated. Apart from the attractiveness of sleeping and window gazing posture and the perspicuity and dependability of window gazing all other posture are either rated below average or bad. This concludes that the seating postures from original environments are not recommended for using in a passenger vehicle. Hence, there is a need to conduct a study to derive new seating postures for each given activity in a passenger vehicle. In the planned study, occupants are requested to position their seat with their personal preference, and then asked to rate the posture using the UEQ scale. Acknowledgments. The study was conducted in co-operation with BMW Group, Germany and RMIT University, Australia.

User-Centered-Design Approach to Evaluate the User Acceptance of Seating Postures

33

References 1. Parida, S., Franz, M., Abanteriba, S., Mallavarapu, S.: Autonomous driving cars: future prospects, obstacles, user acceptance and public opinion. In: Stanton (Hg.) Advances in Human Aspects, vol. 786, pp. 318–328 (2018) 2. SAE International: SAE International’s levels of driving automation for on-road vehicles (2014) 3. Lyons, G., Jain, J., Holley, D.: The use of travel time by rail passengers in Great Britain. Transp. Res. Part A Policy Pract. (2007). https://doi.org/10.1016/j.tra.2006.05.012 4. Pfleging, B., Schmidt, A.: (Non-)Driving-related activities in the car: deﬁning driver activities for manual and automated driving 5. Schoettle, B., Sivak, M.: A survey of public opinion about autonomous and self-driving vehicles in the US, the UK, and AU. The University of Michigan Transport Research Institute (2014) 6. Cyganski, R., Fraedrich, E., Lenz, B.: Travel time valuation for automated driving: a usecase driven study 7. Naujoks, F., Purucker, C., Neukum, A.: Secondary task engagement and vehicle automation – comparing the effects of different automation levels in an on-road experiment. Transp. Res. Part F Trafﬁc Psychol. Behav. (2016). https://doi.org/10.1016/j.trf.2016.01.011 8. Kamp, I., Kilincsoy, U., Vink, P.: Chosen postures during speciﬁc sitting activities. Ergonomics (2011). https://doi.org/10.1080/00140139.2011.618230 9. Parida, S., Mallavarapu, S., Franz, M., Abanteriba, S.: A literature review of seating and body angles for non-driving secondary activities in autonomous driving vehicles. In: Stanton, N. (ed.) Advances in Human Aspects of Transportation. Proceedings of the AHFE 2018. Advances in Intelligent Systems and Computing, vol. 786, pp. 398–409. Springer International PU, [S.l.] (2018) 10. Schrepp, M., Hinderks, A., Thomaschewski, J.: Design and evaluation of a short version of the user experience questionnaire (UEQ-S). IJIMAI (2017). https://doi.org/10.9781/ijimai. 2017.09.001 11. Schrepp, M., Hinderks, A., Thomaschewski, J.: Construction of a benchmark for the user experience questionnaire (UEQ). IJIMAI (2017). https://doi.org/10.9781/ijimai.2017.445

User Evaluation of Industry 4.0 Concepts for Worker Engagement Susanna Aromaa1(&), Marja Liinasuo1, Eija Kaasinen1, Michael Bojko2, Franziska Schmalfuß3, Konstantinos C. Apostolakis4, Dimitrios Zarpalas4, Petros Daras4, Cemalettin Özturk5, and Menouer Boubekeuer5 1

VTT Technical Research Centre of Finland Ltd., P. O. Box 1300, 33101 Tampere, Finland {susanna.aromaa,marja.liinasuo,eija.kaasinen}@vtt.fi 2 Department of Factory Planning and Factory Management, Chemnitz University of Technology, Erfenschlager-Strasse 73, 09125 Chemnitz, Germany [email protected] 3 Department of Cognitive and Engineering Psychology, Chemnitz University of Technology, Wilhelm-Raabe-Strasse 43, 09120 Chemnitz, Germany [email protected] 4 Information Technologies Institute, Centre for Research and Technology Hellas, 6th km Charilaou-Thermi Road, 57001 Thessaloniki, Greece {kapostol,zarpalas,daras}@iti.de 5 United Technologies Research Centre Ireland, Penrose Wharf, Cork, Ireland {Ozturkc,boubekm}@utrc.utc.com

Abstract. Industry is undergoing a digital transition that will change the design and setup of human-machine systems. One part of this change is increasing possibilities of workers to influence their work. In this paper, we present four components of the Factory2Fit project that contribute to this change by engaging workers: (1) knowledge sharing and collaboration via a discussion platform; (2) visualisation of information via augmented reality (AR) glasses; (3) participatory design of workplaces and tasks by means of a 3D simulation software programme, and (4) an on-site training tool utilising a training platform. The demonstrators were evaluated with workers to identify foreseen beneﬁts, challenges and impact on their work. Most of the concepts seem to be well accepted and they have high potential to improve work well-being and work performance. The results of this study are encouraging, but long-term ﬁeld studies with actual prototypes will be needed to evolve the concepts. Keywords: Industry 4.0 Smart factory

Acceptance Engagement Work Manufacturing

1 Introduction Human-centricity is vital for engaging workers in smart factories. European Factories of the Future Research Association [1] see that industry is undergoing a digital transition that will change the role of humans and machines. In the future, tasks will be © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 34–40, 2019. https://doi.org/10.1007/978-3-030-02053-8_6

User Evaluation of Industry 4.0 Concepts for Worker Engagement

35

performed in cooperation between knowledge workers and smart manufacturing technologies. Workers need to be supported at their workplace when adapting to upcoming digitalisation and cooperation requirements. This support to the workers can be given efﬁciently by utilising innovative digital technologies. Human-machine interaction needs to be inherently safe. Open innovation and product co-creation by scaling knowledge from the shop floor, and knowledge sharing and creativity with global engineering teams are highly important. New digital technologies should open new possibilities for the workers without hindering the usage of professional abilities the workers already possess. This study is conducted within the EU-funded project Factory2Fit. The project is developing adaptive, empowering and engaging factory floor solutions, which aim to improve both productivity and work well-being. The goal of the study is to identify conceptions that workers have regarding the proposed concepts to engage workers. In addition, foreseen beneﬁts, challenges and forms of impact are studied.

2 Materials and Methods Used for the Study 2.1

Concepts for Engaging Workers

The ﬁrst concept is knowledge sharing and collaboration via a social media platform (later SoMeP). Its goal is to increase knowledge sharing among workers and to support them in problem solving situations. SoMeP is a software that can be used in the factory via smartphones, tablets or PCs. It provides the options to share messages, photographs and videos. The added value to existing solutions comes from linking the discussions with production-related information (e.g., discussions related to a speciﬁc error code). The second concept is the visualisation of information via augmented reality (AR) glasses (later AR-tool). The purpose of this concept is to provide context related information at the workplace. AR glasses (e.g., Microsoft HoloLens) can be used to display interactive interfaces on real and digital augmentation targets, e.g., point towards the location of an alarm. An AR-tool could also visualise information from other software, for example, the SoMeP. The third concept is participatory design by means of 3D simulation software for workplace and task simulation (later 3D-PD). The goal of this concept is to provide workers the possibility to co-design the workplace and plan working practices with other stakeholders. The 3D simulation software (Visual Components) supports the design of workplaces and work tasks. The fourth concept is the on-site training tool utilising a training platform (later training tool). With machine-learning based search and retrieval algorithms, the worker can ﬁnd relevant and focused instructions in video format. The purpose is to provide support and teach work tasks on-site. It can be used with tablets and smartphones (Fig. 1).

36

S. Aromaa et al.

Fig. 1. The four concepts introduced in Factory2Fit to increase worker engagement.

2.2

Methods Applied in the Study

Two workshops were conducted to discuss the four concepts of engaging factory workers. One workshop was held at a machinery manufacturer in Finland and the other in a metrology lab of a components manufacturer in Germany. Eight male participants took part in the workshop in a Finnish company. Their average age was 46 years (range 33–55 years). The group was comprised of one automation designer, two engineers, a production supervisor, a quality expert, an electrician, an automation assembly worker and a worker with responsibilities during the start-up of the machine. Their experience in their current work role was 13 years on average (range 5–33 years). As a whole, the participants were well familiar with the digital technology related to smartphones, tablets and navigators, but were quite inexperienced regarding newer technology (e.g., AR technology). Eight participants (six male and two female) took part in the German workshop. The work of the participants was mainly related to quality-control tasks in a metrology lab. This group was comprised of two measuring machine operators, two shift supervisors, two mechanics, and one quality controller. One participant’s role was undisclosed. Their experience in their current work role was 4 years on average (range 0.2–18 years). The level of knowledge and experience of digital technology was roughly the same level as the previous group. First, the project was introduced on a general level to the participants and they signed consent forms and answered the demographics questionnaire. Then, the four concepts were introduced briefly one by one. In the ﬁrst workshop, the AR tool concept

User Evaluation of Industry 4.0 Concepts for Worker Engagement

37

was demonstrated with HoloLens, but workers did not want to have a hands-on experience about it. Other concepts were illustrated with slides, photos and videos. In the second workshop, all concepts were illustrated with slides, photos and videos without actual hands-on experience. After the presentation, workshops were conducted with researchers as facilitators. Firstly, each participant ﬁlled in the questionnaire regarding his/her experience of the concept by using a 5-point Likert-scale. Thereafter, the concepts were discussed in the group. In the ﬁrst workshop, discussions were conducted in two groups: factory floor workers (three persons) and other stakeholders (ﬁve persons), whereas in the second workshop, all participants discussed together in a single group. Discussion themes were beneﬁts, challenges and improvement ideas of the introduced concepts. Data analysis was based on the Factory2Fit Work Well-being Framework [2], which was developed in this project. In this study, user experience, user acceptance and safety were addressed from the framework.

3 Results In the Finnish workshop, participants liked the SoMeP. They thought that it would be useful in knowledge sharing, for example, between shifts. With this concept, all information would be stored in one place, which was found to be an important beneﬁt. However, they thought that the co-existence of different languages and conversation groups (local vs. global) might become a challenge. As a practical drawback, the participants wondered how the information would be managed if there is lots of it. Furthermore, some people may be reluctant to write messages with their real names. They also suggested that other information could be added there, e.g., supply schedules and the status of backorders. The participants saw potential in the AR-tool when used in training and maintenance, e.g., providing visual instructions while the maintenance worker is using both hands for the task. The discussions revealed workers’ concerns on occupational safety. They proposed that AR equipment should be chosen based on workplace conditions. This would ensure that the workers always have a clear view of their immediate surroundings, and the device materials can withstand the environmental conditions (e.g., heat). These concerns naturally affected user acceptance (Table 1). The participants thought that the 3D-PD is a nice concept and should be used, for example, when designing assembly. By using the tool, it is possible to foresee problems and everyone can participate. On the other hand, they thought that the tool would be used rarely and that it might be difﬁcult to ﬁnd consensus over many differing opinions. The training tool received many positive comments: it was found to be practical, the logic is clear and the participants assumed it functions well and is easy to use. They saw beneﬁts for novice workers, and that the image- and video-based tool would not have any language barriers. It would speed up the work. As a challenge, they saw the efforts needed in creating and maintaining video content. As a new invention, they considered that this tool could be used to identify parts that are missing identiﬁcation codes.

38

S. Aromaa et al.

Table 1. Questionnaire results from both workshops (5 = strongly agree/strong positive experience; 1 = strongly disagree/strong negative experience). Workshop 1 (N = 8)

Workshop 2 (N = 8)

SoMeP AR-tool 3D-PD Training SoMeP AR-tool 3D-PD Training AVG AVG AVG AVG AVG AVG AVG AVG (STDEV) (STDEV) (STDEV) (STDEV) (STDEV) (STDEV) (STDEV) (STDEV) In general, how did you feel about the tool Using the tool would make the job of factory worker more enjoyable Using the tool would make the factory work more interesting Using the tool would make the factory worker feel more competent at work Using the tool could improve the performance of the factory worker The tool would be well accepted among factory workers The usage of tool in the factory floor would support safety

3.63 (0.74)

3.63 (0.74)

4.13 (0.83)

4.38 (0.52)

3.43 (0.53)

2.57 (1.13)

4.86 (0.38)

3.43 (0.79)

3.63 (0.52)

3.25 (0.46)

4.25 (0.71)

4.63 (0.52)

3.25 (0.46)

2.50 (0.93)

4.75 (0.46)

3.50 (0.53)

3.75 (0.46)

3.88 (0.64)

3.63 (0.74)

4.13 (0.83)

3.75 (0.71)

3.13 (0.64)

3.88 (0.35)

3.38 (0.52)

3.13 (0.83)

3.25 (0.71)

3.63 (0.74)

4.13 (0.64)

3.13 (0.83)

2.50 (0.93)

3.25 (1.04)

2.63 (0.52)

3.88 (0.35)

3.50 (0.76)

4.38 (0.52)

4.63 (0.52)

3.25 (0.71)

2.25 (0.46)

4.38 (1.06)

3.50 (0.53)

3.13 (0.64)

2.88 (0.99)

4.00 (1.07)

4.63 (0.52)

3.13 (0.35)

2.13 (0.64)

4.50 (0.53)

3.25 (0.89)

3.38 (0.74)

2.38 (0.74)

3.75 (0.46)

3.88 (0.35)

2.88 (0.83)

1.50 (0.53)

4.38 (0.74)

3.13 (0.35)

User Evaluation of Industry 4.0 Concepts for Worker Engagement

39

In the German workshop, the participants thought that the SoMeP could make sense in maintenance and as an error catalogue. If it were used with a tablet, it would be better available, compared to a PC, and without extra paperwork. However, they were concerned about the quality and correctness of the information and the amount of work it would take to add information. In the measurement lab, there is no time for extra work. There should be instructions and etiquette concerning how to add information to SoMeP and maybe not everyone should be allowed to add content. The AR tool was considered useful in both maintenance and training applications. Similar to the Finnish workshop, the workers underlined the importance of solutions to be tailored towards ensuring occupational safety. In particular, situational awareness should not be hindered by the additional cognitive load of the holograms. This is again reflected in their respective scores shown in Table 1. The 3D-PD was liked the most in the German workshop. The participants supposed it would help in planning the measurement room layout, enabling the trying out of different alternatives without stopping machines. The main concern was the amount of effort it may require to create simulations. The experience of the training tool was positive. The participants said that it could be used in maintenance for simple disturbances. They agreed that the tool could be used to recognise parts which are around the workplace without identiﬁcation information. Some concerns were related to the accuracy: if there are similar parts with only small variances, does the system identify them correctly?

4 Discussion and Conclusions The purpose was to identify conceptions that the workers have regarding the four concepts to engage workers including beneﬁts, challenges and foreseen impacts on their work. The results of the two workshops in the two countries are somewhat similar and cultural differences were not observed. However, the differing nature of the work tasks had an effect on their preferences; the quality control work with precision and detailed processes is different from manufacturing machines in general. Based on the results, user experience of the concepts was positive. The participants agreed that most tools would make the work more interesting and enjoyable. The participants especially liked the 3D-PD and the training concepts. Regarding user acceptance, most concepts were considered to become well accepted, indicating that as concepts, they would be accepted and possible issues, if any, relate to practicalities realized at the workplace. Presumably, as AR technology is not familiar among workers, its’ acceptance was not as high as with other tools. However, based on the comments during the workshops, means to raise user acceptance were found, such as by pre-training users in the use of AR technology and by designing the tool in a manner, which supports occupational safety and situational awareness. Some concepts were seen to improve safety. However, in the SoMeP concept, the moderation of the information was seen important to prevent sharing incorrect information, which could lead to accidents. Similarly, the image recognition in the training tool needs to be accurate to prevent the possibility of using false instructions. Safety

40

S. Aromaa et al.

issues were also raised in the AR use, indicating that increasing the ﬁeld of view is needed to mitigate safety hazards. Most of the concepts to engage workers seem to be well accepted and they carry a high potential to improve work well-being and work performance. The results of this study are encouraging and it was also possible to ﬁnd remedies to the identiﬁed challenges. However, long time ﬁeld studies with actual prototypes will be needed to develop the concepts further with actual users. Acknowledgments. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 723277 (project Factory2Fit). This paper reflects only the authors’ view and the Commission is not responsible for any use that may be made of the information it contains. The authors are grateful to all researchers, concepts’ developers and company representatives who have contributed to and supported the work presented in this publication.

References 1. EFFRA - European Factories of the Future Research Association: Factories 4.0 and Beyond: Recommendations for the work programme 18-19-20 of the FoF PPP under Horizon 2020 (2016) 2. Kaasinen, E., Liinasuo, M., Schmalfuß, F., Koskinen, H., Aromaa, S., Heikkilä, P., Honka, A., Mach, S., Malm, T.: A worker-centric design and evaluation framework for operator 4.0 solutions that support work well-being. Accepted paper to Human-Work Interaction Conference, Espoo, Finland (2018)

FatigueWatcher: Interactive Fatigue Detection for Personal Computer and Mobile Device Ayumu Tanaka, Takashi Yokogawa, and Hiroaki Tobita(&) Advanced Institute of Industrial Technology, 1-10-40 Higashioi, Shinagawa-ku, Tokyo, Japan {a1623at,a1630ty,tobby}@aiit.ac.jp

Abstract. We introduce our FatigurWatcher that detects fatigue of PC and mobile device users. To achieve our system, we use a web camera embedded into most of PC and mobile devices. The system captures user’s face in front of the devices and then quantiﬁes face characters such as blinks, yawns, and facial inclinations every second. Although our fatigue detection is simple, we could achieve enough accuracy to use it for a wide variety of applications. Moreover, through our initial evaluations, we were able to determine that the number of blinks is the strongest indicator to detect possible fatigue. By running Bayesian estimations from the obtained data, we were able to determine that this system can detect fatigue with an accuracy of 87.5%. In this paper, we describe our FatigureWatch focusing on the concept, implementation, and evaluation. Keywords: Fatigue detection Facial features Image processing Health care Personal computer Mobile device Interactive system Bayesian estimation

1 Introduction With the development of computer and network technologies, people can work anytime and anywhere. Especially, working time with PC has been getting longer than past decade. Recently, workaholic has been serious problems for ofﬁce workers, because it is difﬁcult for them to notice their fatigue by themselves (Fig. 1 (left)). Accumulation of fatigue by ofﬁce workers caused by long working hours, which in the worst case can end in death, is a raising social problem in modern society. Unfortunately, most of the works pay attention their work output rather than their health. Also, many people play online video games with smartphone. During playing video games, people are too concentrating to forget time and their health. However, there is no effective technique to detect users’ fatigue for personal computer (PC) and mobile device. To tackle these problems, several researches on fatigue detection have been developed, especially focusing on driving cars. DepeiBao’s study [1] proposed to use monocular cameras to measure drivers fatigue by analyzing the driver eyes and face direction. On the other hand, Sultan Alhazmi’s [2] focused on driving parameters such as the angular velocity of handling, the power of stepping on accelerator and breaking pedals, and the power of gripping the handle. However, there is no method to detect fatigue for PC and mobile device users and workers. As PC and mobile devices © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 41–47, 2019. https://doi.org/10.1007/978-3-030-02053-8_7

42

A. Tanaka et al.

Fig. 1. Fatigue sign such as facial inclinations and yawning (left) and Facial characteristics (right).

contains many applications, people naturally depend on the devices in their daily life. We can easily predict that spending time using such device would increase. Thus, it is necessary to detect fatigue and to provide effective applications to maintain user’s health. Therefore, we have developed our FatigureWatcher that detects user’s fatigue through simple web camera attached to PC and mobile devices. Our goal is to reduce work time by detecting fatigue, so we are interested in whether it is possible to detect fatigue through simple image processing and what kind of element is the most effective to detect fatigue. To know whether it is possible to detect fatigue through simple image processing, we used the equipped by standard camera on personal computers to determine the face of users, analyze number of blinks, yawns and facial inclinations every second. We focus on six facial features (e.g., upper eyelid, lower eyelid, distance between the eyes, nose bridge, upper lip, and lower lip). To know what kind of element is the most effective to detect fatigue, we evaluated our prototype system focusing on the average number of blink, facial inclination, and yawning. Based on the results obtained from data equivalent to 25 h from 8 volunteers, we were able to determine that the number of blinks is the strongest indicator to detect possible fatigue. Also, by running Bayesian estimations from the obtained data, we were able to determine that this application can detect fatigue with an accuracy of 87.5%. Although our fatigue detection is simple, we could achieve enough accuracy to use it for a wide variety of applications. For example, once fatigue is detected, the application suggests stop working via prompting a message on the personal computer. We think this method would be useful to restrict work time and keep user’s health.

2 Fatigue Detection 2.1

Facial Characteristics

We implement our system as an application software for PC and mobile device that contains web camera on the top of the display. The application works behinds main applications, and the camera captures user’s face every second. Our current prototype supports Mac OS and iOS. We use Open Source Computer Vision Library (OpenCV) for image processing to detect user’s face and the facial features. We focus on six

FatigueWatcher: Interactive Fatigue Detection

43

characteristics such as upper eyelid, lower eyelid, distance between the eyes, nose bridge, upper lip, and lower lip (Fig. 1 (right)). The system evaluates users fatigue by analyzing the six points from images captured by the embedded camera on PC [3]. Figure 2 shows the overview of our system. Figure 2 (left) shows how our system recognizes user’s fatigue. After capturing images by the camera, the system retrieves facial features in the images, and detect fatigue by using Bayesian network model. After calculating each data, our system stores such data in a database focusing on the 4 fatigue characteristics values (e.g., eye, line, mouse, and time) (Fig. 2 (right)).

Fig. 2. System overview: The system analyzes captured image with Bayesian network (left), and stores 4 types of information in a database (right).

2.2

Fatigue Characteristics Values

Next, we describe how to calculate fatigue characteristics values in detail. We calculate fatigue elements such as inclination, blinks, and yawns by the combination of six facial characteristics. We determine each element as the flowing rules. Distance Between Upper Eyelid and Lower Eyelid f1 The difference between the upper eyelid’s y coordinate a1 and the lower eyelid’s y coordinate a2. In order to normalize the calculation result without being affected by the size of the face image, a coefﬁcient L1 needs to be multiplied. L1 is calculated by dividing the image height H by the difference between the distance between eye’s. y coordinate b1 and the nose bridge’s y coordinate b2. L1 ¼ H=ðb1 b2 Þ; f1 ¼ L1 ða1 a2 Þ Facial Inclination f2 Calculated as the difference between x coordinate of distance between eyes c1 and nose bridge’s x coordinate c2 (Fig. 3 (1)). To avoid that the characteristics values are affected by the face image size, the value is normalized by multiplying coefﬁcient L2, which is the result of the division between the width W of the image and the horizontal distance of the face (d1 – d2). L2 ¼ W=ðd1 d2 Þ; f2 ¼ jL2 ðc1 c2 Þj

44

A. Tanaka et al.

Fig. 3. Calculation of Facial Inclination, Blinks and Yawns.

Blinks f3 When the distance between the y coordinate of the upper eyelid a1 and the y coordinate of the lower eyelid a1 is 0, it is deﬁned as a blink f3 = 1 (Fig. 3 (2)). Any other case is deﬁned as f3 = 0. Yawns f4 When the value acquired from multiplying L to the distance between the y coordinate of the upper lip and the y coordinate of the lower lip is more than 200 for more than 2 s it is deﬁned as a yawn f4 = 1 (Fig. 3 (3)). Any other case is deﬁned as f4 = 0.

3 User Test 3.1

Goal

We calculated three facial elements such as facial inclination, blinks, and yawns. We are interested in two questions. – Do facial elements classify between focus and fatigue states? – Which is the most effective facial element to determine fatigue? We think there are mainly two states for PC users: focus and fatigue. The focus state means that a user just concentrates on PC works, and fatigue state means that a user is tired. Thus, we are interested in whether our system can classify between two states. Next, we examined how the system detects user’s fatigue correctly. We think the accuracy is important to achieve applications based on user’s fatigue. We focus on three facial elements, so we are interested in which element of the facial characters or combination between different features is the most effective to detect user’s fatigue. To evaluate our system, we deﬁned three states: start, focus, and fatigue states as the followings. (1) The start state is 10 min from starting working. (2) The fatigue state is 10 min before the user takes a break. (3) The focus state is any other work time except the above two states.

FatigueWatcher: Interactive Fatigue Detection

3.2

45

Participants

We recruited eight participants—six men and two women, aged 30–40 years—from local companies. Though not all participants were students in computer science, they were familiar with computer manipulation. All participants had extensive experience with applications in PC. They usually use the devices more than 8 h. 3.3

Tasks

We are interested in detecting fatigue in their natural environment. Each participant run both our system and his/her main application software that supports his/her task in his/her own PC. The system collects and analyzes facial images every second. We analyzed the average number of blinks, facial inclinations and yaws per minute per each group of data. 3.4

Setup

Each participant installed the developed application in his/her PC and used it in his/her work environment with his/her own desk and chair. 3.5

Summary of Experimental Findings

We show three ﬁgures focusing on the average number of blink, facial inclination, and yearning resulted in our user tests. Each ﬁgure shows counts per minuets and contains both focus and fatigue state. Figure 4 (1) shows the result of the average number of blinks. In this case, the system detected both focus and fatigue. In most of the participants, the fatigue parameter was more than the focus one. Figure 4 (2) shows the result of the average number of facial inclination. In this case, we can see only fatigue parameter of half of the participants. Thus, we cannot ﬁnd signiﬁcant different between focus and fatigue counts. Figure 4 (3) shows the result of the average number of yawning. Although the ﬁgure shows both focus and fatigue parameters, the result depended on the participant. Some participants showed an increase when in focus state, the others showed an increase when in fatigue state. From the result of the user tests, we found that the average blink numbers related with fatigue comparing the facial inclinations and yawning. Next, we composed a Bayesian network with each data to estimate fatigue. Table 1 shows the percentage of times that fatigue was detected correctly within the Bayesian network. We were able to determine that the number of blinks is the strongest indicator to detect possible fatigue. By running Bayesian estimations from the obtained data, we were able to determine that this application can detect fatigue with an accuracy of 87.5%. On the other hand, the accuracy of detecting facial inclination and yawning is 68.76% and 50.0% respectively. About a couple of cues such as blink and facial inclination, blink and yawning, and facial inclination and yawning are less than 87.50%. Mover, by concerning all elements, the accuracy of fatigue was 75.0%.

46

A. Tanaka et al.

Fig. 4. Average blink numbers (1), average facial inclination numbers (2), and average yawning numbers per a minute of 8 participants (3). Table 1. Correctly detected fatigue rate per cues with Bayesian network. We evaluated single, two, and all cues respectively. Single cue Cues Blink Inclination Yawning

Result 87.50 68.75 50.00

Two cues Cues Blink & Inclination Blink & Yawning Inclination & Yawning

ALL Result Cues Result 75.00 ALL 75.00 81.50 62.50

4 Conclusion and Future Works In this paper, we described our FatigueWatcher that detects fatigue of PC and mobile device users and divides captured facial images into “Fatigue” and “Focus” states. We mentioned our concept and implementation focusing on how to determine fatigue. We also showed the result of our initial user tests. Through our initial evaluation with Bayesian network, we could ﬁnd blink was a signiﬁcant element to determine fatigue and detect fatigue correctly in 87.5% of the cases. To increase the accuracy rate of fatigue, we are planning to add other parameters not related to the face characteristics such as physical condition. Also, we will integrate our approach with practical software applications. In such environment, the system warns user’s fatigue effectively as a part of the software application.

FatigueWatcher: Interactive Fatigue Detection

47

References 1. Bao, D., Yang, Z., Song, Y.: Projection function for driver fatigue monitoring with monocular camera. In: Proceedings of ACM SAC, pp. 11–15, March 2007. https://doi.org/10.1145/ 1244002.1244022 2. Alhazmi, S., Saini, M., El Saddik, M.: Multimedia fatigue detection for adaptive infotainment. In: Proceedings of ACM HCMC, pp. 15–24, October 2015. https://doi.org/10.1145/2810397. 2810440 3. King, D.E.: Dlib-ml: a machine learning toolkit. J. Mach. Learn. Res. 10, 1755–1758 (2009)

Introduction of Service Design in a Public Hospital’s Medical Oncology Service Carlos Romero-Piqueras1(&), Jorge Sierra-Pérez2, and Eduardo Manchado-Pérez1 1

Design and Manufacturing Engineering Department, Zaragoza University, C/Maria de Luna 3, 50018 Zaragoza, Spain {carlos.romero,manchado}@unizar.es 2 Centro Universitario de la Defensa, Ctra. De Huesca s/n, 50090 Zaragoza, Spain [email protected]

Abstract. The purpose of this paper is to present an initial phase of the introduction of service design practices in a Spanish public hospital. Service design helps to innovate and improve the users’ experience and processes that build a service. Experience is deﬁned as interactions and involvement of the users with the service, while the processes represent all the operations, protocols, and work tools needed to deliver the experience. Offering experiences focused on patients is considered as a synonym of quality patient care. Facing the design process focusing on user needs is one of the principles of service design so their applications to public health care services can have a positive impact on patients’ experience. As a ﬁrst step to introduce service design widely in the hospital, a pilot case was proposed to redesign one of the hospital’s services, to then extend the project to further potential services. To decide which hospital service was likely to be redesigned, several factors were taken into account, as the interest for the hospital, the patients and the contribution of the research. Keywords: Service design Oncology Healthcare

Patient experience User-centered design

1 Introduction Service design helps to innovate or improve services to make them more useful, usable and desirable for customers, as well as make them efﬁcient and effective for organizations. The process and tools used in service design involve different stakeholders of the service from the process of creation [1], in order to obtain better results in the implementation of the solutions and an overall improvement in the relationship between service providers and users [2]. Service design is thus considered a holistic, interdisciplinary and integrating ﬁeld [3]. The work includes Design Thinking methodology, which improves and innovates services beginning with the analysis of the experience of the people implied among them [4]. This type of design methodology has a high potential in the process of creating private and public services since it has the

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 48–53, 2019. https://doi.org/10.1007/978-3-030-02053-8_8

Introduction of Service Design in a Public Hospital’s Medical Oncology Service

49

capacity to integrate knowledge of several disciplines such as marketing, human resources, operations, organizational structure and technology [5]. Within public services, an incipient ﬁeld of application linked to the development and analysis of services is the area of health care. At designing health services, outstanding cases of work such as the mourning room for relatives of the University Hospital Valle de Hebron are ﬁnd. Another remarkable hint is the Philips Healthcare project that revolutionizes the experience of body scanning for children through a playful process in which children learn to scan stuffed animals and understand the process they will experience later [6]. Another relevant project is the set of services incorporated into the Hospital de Sant Joan de Déu in Barcelona, such as animal therapy, the volunteer service to maintain the school routines or the internal consultation spaces, all of which result of projects in which various design methodologies have been applied to the development of speciﬁc cases [7]. By patient experience we mean: a set of experiences, sensations, evaluations and patient satisfaction with regard to health services; and the result of the phenomenon of interaction with the people and points of contact (spaces, objects, information, etc.) that conform this service. Several authors consider providing patient-centred experiences [8] to be synonymous to quality patient care and a relevant holistic approach based on human values. The working principles of service design, adapted from those stated by various authors [9] and applied in this context are: (1) People, patients, families and health care workers are at the heart of the design process. Their needs are studied from a mainly qualitative approach. (2) The projects are addressed in multidisciplinary working groups in which the different agents involved in the services identify problems and cocreate solutions. (3) The services are studied as a sequence of events, identifying “moments of truth” and points of improvement. (4) The solutions are prototyped and constantly evaluated in the design process in order to learn from mistakes and continue improving. (5) Development of work is under a holistic approach that allows a global vision of the service in its context and on a larger scale. This is the conceptual framework from which we propose and base the interest of approaching a project of design of services for the continuous improvement of the patient experience in Miguel Servet University Hospital. After a process of analysis of the different hospital services that could be redesigned, it is concluded that the Oncology Service represents an opportunity to undertake a pilot project.

2 Methodology The methodology used is the action-research model following Coghlan and Brannick [10]. In this collaborative approach the researcher solves a problem of a client, (in this case redesigning a hospital service and generating scientiﬁc know-how). The model stages are: (1) Context and purpose, where the action will take place and what are the purposes of the researcher and the organization. (2) Diagnosis, justiﬁcation of the need for the action. (3) Action Planning, phases and action tasks. (4) Implementation of the action. (5) Evaluation of the action, and analysis of the results of the action. This paper describes steps 1, 2 and 3 of the case presented.

50

2.1

C. Romero-Piqueras et al.

Context and Purpose

The research development context is a reference hospital in the city of Zaragoza (Spain), the Miguel Servet University Hospital, with 1400 beds and 28 operating rooms. This hospital has a staff of around 5000 professionals, caring a population of approximately 400000 inhabitants, belonging to 21 health areas, (4 of them rural areas), in addition to providing support to other Spanish provinces. The group formed by the primary health centers and the Hospital is known as Sector II. Sector II has a Citizen Participation Group, which, together with the Management of the Sector, the Patient Care Unit and the Nursing Directorate, have been key factors in framing the context of the service to be redesigned and its purpose. These stakeholders expressed the need to evolve the culture of the hospital towards the patient experience. Thus, the research team proposed the introduction of service design as a basis for managing a patient-centred hospital culture, developing a ﬁrst step for a bigger working methodology that the hospital will incorporate to create a future Observatory of Patient Experience. This ﬁrst step required for the creation of the Observatory of Patient Experience is to have at least a ﬁrst case of partial success by means of the application of service design in a project carried out in one of the hospital’s services. The objective of this pilot or prototype is to lay the foundations of the work methodology to be followed in the future. To decide which hospital service was most likely initially to be redesigned, different methods of data collection were used: semi-structured interviews, contextual observation and a focus group, which served to conclude in the need to tackle the case study of the oncology service. 2.2

Diagnosis

In this phase, one of the main objectives is to communicate the project proposal to the medical oncology service team, gaining their implication by showing the potential of service design. Next step is to form a team of healthcare personnel to collaborate with the research team to facilitate the different research actions and to describe a ﬁrst diagnosis of the state of the service from the point of view of the service personnel. A series of presentations were made so that the service staff could understand the potential interest of the project and its character as a pilot project for a more ambitious project (the Observatory of Patient Experience). The beneﬁts of designing health services using a theoretical framework and some success stories of similar experiences carried out in other hospitals were presented and argued. This process served to activate the interest of the staff in the project and its subsequent complicity with the working group. A series of semi-structured interviews were carried out with each of the members of the working group, one hour long interviews, which were recorded for later analysis by the research team and which served as a ﬁrst diagnosis of the service and its functioning. This information served to begin to deﬁne the process of actions to be taken to understand the oncology patient’s experience.

Introduction of Service Design in a Public Hospital’s Medical Oncology Service

2.3

51

Action Planning

The previous phase allowed the research team to have enough knowledge to propose a series of actions that would improve the oncology service and at the same time improve the patient’s experience. Action planning is divided into three blocks: (1) Learning about the cancer patient’s experience. (2) Mapping the patient experience using the service blueprint tool, which represents the different patient actions, contact points and staff interactions, as well as the work of processes not visible to the patient [11]. (3) Cocreate the redesign of the service by involving the work team, patients and relatives in the creation process. The actions taken to learn about the patient’s experience began with an immersion session in the context in which a member of the research team experienced the process of a ﬁrst impact consultation in which the patient is notiﬁed of his or her pathology and the treatment process. The action helped the researchers to generate greater empathy with the type of patient. Contextual observation sessions, patients and families were observed in waiting rooms, consultations, treatment rooms and oncology plant rooms. The data collected in these actions were the basis for developing an interview guide to deepen the problems and needs of patients. Subsequently, 30 in-depth interviews were conducted with patients and families. Prior to mapping the patient experience, focus groups with doctors and nurses were conducted. Based on this work, a ﬁrst version of service blueprint was developed and shared with more than 60 staff members for reﬁnement. The research team then represented the service in a ﬁnal blue print identifying the main areas for improvement. Table 1 shows a summary of the actions carried out so far. Co-creation phase has not been carried out yet, but will be carried out shortly through participatory workshops using Design Thinking tools. Table 1. Summary of actions, stakeholders and purposes Action Context immersion Context observations In-depth interview Focus group interviews Service blueprint

Stakeholders Research team Research team Research team, patients and relatives Research team, nurses and medical team Research team, nurses and medical team

Purposes Generate empathy Environment and products insights Patients and relatives insights Nurses and medical team insights Patient’s experience visualization

3 Results and Discussion The results obtained in the process of describing the patient’s experience can be studied from various perspectives and require classiﬁcation, using the different approaches described by DiJulius [12]. From environmental, experimental, functional, operational

52

C. Romero-Piqueras et al.

and technical approach. The results related to the environment: everything related to sign posts, accessibility, lighting, acoustics, furniture and entertainment. The experimental results: team members actions when they are interacting with the patient, such as shared decision making, vocabulary, empathy, treatment, aptitude to help or provision of information. The functional results: the ease of accessing to the service for the patient, hours of operations, professional accessibility and procedures. The operational results: team members’ actions must be taken behind the scenes. And ﬁnally, the technical results: level of experience of employees and work tools. Environmental. An oncology service is supported by different spaces, some of which are key to the patient experience, such as waiting rooms, consultation rooms and treatment rooms. The need for waiting rooms to promote the relationship between people, relaxation and privacy of patient data was noted. The consultations must also be humanized, elements such as light, paintings and plants are valued by patients and families, as well as a layout of the furniture that promotes a more empathetic approach and treatment by health personnel. With regard to treatment rooms, it is important to differentiate between two types of patients, those who want to rest and relax and patients who require accompaniment, entertainment and relationships. Experimental. The need for information throughout the whole process is one of the main demands of patients. Knowing what is the next step in treatment in the short and long term, knowing what needs and side effects will be derived from the treatment process and being actively involved in decision-making are very important issues, but equally important is how this information is provided. Physicians’ understanding of how to inform each type of person and more empathetic treatment are required, especially in consultations involving the communication of results or ﬁrst impact consultations. Functional. The improvements regarding access to the service are: An appointment system that takes the patient’s personal life into account. Telephone attention with a longer schedule to solve doubts and give support to the care at home. An escort service for people who have no family and for the ﬁrst consultations and treatments of new patients. Families want to be able to access scheduled interviews with doctors. One of the most desired changes is a special circuit of the hospital emergency department for the oncology patient allowing direct access from the emergency department to the oncology department. Operational and Technical. At the operational and technical level, the staff requires greater communication and teamwork, considering the creation of autonomous and multidisciplinary teams that work together with the appointment services, organizing realistic consultation agendas adapted to the needs, which can result in a better distribution of time with fewer waiting times for the patient and better care in consultation. Some work tools need to be updated and certain processes digitized, eliminating paperwork that creates problems of access to information and loss of time. The training of new personnel in the service is another point to work on, other aspects that require extra training are the management of the end of life and the accompaniment in mourning.

Introduction of Service Design in a Public Hospital’s Medical Oncology Service

53

4 Conclusions Service design is a powerful methodology for implementing changes in public services, achieving more fluid and meaningful user experiences. While improving the functioning of organizational processes it is able to help to a best coordination of the work of staff, obtaining better efﬁciency. In the phases of research carried out in this project, some beneﬁts were obtained by merely carrying out the work itself: patients and families have valued very positively the empowerment that the design of services provides by involving them as a fundamental part of redesigning and improving the service. This also generates some expectations that will have to be fulﬁlled not to create frustration. The staff has also stated how the research process itself is already useful in setting in motion changes that can be easily assumed on a day-to-day basis. As future work, the co-creation actions and points 4 (Implementation) and 5 (Evaluation) of the methodology are still pending. Once this process is completed, it will be scaled to be applied to other hospital services and, if possible, to other public hospitals in the surrounding area.

References 1. Williams, I.: Travelling an unfamiliar road: Implications for the entry of design practitioners into healthcare. Diss. Queensland University of Technology (2016) 2. Grifﬁoen, I., Melles, M., Stiggelbout, A., Snelders, D.: The potential of service design for improving the implementation of shared decision-making. Des. Health 1(2), 194–209 (2017) 3. Moritz, S.: Service Design: Practical Access to an Evolving Field. Köln International School of Design (2005) 4. Brown, T.: Change by Design: How Design Thinking Transforms Organizations and Inspires Innovation, vol. 31 (2009) 5. Patrício, L., Gustafsson, A., Fisk, R.: Upframing service design and innovation for research impact. J. Serv. Res., 1–14 (2017) 6. KittenScanner. Involving kids in their own care. https://www.90yearsofdesign.philips.com/ article/30 7. SJD Barcelona Hospital. Pet-assisted Interventions. https://www.sjdhospitalbarcelona.org/ en/pet-assisted-interventions 8. Edvardsson, D., Winblad, B., Sandman, P.: Person-centred care of people with severe Alzheimer’s disease: current status and ways forward. Lancet Neurol. 7(4), 362–367 (2008) 9. Stickdorn, M., Schneider, J., Andrews, K., Lawrence, A.: This is Service Design Thinking: Basics, Tools, Cases, vol. 1. Wiley, Hoboken (2011) 10. Coghlan, D., Brannick, T.: Doing Action Research in Your Own Organization. SAGE Publications, London (2014) 11. Shostack, L.: How to design a service. Eur. J. Mark. 16(1), 49–63 (1982) 12. DiJulius, J.R.: What’s the Secret?: To Providing a World-class Customer Experience. John Wiley & Sons, Hoboken (2011)

Early Validation of User Needs in Concept Development: A Case Study in an InnovationOriented Consultancy Marianne Kjørstad1(&), Kristin Falk1, Gerrit Muller1, and José Pinto2 1

University of South-Eastern Norway, Hasbergs vei 36, 3616 Kongsberg, Norway {Marianne.Kjorstad,Kristin.Falk,Gerrit.Muller}@usn.no 2 Semcon Devotek AS, Dyrmyrgata 47, 3611 Kongsberg, Norway [email protected]

Abstract. An innovation consultancy applies human-centered methods to explore user needs in the early phase of concept development. This paper compares methods applied by the consultancy with theory from the body of knowledge within Design Thinking and Systems Engineering. The basis for this research is observations and interviews for three speciﬁc cases for three different customers. This paper presents criteria and impacting factors on how effective the innovation consultancy performs early validation of user needs. A properly planned co-creation session with the customer is the core. Using a key driver graph we found the main impacting factors to be research on user needs, technology and market trends, techniques used for analyzing the problem and solution domain, selection of participants, and the competence of the facilitator. We conclude that in these three cases the methods are effective in communicating innovative ideas and concepts with the purpose of early validation of user needs. Keywords: Human centered design Systems Engineering Early validation Design Thinking User needs Innovation Co-creation sessions

1 Introduction This paper presents a case study in an innovation-oriented consultancy for early validation of user needs in the concept phase. The innovation consultancy develops concepts, prototypes, and full-scale systems to customers within several domains. Early validation of user needs is essential to avoid costly design changes and to develop systems that fulﬁll their purpose for humans. Systems Engineering emphasizes the importance of identiﬁcation of stakeholders, among them the users, and their needs to understand all perspectives related to the system of interest. These user needs must be identiﬁed and clearly communicated. A challenge is that the softer human values may lose in a trade-off with the more speciﬁc technical requirements. The following research question is the foundation of this research: how effective does the innovation consultancy apply the methods for early validation of user needs in the concept phase? © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 54–60, 2019. https://doi.org/10.1007/978-3-030-02053-8_9

Early Validation of User Needs in Concept Development

1.1

55

Research Methodology

Case studies [1] form the basis for this research within an innovation consultancy providing innovation services to customers within different domains. This research focuses on three speciﬁc cases for three different customers. To determine what impacts the effectiveness of the early validation method used by the innovation consultancy, we ﬁrstly conduct a literature review on the state of the art of the various early validation methods. Through observations, interviews and discussions with technical engineers and designers, we investigate how the innovation consultancy performs early validation of user needs and why they are doing it this way.

2 State of the Art Early Validation of User Needs Early validation of user needs is a fundamental concept within Systems Engineering and Design Thinking. A major difference between the approaches is the applied industrial domain. Systems Engineering validates user needs by reviewing user requirements with customers and/or users [2]. Furthermore, the Systems Engineering approach applies ConOps [2] and/or OpsCon [3] to describe the operational concept of a system using scenarios. Traditionally, ConOps and OpsCon are highly textual-based methods originating from the defense industry. Several variants of the ConOps use less text and are less time consuming, such as agile ConOps [6] and illustrative ConOps [4]. Stakeholder analysis is applied for early validation in the Systems Engineering approach [5]. Storytelling and narratives [6] are early validation methods of user needs applied within Systems Architecting and agile forms of Systems Engineering, but also common in consumer-, IT-, and health care domain. These methods are used to understand the context of use. Conceptual modeling is another early validation method commonly applied within Systems Architecting [7]. This method provides an early validation of the most relevant quality attributes at customer/operational level. Rapid prototyping is typically used within Design Thinking [8, 9]. This provides quick and dirty validation of ideas using low-cost equipment in rapid iterations. Design Thinking is also advocating for releasing prototypes into the market in order to validate user needs at an early stage [10]. Virtual prototyping is another type of early validation method, based on a visual or software model of the system [11]. Business modelling canvas [12], value proposition canvas [13] and the Lean Canvas [14] are techniques used for early validation in business theory among others based on a lean approach. Within the IT and enterprise sector, we ﬁnd the workflow analysis used for early validation as it provides a systematic way of mapping the use of the system.

56

M. Kjørstad et al.

3 Early Validation Using Co-creation Sessions Based on the state of the art of early validation of user needs, we ﬁnd existing methods that have proven applicable, and useful in several domains such as defense and aerospace. Based on the innovation consultancy’s need for rapid validation in concept phase, we ﬁnd the traditional early validation methods within Systems Engineering to be time consuming and comprehensive. The innovation consultancy has developed innovation services for early validation of user needs which are heavily inspired by more rapid approaches, such as Design Thinking [8, 9], Systems Architecting [6] and business theory [12, 13]. The innovation consultancy offers co-creation sessions to customers for early validation of user needs for technological product development. Their vision is building the brand of an innovation consultancy that provides product development based on human behavior. Figure 1 shows the co-creation session with main input and output.

Fig. 1. Co-creation session with input and output

The insight phase on the left side is focusing on the three aspects of innovation; technology, business, and human values. This phase is an important input to the cocreation session, and typically includes stakeholder mapping and analysis, ﬁeld visits, interviews, research on market trends and enabling technology. The duration of this phase is 1–2 weeks and the consultancy performs it in close communication with the customer. The co-creation session with the customer is a direct interaction with several stakeholders from the customer and lasts for 1–3 days. The consultancy carefully plans the agenda of the session, choosing techniques from an internal library and adapting to the context and the participants. They select participants based on experience, role and if possible on personality. External stakeholders may participate if the consultancy expects added value from their participation. One or two people from the consultancy facilitate the co-creation session. The session may include other participants from the consultancy as well. The delivery phase involves mapping and analyzing all collected data produced in the co-creation session. Typically, this phase has a duration of 1–3 weeks and the consultancy performs it. The outcome is reﬁned concepts and a plan for further actions.

Early Validation of User Needs in Concept Development

57

We base our research upon three separate cases within three different domains. Table 1 describes the proﬁles of each of our cases. Table 1. Co-creation session case proﬁles Case no. 1.

Domain

Objective

Participants

Cabin tourism

Innovative cabin resort

2.

Chemical plant

Increased loading efﬁciency

3.

Demolishing plant

Effective and efﬁcient demolishing

7 (customer), 2 (consultancy) 8 (customer), 5 (consultancy) 18 (customer), 4 (consultancy)

4 Criteria and Impact Factors of Co-creation Sessions By using a key driver graph, we derive the criteria for an effective method from the innovation consultancy’s perspective. The key drivers provide the objectives of applying the method, which makes them good candidates as criteria for evaluation. We discuss how these factors play a role in achieving the criteria. The key drivers shown on the left-hand side in Fig. 2 derive from observations and interviews of facilitators and participants of the three separate cases within the innovation consultancy. The key drivers represent the criteria of effectiveness for the early validation of user needs. To realize the key drivers (criteria) the innovation consultancy applies several application processes using the means shown on the right-hand side in Fig. 2.

Fig. 2. Key driver graph of the co-creation session

58

M. Kjørstad et al.

Enable thinking outside the box to generate new ideas is a key driver. The innovation consultancy applies several means to realize this key driver, such as a warm up exercise that contribute to the application drivers remove self-induced limitations and provide a creative and trusting environment. The selection of participants for the co-creation session is a mean to a creative and trusting environment, as well as a broad perspective to realize the key drivers enable thinking outside the box and build a common understanding of user needs. The means shown on the right-hand side in Fig. 2 are representative of the impact factors on the effectiveness of the early validation method. An important part of cocreation sessions is playful (warm up) exercises aiming to prime the participants with some subconscious information and set the mood to achieve the session’s goals. Examples of these exercises include describing one’s superpowers as a superhero, explaining why one has gotten an imaginary gift from another or a physical activity that require negotiating a team strategy to win a competition. These types of tasks function as an exercise to remove self-induced limitations, open for creativity and create a trust between the participants. The equipment and the room(s) used in the co-creation session need to support more practical issues like a large wall for mapping activities, sticky notes in diverse colors, drawing ink instead of common pens to minimize use of word on sticky notes (be speciﬁc and easier to read for all), rapid prototype equipment like tape, carton, and paint. The location of the co-creation session is also important to remove the participants from their everyday controlled working environment. This creates space for wonder, curiosity, and play. Building the problem landscape takes place in the ﬁrst phase of the co-creation session and building the solution landscapes takes part in the later phase. The innovation consultancy has experience with various techniques for this purpose, such as user research based on interaction and interviews with users in their operational context (part of insight phase prior to co-creation session), canvas for eliciting user needs, mapping current and better view of the situation on a timeline, and canvas for understanding pains and gains. When the focus is turning more towards the solution landscape, the innovation consultancy applies techniques for ideation, evaluation, and selection of ideas. These techniques include categorization of ideas based on effort, tangible value (revenue) intangible value (brand awareness or customer loyalty), selecting ideas by voting with stickers etc. Another impact factor of the co-creation session is the selection of participants. To provide a trusting and creative environment, as well as a broad perspective, the facilitator has to consider the group dynamic needs based on personality, experience, and competence carefully. Doing business on providing innovation services like the co-creation session, also require facilitators with competence and experience to guide, inspire and lead the participants through the session.

Early Validation of User Needs in Concept Development

59

5 Discussion and Conclusion This paper explores criteria for an effective early validation of user needs from an innovation consultancy perspective. Three speciﬁc cases in three different domains are the basis for this study. As part of the insight phase for the co-creation session, the innovation consultancy performs user research by interviewing and interacting with users in their operational context. This is well aligned with the empathize phase in Design Thinking [8] and stakeholder analysis in Systems Engineering [5]. The insight phase also includes research on enabling technology, market trends and competitors, as we ﬁnd in business model theory [12]. During the co-creation session, the innovation consultancy performs exercises for removing self-induced limitations and makes use of different techniques to analyzing the problem and solution domain, such as considering pains and gains. These methods are familiar in both Design Thinking [8, 9], Systems Architecting [7] and business theory [13]. Playful (warm up) exercises are however more common in Design Thinking than in Systems Architecting. By using a key driver graph, we derived criteria and impact factors of the effectiveness of the early validation method applied by the innovation consultancy. The derived criteria are: think outside the box to generate new ideas, build common understanding of user needs, and generate tangible concepts with ownership from the customer. The main impact factors are: research on user needs, technology and market trends, techniques used for analyzing the problem and solution domain, selection of participants, and the competence of the facilitator. We conclude that in these three cases the methods are effective in communicating innovative ideas and concepts with the purpose of early validation of user needs. Acknowledgments. This research is part of a larger research project on Human Systems Engineering Innovation Framework (H-SEIF), funded by the Norwegian government through Oslofjordfondet.

References 1. Yin, R.K.: Case Study Research: Design and Methods. Sage, Los Angeles (2014) 2. Kossiakoff, A., Sweet, W.N., Seymour, S.J., Biemer, S.M.: Systems Engineering: Principles and Practice. Wiley, Hoboken (2011) 3. Wheatcraft, L.: ConOps vs OpsCon – What’s the Difference? https://reqexperts.com/2013/ 06/25/conops-vs-opscon-whats-the-difference/ 4. Solli, H., Muller, G.: Evaluation of illustrative ConOps and decision matrix as tools in concept selection. In: INCOSE International Symposium, pp. 2361–2375 (2016) 5. INCOSE: Systems Engineering Handbook. Wiley, Hoboken (2015) 6. Cockburn, A.: Writing Effective Use Cases. Addison-Wesley, Boston (2005) 7. Muller, G.: Challenges in teaching conceptual modeling for systems architecting. In: Jeusfeld, M., Karlapalem, K. (eds.) Advances in Conceptual Modeling, ER 2015. Lecture Notes in Computer Science, pp. 317–326. Springer, Cham (2015) 8. Plattner, H.: An Introduction to Design Thinking PROCESS GUIDE (2010) 9. Brown, T., Wyatt, J.: Design thinking for social innovation (2010)

60

M. Kjørstad et al.

10. Kelley, T., Kelley, D.: Creative Conﬁdence: Unleashing the Creative Potential Within us All. William Collins, London (2015) 11. Beckers, J.M.J., Muller, G.J., Heemels, W.P.M.H., Bukkems, B.H.M.: Effective industrial modeling for high-tech systems: the example of happy flow. In: INCOSE International Symposium, pp. 1758–1769 (2007) 12. Osterwalder, A., Pigneur, Y.: Business Model Generation. John Wiley & Sons, Hoboken (2010) 13. Osterwalder, A., Pigneur, Y., Bernarda, G., Smith, A., Papadakos, T.: Value Proposition Design: How to Create Products and Services Customers Want. Wiley, Hoboken (2014) 14. Maurya, A.: Running Lean: Iterate from Plan A to a Plan That Works. O’Reilly, Sebastopol (2012)

Introduction of User Experience into the Design of Academic Services at University Centre of Defence Jorge Sierra-Pérez2(&), Carlos Romero-Piqueras2, Myriam Cilla1, Silvia Guillén-Lambea1, and Marcos Pueo1

2

1 Centro Universitario de la Defensa, Ctra. De Huesca s/n, 50090 Zaragoza, Spain {jsierra,mcilla,sguillen,mpueo}@unizar.es Department of Design and Manufacturing Engineering, Zaragoza University, C/Maria de Luna 3, 50018 Zaragoza, Spain [email protected]

Abstract. The introduction of user experience (UX) approach into public services is still a not widespread ﬁeld. This paper presents a real experience in a University institution, re-designing an academic service that provides to its students from an UX approach. Speciﬁcally, the main objective is to improve the efﬁciency in the use of different resources during the provision of the management of the degree Thesis. The initial diagnosis reveals the asymmetry of the workload among different actors, highlighting the administrative ofﬁce. Moreover, there is no any computerized management system, which would speed up the performance of certain tasks, reducing the risk of error. This service’s redesign is based on the introduction of an information management system that simpliﬁes the tasks related to all information and documentation generated during the process. In this way, this system allows the re-distribution of workload giving greater importance and responsibility to each stakeholder where it is necessary. Keywords: Design thinking Academic services

User-centered design Service design

1 Introduction Design Thinking is a general human-centered approach to problem solving, creativity and innovation that puts the observation and discovery of human needs at the forefront of the innovation process [1], based on different principles, practices and tools, thinking styles and mindsets [2]. The design thinking approach can be differentiated into three essential objectives; (1) deep and holistic user understanding; (2) visualization of new possibilities, prototyping and reﬁning; and (3) the creation of a new activity system to bring the nascent idea to reality and proﬁtable operation [3]. In recent years, this type of thinking has been revealed as a highly powerful tool because its capacity is focused on the understanding of the end user, and their needs. In this regard, the introduction of this approach into the provision of services by public or private organizations is considered a ﬁeld with a high potential, incorporating © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 61–67, 2019. https://doi.org/10.1007/978-3-030-02053-8_10

62

J. Sierra-Pérez et al.

multidisciplinary contributions from marketing, human resources, operations, organizational structure, and technology disciplines, integrated through design-based methods and tools [4]. In service design processes, the UX is combined with other service elements such as the physical environment to help designers and customers co-create optimal and memorable experiences for the users creating new links between the different agents of the process [5]. Traditionally, the introduction of UX approach for the design of services has been focused on private organizations, but recently it has notably increased its foray into public entities at the state, regional and local levels [6]. The Government of the United Kingdom redesigned all the online procedures (gov.uk) that its citizens carried out; designing, writing and organizing their requirements according to the needs of citizens, and not following the way of operating the state [7]. U.S. Department of Health and Human Services (HHS) had launched a web site (https://www.usability.gov/) which provides overviews of the user-centered design process and various UX disciplines. This website collects related documentation on methodology and tools to develop digital services under usability terms. At Spanish level, the introduction of UX approach is still limited. There are isolated examples at the local and regional level within the country. On the one hand, the region of Aragón (Spain) has a program to improve the services that it provides to its citizens; starting with the training of its civil servants in service design in the highest levels of the administration. In 2017 the ﬁrst edition was carried out, the “Zaragoza GovJam” (http://zaragozagovjam.org/), where through the collaborative design between different departments, it was conceptualized how some of the services they provide in the news could be redesigned. In 2018, previously to the second “Zaragoza GovJam”, a speciﬁc formation about UX and service design was carried out to public employees. On the other hand, the local government of Sant Quirze del Vallès (Spain) introduced UX in public service design to organise the administrative procedures that it provides to its citizens. The idea arose from the continuous technological revolution in which we are immersed, since it provokes the coexistence of two parallel worlds: that of attention through social networks, fast, close and empathetic, with that of face-to-face attention. Therefore, the new Citizen Attention Ofﬁce of this city council was redesigned following UX criteria. It included the necessary meeting between on-line and off-line, allowing the revision and improvement of citizen attention protocols (http://www. santquirzevalles.cat/). At academic level, Stanford University, through the “d.school”, is one of the focuses on Design Thinking, not only for its internal implementation, but also for being a reference in the training in this discipline. In addition, it is in charge of advising reference companies to implement this methodology in its structure (https://dschool. stanford.edu/). In the Spanish context, the Polytechnic University of Catalonia has since 2011 a collaborative work network of administrative staff called Nexus24 (http:// www.upc.edu/nexus24). The program promotes and facilitates collaborative work in the public administration network and it is aimed at the 1,500 people who perform management work in the University. Its objective is to create a dual operating system, hierarchical and rediarchic mode, putting their students in the centre of the system. Taking all this into account, this study focuses on the evaluation, from a usercentered approach, of different administrative and academic services provided by the

Introduction of User Experience into the Design of Academic Services

63

University Centre of Defence (CUD) of Zaragoza (Spain) to its students. Due to its recent creation and its singular nature, the institution has generated a great amount of academic, informative and administrative services in a short period of time. Therefore, a reordering, identifying possible overlaps or optimising the use of resources could be needed. The main objective of service design is to improve the efﬁciency of the selected services, optimising the use of resources in the different levels of the system: student, administration, faculty, website, printed, etc.

2 Methodology The methodology used is the research-action following the model of Coghlan and Brannick [8]. It is a collaborative approach in which the research team solves a client problem: the redesign of an academic management service, and in turn, scientiﬁc knowledge is generated. In order to achieve the project objectives, the following stages have been implemented (Fig. 1): (1) Context and purpose, where the action was carried out and the purposes of the researcher and the organization were deﬁned. (2) Diagnosis, justiﬁcation of the need to perform the action in that context. (3) Planning of the action, phases and tasks of the action. (4) Launch of the action, the action was taken. (5) Evaluation of the action, the results of the action were analysed.

Fig. 1. Design thinking stages (https://dschool.stanford.edu)

2.1

Context and Purpose: Analysis of the Starting Situation

The present project introduces the Design Thinking methodology through the design of the different administrative processes and/or academic management that take place in CUD, always focused from the students’ perspective. Through the experience of students is redesigned the functioning of some of the current services that are offered to them from the moment they enter as students in the CUD until they are graded. The main objective of the introduction of service design is to improve the efﬁciency in the provision of the selected services, reinforcing the efﬁciency in the use of resources in the different levels of the system: student, administration, faculty, web sites, paper forms, etc. The services that can be redesigned will be academic services provided to students can be either academic or informative services.

64

J. Sierra-Pérez et al.

During the project development, it is necessary to involve both the ﬁnal user, in this case the CUD’s students, and the institution management team, which authorises the present proposal. Moreover, the Administration and Services Staff related to the services evaluated as main actors, as well as part of the teaching staff as support participants in some procedures. 2.2

Initial Diagnosis

The main objective of this phase is to obtain a broad picture of the existing service, mapping all actions of each actor and related interactions. From this, it can be identiﬁed problems or improvement areas to address in the following phase, the action planning. The tools used in this stage are semi-structured individual and group interviews to the different actors involved, contextual observation, immersion techniques where the researcher lives the experience in the ﬁrst person, as well as focus groups. Firstly, the research team develops an initial service mapping from its experience as service’s actor. From this initial mapping, the relevance of each actor is obtained and thus, the priority actions are established. That is, ﬁrst, the most relevant actors are interviewed, to build the service structure as more comprehensive as possible. Then, other less relevant actors complete this mapping with more speciﬁc information. This mapping allows to identify critical points of service, and needs and problems of users. The conclusions of this phase are the basis on which develop the action to carry out in order to obtain the establish objectives. 2.3

Action Planning

Once all the academic services of CUD have been analysed, the service of the management of the degree Thesis (TFG) has been selected as candidate to develop the UX experience process. This service is one of the most complicated processes of the institution, since its duration is one year and involves several actors such as students, professors, administrative staff, coordinators and external staff from the place where the student develop his/her work. The students carry out the degree Thesis during their stay on different military units all around Spain. Therefore, they are not located at the academic center. Before designing the actions that respond to the conclusions obtained in the diagnosis of the service, it is necessary to deﬁne the economic availability, of resources, as well as the disposition of all actors involved in the service. For this purpose, interviews with the different engaged actors and with the centre management team have been performed in order to map the complete academic service. The different proposed actions to carry out have to classify as short, medium or long term. And also, the impact of their application, in order to identify collateral effects in other institution’s services. The set of actions must be accompanied by a planning of its implementation in the prototyping phase, as well as the resources necessary to carry it out.

Fig. 2. Network service architecture of the management of degree thesis

Introduction of User Experience into the Design of Academic Services 65

66

J. Sierra-Pérez et al.

3 Results and Discussion The initial diagnosis of the current situation of the service reveals the asymmetry of the workload among the different actors of the service, with the administrative ofﬁce performing the majority of the procedures, concentrating the majority of actions and also they are involved in the most actions of the rest of actors (Fig. 2). The number of student who carry on the degree Thesis per year is around 150. Thereby the process implies a great amount of bureaucratic paperwork. The majority part of this paperwork is carried out by e-mail, obstructing the communication between the different actors. The informative process to the rest of service’s actors should be more fluent. Moreover, there is no any computerized management system, which would speed up the performance of certain tasks and reduce the risk of human error. The available initial information for students and professors is not clear and its accessibility is not easy. Most of the procedures are included in CUD website, and users should download the forms, ﬁll out and sign, scan, and ﬁnally, send by email. Then, the administrative staff have to check all the ﬁelds one by one in order to assure the correctness of the information. The students complain about the communication of their results (marks) because the process is very slow and the information is not received at the same time for all the students. The redesign of this service is based on the introduction of an information management system that simpliﬁes the tasks related to all information and documentation generated in the process. In this way, this type of system allows the workload to be distributed giving greater importance and responsibility to each stakeholder where it is considered necessary. Moreover, the document manager will hold online intelligent formularies for all service procedures, in which all actors will ﬁll them the most updated version.

4 Conclusions The UX is a powerful design methodology for the academic service of a university institution. This methodology has been applied to the management of the degree Thesis for the students of the CUD. The analysis, which is based on the different points of views of the involved actors, has provided different potential proposals to improve the service. In future academic courses, it is intended to incorporate the new measures, as well as work on improving the rest of the academic services that the CUD provides to its students.

References 1. Brown, T.: Change by Design: How Design Thinking Transforms Organizations and Inspires Innovation, vol. 31 (2009) 2. Carlgren, L., Rauth, I., Elmquist, M.: Framing design thinking: the concept in idea and enactment. Creat. Innov. Manag. 25(1), 38–57 (2016) 3. Lockwood, T.: Design Thinking: Integrating Innovation, Customer Experience, and Brand Value. Culture. p. 256 (2009)

Introduction of User Experience into the Design of Academic Services

67

4. Patrício, L., Gustafsson, A., Fisk, R.: Upframing service design and innovation for research impact. J. Serv. Res., pp. 1–14, December 2017 5. Teixeira, J., Patrício, L., Nunes, N.J., Nóbrega, L., Fisk, R.P., Constantine, L.: Customer experience modeling: from customer experience to service design. J. Serv. Manag. 23(3), 362–376 (2012) 6. Doughty, H.A.: Leading public sector innovation: co-creating for a better society. Innov. J. 15(3), 1–6 (2010) 7. Commission, Design: Restarting Britain2: Design and Public Services. Annu. Rev. Policy Des. 2, 1–10 (2014) 8. Coghlan, D., Brannick, T.: Doing Action Research in Your Own Organization. SAGE Publications, London (2014)

Research on Aesthetics Degree Evaluation Method of Product Form Ming Li(&) and Jie Zhang East China University of Science and Technology, Shanghai 200237, China [email protected]

Abstract. In order to analyze the implicit aesthetic perception of consumers objectively, an evaluation model of aesthetics measure was presented based on the explicit product form. The aesthetics evaluation knowledge was summarized and ten indexes of aesthetics degree to evaluate product form were acquired. The index dimension was reduced with cluster analysis, which would obtain the representative aesthetics degrees for target product. The structural equation model was applied to test the rationality of the index. The mapping relationship of product form to its aesthetic measure was established with neural network, so that the cognitive processes of aesthetic measure could be simulated and evaluation of aesthetic measure conducted. Taking purple clay teapot as an example, the evaluation of its aesthetic measure was veriﬁed and analyzed. The result showed that the model established would be reliable and could provide an effective aid to the design of product form. Keywords: Product form Neural network

Aesthetic degree index Evaluation model

1 Introduction Consumer products of the emotionalization and personalization as the primary selection principle has surpassed function of products. Signiﬁcantly changed in consumption patterns have directly aroused the rethinking of product design by enterprises and designers. But how to ﬁnd and grasp the spiritual needs of consumers, and how to improve the aesthetic sense of product form is a very important topic for designers. On the other hand, how to objectively describe, express and evaluate the aesthetics of products will have a signiﬁcant impact on product design. The deﬁnition and calculation of aesthetics has made various explorations by a large number of scholars. George David Birkhoff proposed a mathematical model of aesthetic degree in 1933, the “Aesthetic measure” was expressed as the ratio of order to complexity, i.e. M = O/C [1]. Bauckhage and Kersting applied the golden rectangle to evaluate images in the process of aesthetic evaluation [2]. Ngo and teo proposed and calculated the formal rules of the screen interface elements [3]. At present, the evaluation of product form aesthetic hasn’t synthesized various aesthetic indicators and formed practical evaluation methods system.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 68–75, 2019. https://doi.org/10.1007/978-3-030-02053-8_11

Research on Aesthetics Degree Evaluation Method of Product Form

69

2 Study Method 2.1

Aesthetic Measures

The formal beauty law is the experience summary and abstract of formal in the process of creating beauty, some of its usage include balance and equilibrium, contrast and blend, change and unity, cadence and rhythm, etc. Studying and exploring the effects of formal beauty law on human aesthetic perception can guide people to better create beautiful things. In this study, ten indexes were proposed for the aesthetic of product morphology. Degree of Balance The degree of balance is computed as the difference between center of gravity of components on each side of the X-axis, Y-axis and Z-axis and is given by WL WR WT WD WF WB maxðjWL j;jWR jÞ þ maxðjWT j;jWD jÞ þ maxðjWF j;jWB jÞ DB ¼ 1 3

ð1Þ

where L, R, T, D, F, and B stand for left, right, top, bottom, face, and back, respectively. Degree of Equilibrium The degree of equilibrium is computed as the difference between weight of components on each side of the horizontal and vertical axis and is given by: DE ¼ 1

JT JD JL JR maxðjJT j;jJD jÞ þ maxðjJL j;jJR jÞ 2

ð2Þ

With: Jj ¼ Sj Dj ; j ¼ T; D; L; R

ð3Þ

where T, D, L, R, and T stand for top, bottom, left, and right, respectively; Sj is the cross-sectional area on side j; Dj is the distance between the central lines of the object on each side and the whole product. Degree of Unity The degree of unity, by deﬁnition, is computed as the relationship between product components and product unity and the compactness of the distribution of product components, and given by

DU ¼

P n ai 1 n2 þ maxðui Þminðui Þ þ i aframe n uproduct 3

ð4Þ

where n is the number of product components; ui and uproduct are the volume of object i and product, respectively; ai and aframe are the areas of object i and cross-section of product.

70

M. Li and J. Zhang

Degree of Coordinate The degree of coordinate is computed as the difference between the physical center of product components and the physical center of product unity and given by: n n P P 2 ai ðXi Xc Þ 2 ai ðYi Yc Þ i i þ n n bframe P ai hframe P ai i i ð5Þ DC ¼ 1 2 where (Xi, Yi) and (Xc, Yc) are the coordinates of object i and product center, respectively; bframe and hframe are the width and height of product, respectively. Degree of Deviation The degree of deviation is computed the deviation of the volume and area of a product from similar products and given by n P DD ¼

Vi Vi þ

1

n P

Si Si

1

ði N Þ

2

ð6Þ

where N is the number of study samples; Vi and Si are the volume and area of object i. Degree of Economy The degree of economy, by deﬁnition, is a measure of how economical the product is and is given by DEY ¼ 1

1 Nsize

þ

1 Nmaterial

þ

1 Nobject

3

ð7Þ

where Nsize, Nmaterial and Nobject are the number of sizes, materials and components of product, respectively. Degree of Homogeneity The degree of homogeneity, by deﬁnition, is a measure of how evenly the objects are distributed among the quadrants and is given by DH ¼ k ln

n! nTL !nTR !nDL !nDR !

ð8Þ

where k is a constant, known as Boltzmann’s constant; n is the number of product components; nTL, nTR, nDL, and nDR are the numbers of objects on the top-left, topright, down-left, and down-right quadrants, respectively. Degree of Symmetry The degree of symmetry, by deﬁnition, is the extent to which the product is symmetrical in three directions: vertical, horizontal, and diagonal and is given by

Research on Aesthetics Degree Evaluation Method of Product Form

DS ¼ 1

jSvertical j þ jShorizontal j þ jSradial j 3

71

ð9Þ

where Svertical, Shorizontal, and Sradial are, respectively, the vertical, horizontal, and radial symmetries with Svertical

0 0 0 0 0 X X 0 Y Y 0 Y Y 0 XTR 1 XTL DL DR TL TR DL DR ¼ 0 ; X 0 Þ þ maxðX 0 ; X 0 Þ þ maxðY 0 ; Y 0 Þ þ maxðY 0 ; Y 0 Þ 8 maxðXTL TR DL DR TL TR DL DR 0 0 0 0 # h h0 h h0 R R0 R R 0 TL TR DL DR TL TR DL DR þ þ þ þ maxðR0TL ; R0TR Þ maxðR0DL ; R0DR Þ max h0TL ; h0TR max h0DL ; h0DR

Shorizontal

Sradial ¼

ð10Þ

0 0 0 0 0 X X 0 Y Y 0 Y Y 0 XDL 1 XTL TR DR TL DL TR DR ¼ 0 ; X 0 Þ þ maxðX 0 ; X 0 Þ þ maxðY 0 ; Y 0 Þ þ maxðY 0 ; Y 0 Þ 8 maxðXTL DL TR DR TL DL TR DR 0 0 0 0 # h h0 h h0 R R0 R R 0 TL DL TR DR TL DL TR DR þ þ þ þ maxðR0TL ; R0DL Þ maxðR0TR ; R0DR Þ max h0TL ; h0DL max h0TR ; h0DR

ð11Þ

0 0 0 0 0 X X 0 Y Y 0 Y Y 0 XDR 1 XTL TR DL TL DR TR DL þ þ þ 0 ; X0 Þ 0 ; X0 Þ 0 ; Y0 Þ 0 ; Y0 Þ maxðXTR maxðYTL maxðYTR 8 maxðXTL DR DL DR DL 0 0 0 # 0 0 h h h h R R 0 R 0 R 0 TL DR TR DL TL DR TR DL þ þ þ þ maxðR0TL ; R0DR Þ maxðR0TR ; R0DL Þ max h0TL ; h0DR max h0TR ; h0DL

ð12Þ

Xj0 ; Yj0 ; h0j ; and R0j are, respectively, the normalised values of Xj ¼ Xj Xc ; j ¼ UL; UR; DL; DR

ð13Þ

Yj ¼ Yj Yc

ð14Þ

Yj Yc hj ¼ Xj Xc

ð15Þ

Rj ¼

qﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2 2ﬃ Xj Xc þ Yj Yc

ð16Þ

where TL, TR, DL and DR stand for top-left, top-right, down-left and down-right, respectively; (xj, yj) and (xc, yc) are the co-ordinates of the centers of product components on quadrant j and the product unity. Degree of Proportion The Degree of proportion, by deﬁnition, is the comparative relationship between the dimensions of the proportional shapes and is given by DP ¼ tj ; min tj t; j ¼ sq; r2; gr; r3; ds

ð17Þ

72

M. Li and J. Zhang

with

t¼

r; r 1

;r 1 r ;r[1

¼

H B

ð18Þ

where H and B are the width and height of the product. pj is the proportion of product j with

1 1 1 1 1 ; ; ; ; psq ; pr2 ; pgr ; pr3 ; pds ¼ ð19Þ 1 1:414 1:618 1:732 2 where sq, r2, gr, r3, and ds stand for square, square root of two, golden rectangle, square root of three, and double square, respectively. Degree of Order Degree of order, by deﬁnition, is the extent to which the objects are systematically ordered and is given by Ds;q ¼ 1 with

ð20Þ

0 0 0 0 0 1 ITL ITL ITL ITR ITR 0 ; I 0 Þ þ maxðI 0 ; I 0 Þ þ maxðI 0 ; I 0 Þ þ maxðI 0 ; I 0 Þ þ maxðI 0 ; I 0 Þ 6 maxðITL UR TL DL TL DR TR DL TR DR

0 IDL þ 0 ; I0 Þ maxðIDL DR

ð21Þ

I ¼ X; Y

ð22Þ

" min J10 ; J20 min J20 ; a03 min J30 ; J40 min J10 ; J30 min J10 ; J40 n! 0 0 þ 0 0 þ 0 0 þ 0 0 þ 0 0 2 ðn 2Þ max J1 ; a2 max a2 ; a3 max J3 ; J4 max J1 ; J3 max J1 ; J4 0 # min J20 ; J40 min Jn1 ; Jn0 þ þ 0 þ max J20 ; J40 max Jn1 ; Jn0

ð23Þ

J ¼ A; V

ð24Þ

RI ¼

RJ ¼

jRX j þ jRY j þ jRA j þ jRV j 4

where An and Vn are the cross-sectional area and volume of product components n on each quadrant; Xj0 ; Yj0 ; A0j ; and Vj0 are respectively, the normalised values of Xj ¼

nj X Xij Xc

ð25Þ

i

Yj ¼

nj X Yij Yc ; j ¼ TL; TR; DL; DR

ð26Þ

i

where TL, TR, DL and DR stand for top-left, top-right, down-left and down-right, respectively.

Research on Aesthetics Degree Evaluation Method of Product Form

73

3 Case Study 3.1

Selection Sample

This study selected teapot as an example, and the purple clay teapot has been collected from masters and workshops in china, it has been selected 20 representative teapots and remodeled by grayscale process. Samples of all teapots are shown in Fig. 1. 31 students who have design background have been interviewed by semantic differential and analyzed result of data.

Fig. 1. Samples of purple clay teapot

3.2

Fig. 2. Model of indexes veriﬁcation

Determination Representative Indexes

The purple clay teapot has been measured through aesthetic index, and degree of deviation, degree of symmetry, degree of proportion and degree of order as the representative indexes are selected by cluster analysis. The result that cluster analysis obtained 4 aesthetic degree indexes has been validated through structural equation model, this model’s v2 and RMSEA value are used as the validation parameters and model is shown in Fig. 2. This model has a good result which v2 and RMSEA value is 39.857 and 0.06 (below 0.08), respectively, and the degree of freedom is 17. The results shown that model of 4 representational indexes is reasonable and reliable. 3.3

Establish BP Neural Network Model of Aesthetic Evaluation

This study used BP neural network toolbox in MATLAB. The input layer parameters are 4 representative degree of aesthetic value, and the output layer parameters are aesthetic feeling by semantic differential in trained process.

74

M. Li and J. Zhang

3.4

Veriﬁcation Analysis

The value of two designed teapots aesthetic evaluation is 0.9612 and 0.6379 through 31 students who have design background, and tested sample 1 aesthetic evaluation higher than tested sample 2 (see Fig. 3). We set up a prototype system for evaluating the aesthetic of teapot, it is shown in Fig. 4. The prediction of aesthetic evaluation is 0.6850 and 0.6414 (see Table 1). The prototype model system prediction value is close to the survey value, and both have the same trend. The result showed that the model established would be reliable and could provide an effective aid to the design of product form.

Fig. 3. Design samples

Fig. 4. Prototype system

Table 1. Testing of aesthetic degree evaluation No. Symmetry Proportion Order Deviation System prediction Aesthetic 1 0.771 0.707 0.404 0.036 0.685 0.691 2 0.827 0.500 0.333 0.361 0.641 0.637

4 Conclusions In this paper, we have presented a prototype system of evaluating teapot aesthetics and introduced 10 aesthetic measures. And we have applied cluster analysis, structural equation model and BP neural network to establish aesthetic degree of evaluation method. Conclusion 1, application summary of aesthetic judging knowledge built a series of evaluation functions based on explicit product forms, so objective evaluation of the implicit beauty can be made; Conclusion 2, the representative aesthetic indexes can be obtained by cluster analysis and structural equation model in speciﬁc product form; Conclusion 3, the BP neural network can be built relationship between explicit product form and implicit aesthetic sense at quantitative level, the system can simulate human aesthetic cognitive process and calculate aesthetic evaluation, so that it could provide an effective aid to the design of product form.

Research on Aesthetics Degree Evaluation Method of Product Form

75

References 1. Birkhoff, G.D.: Aesthetic Measure. Cambridge University Press, Cambrige (1933) 2. Bauckhage, C., Kersting, K.: Can computers learn from the aesthetic wisdom of the crowd? Künstliche Intelligenz 27(1), 25–35 (2013) 3. Ngo, D.C.L., Teo, S.L., Byrne, J.G.: Modelling interface aesthetics. Inf. Sci. 152, 25–46 (2003)

Sensemaking on the Bridge: A Theoretical Approach to Maritime Information Design Brit-Eli Danielsen(&) Department of Design, NTNU Norwegian University of Science and Technology, Kolbjørn Hejes vei 2B, 7491 Trondheim, Norway [email protected]

Abstract. In the maritime sector 75–96% of accidents have been attributed to “human error”. Behind the label “human error” there are humans struggling to make sense of their environment, which is often a complex system comprising people, organizations and technology. Sensemaking is a concept that can help us understand human behavior in organizations. Knowledge about humans’ capabilities and limitations are crucial for designing resilient systems. The majority of sensemaking research has focused on the cognitive and linguistic sphere. However, an emerging topic in the sensemaking literature is embodied sensemaking, which looks into how also intuitive and bodily sensations and emotions are influencing how we interpret and act in the world. Embodied sensemaking may be especially relevant in the maritime sector where the environment is highly dynamic. This article forms a foundation for further research on sensemaking and maritime information design. Keywords: Sensemaking

Maritime information design Human factors

1 Introduction International shipping transports approximately 90% of the world trade [1], so the safety of vessels is important for the environment and human lives as well as for the global economy. Shipping is becoming safer every year as the world fleet has increased while the number of ships totally lost each year have declined [2]. Still there were 85 total shipping losses in 2016 [1]. According to AGCS (2017), 75–96% of marine accidents can be attributed to human error as “a number of incidents have occurred where crews have relied too much on technology, particularly involving electronic navigation tools.” [1]. The Norwegian Maritime Authority registered 219 vessels involved in accidents in 2016, and 66% of these were characterized as navigation based incidents [3]. Dhillon [4] reported that over 80% of marine accidents are caused or influenced by human and organizational factors. There are numerous statistics like these showing that humans go about making mistakes, slips and lapses [5], often when interacting with technology, like navigation tools. This might make it rational to conclude that in order to increase maritime safety it is necessary to remove the human. However, there are no statistics on “human recoveries” which are incidents that have been averted by the actions of humans and there is increasing evidence that humans add safety to systems [6]. Within safety science it is now more common to view “human © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 76–81, 2019. https://doi.org/10.1007/978-3-030-02053-8_12

Sensemaking on the Bridge

77

error” as a symptom, not a cause, a starting point for further investigations into how a complex system comprising people, organizations and technologies both can have successful and unsuccessful outcomes [7]. There is currently an increased focus on human factors in the maritime sector. Human factors can be deﬁned as “the scientiﬁc discipline concerned with the understanding of the interactions among humans and other elements of a system and the profession that applies theory, principles, data and methods to design in order to optimize human well-being and overall system performance” [8]. A vital element is ﬁtting the task, equipment or environment to the human, as opposed to ﬁtting the human to the environment [9]. Hence, it is crucial to understand the capabilities and limitations of humans. Sensemaking is a concept that can help us understand human behavior in organizations [10] and may be important in contributing to an organizations resilience [11]. Sensemaking involves the processes through which people work to understand and create meaning to issues or events [10, 12]. Sensemaking is about “the interplay of action and interpretation rather than the influence of evaluation on choice” [10]. In his analysis of a friendly ﬁre incident where two U.S. Air Force F-15 ﬁghters shot down two U.S. Army helicopters over northern Iraq in 1994 [13] Snook points out that the pilots did not “decide” to pull the trigger: “Why did they decide to shoot?” quickly becomes “Why did they make the wrong decision?” Hence, the attribution falls squarely onto the shoulders of the decision maker and away from potent situation factors that influence action. Framing the individual-level puzzle as a question of meaning rather than deciding shifts the emphasis away from individual decision makers toward a point somewhere “out there” where context and individual action overlap. (…) underscores the importance of initially framing such senseless tragedies as “good people struggling to make sense,” rather than as “bad ones making poor decisions” (pp. 206–207). Snook emphasizes that the individual sensemaking occurs as an interplay with the environment or context. In the maritime context, safety-critical information is increasingly being presented to the bridge crew via Information and Communication Technology (ICT). The safety aspect of the human-machine interface (HMI) are thus of utter importance as “we acquire more information through vision than through all of the other senses combined” [14]. However, many factors in addition to visual perception influence how a seafarer make sense of his environment: individual factors (human senses, fatigue, stress), communication and teamwork, work environment, and cultural aspects [9]. The following section will shed some light on the concept of sensemaking.

2 Sensemaking Karl E. Weick made sensemaking prominent in the organization literature in 1995, with his seminal book Sensemaking in Organizations [10]. Sensemaking has since been the subject of considerable research, not only in organizational studies, but also within education, health care, communication, leadership and management studies. We are all continuously and effortlessly making sense of our surroundings and everyday life. This mundane sensemaking process may be difﬁcult to study. Hence, most sensemaking studies have focused on speciﬁc episodes, where an interruption of

78

B.-E. Danielsen

an ongoing activity sets off a deliberate or active form of sensemaking [15]. Events, issues or situations become triggers for sensemaking when they “interrupt people’s ongoing flow, disrupting their understanding of the world and creating uncertainty about how to act” [12]. The discrepancy between expectations and reality is important, whether being an occurrence of an unexpected event or a non-occurrence of an expected event that creates ambiguity or uncertainty [10, 12]. There are many factors that influence whether the violated expectations or cues trigger sensemaking, like individual or organizational identity, cognitive frames, personal or strategic goals and technology [10, 12, 15]. Organizational crisis are particular powerful occasions for sensemaking as they are often characterized by ambiguity of cause, effect and means of resolution [16]. Weick describe crises as “low probability/high consequence events that threaten the most fundamental goals of an organization” [17]. Sensemaking consists of the interrelated processes: creation, interpretation and enactment. People ﬁrst create what they focus on (by extracting cues), then interpret these cues and act on those interpretations [15]. This is an ongoing cycle. The outcome of the sensemaking process is a “more ordered environment from which further cues can be drawn” [12] and/or restored organizational activity. The outcome may also be non-sense, where activity is not restored, and that can have fatal consequences [15]. Research on sensemaking as it unfolds during crisis spans a range of contexts: mining disasters [18], climbing disasters [19], space shuttle accidents [20], maritime accidents [21], as well as Weick’s work on the Bhopal accident [17, 22], the Tenerife air crash [23], the Mann Gulch ﬁre [24] and the medical disasters of Bristol Royal Inﬁrmary [25]. Emotions in crisis are often strong and negative like anxiety, fear, panic and desperation. The arousal the emotion trigger in the autonomic nervous system can consume cognitive information processing capacity, which in turn reduces the number of cues that can be noticed and become triggers for sensemaking [16]. As seen in the Mann Gulch ﬁre disaster, people put under life-threatening pressure return to habituated ways of responding, like flight [24]. Positive emotions may broaden individuals’ scope of attention [16] which should lead to a sensemaking process that can contribute to averting crisis and accidents. However, overly positive emotions may cause individuals to be overly optimistic and to overlook important cues and thereby misinterpret the situation [16]. Emotions involve changes in bodily states and “If we take seriously the idea of sensemaking as an emotional process, then we must also understand it as an embodied one” [16].

3 Embodied Sensemaking The research on sensemaking efforts has mainly been limited to the cognitive or linguistic sphere [15]. However, over the recent years focus on embodied sensemaking has emerged [12]. This research is connected to cognitive science and the related embodied cognition, where cognition is seen as partly grounded in bodily states. The cognition, body, and context are viewed as three interrelated concepts that are in constant interactions with each other [26]. Niedenthal [27] claimed “there is a reciprocal relationship between the bodily expression of emotion and the way in which emotional information is attended to and interpreted” [27]. This is supported by

Sensemaking on the Bridge

79

experimental ﬁndings e.g. when receiving good news in a slumped posture experimental participant felt less proud and reported being in a worse mood that participants in upright posture [27]. Cunliffe and Coupland [28] argued that “embodiment is an integral part of sensemaking” and that we make sense of our lives and ourselves through embodied or bodily interpretations in our ongoing everyday interactions and experiences. They theorized the process as “embodied narrative sensemaking”. Through an analysis of a documentary from a rugby tour they demonstrated how the players made sense of their surroundings and experiences in sensory as well as intellectual ways. The authors describes that the players’ sensemaking “draws on an intuitive and informed feeling in his body”. Our bodies cannot be separated from the context, in addition to the cognitive sphere “organizing also operates on a sensory level through sensory knowing and bodily sensations” [28]. Roberts [29] have recognized the value of studying embodied sensemaking in the maritime sector and have proposed a research project on how contemporary seafaring leaders engage in embodied sensemaking to resolve critical events [29]. 3.1

Ship Sense

In the maritime sector, embodied sensemaking may be congruent with the term ship sense. According to Prison et al. [31] ship sense is a form of tacit knowledge regarding the maneuvering of a ship that includes “knowing what information to look for, where to ﬁnd it and how to use it to be able to manoeuvre the ship” [31]. Studies on ship sense has identiﬁed experience and training, perceptual and visual abilities as well as proprioceptive cues as important factors for its achievement and it is presumed to play an important role in the dynamic interaction between the ship and the shiphandler [31]. Prison et al. [31] describe the shiphandlers work to accomplishing a safe journey as “striving for harmony” between the ship and its surrounding environment. A sea voyage have different levels of complexity and required effort from the crew depending on which phase it is in: departing/berthing in harbor, sailing in the archipelago or sailing in open sea. Different ships have different maneuverability and different navigation instruments available. The shiphandler must account for the dynamic factors such as wind, waves, current and visibility that affect each other and the ship. The ability to deal with these factors depends on the shiphandlers personal prerequisites, his education and experience, as well as spatial awareness [30]. The spatial awareness describes the shiphandlers ability to predict the future ship position in relation to waves and objects. When sailing in open sea in fair weather the navigation tasks will mainly be monitoring the navigational instruments. However, if the weather changes to strong winds and high sea-state the effort required by the shiphandler may change dramatically. The autopilot may be disengaged in order to steer the ship manually. In these situations it is important to ‘get a feel for’ the ship’s movement. The shiphandler listens and feels the slamming when the bow falls into the sea after being raised high by the last passing wave. Ship sense is needed to know when it is time to slow down or to slightly alter course in relation to the direction of the oncoming waves, thus it is vital for the safety at sea [30].

80

B.-E. Danielsen

4 Sensemaking in Future Maritime Information Design Currently there is considerable research activity regarding the development of autonomous ships with varying degree of manning level, ranging from manned bridge with speciﬁc functions automated to completely unmanned ships that may be monitored from on shore control centres [2]. There are several drivers for developing unmanned ships, like potential cost reductions and reduction of greenhouse gas emissions. The most widely used argument is the potential for increased safety by removing the human [32]. It has been recognized that human sensemaking will need to be considered even if humans are restricted to be on shore [2, 32]. Humans will design, construct, install, test, verify and perform maintenance within the autonomous system. When an autonomous ship is sailing, the operator in an onshore control centre will be the human in the sharp end. In the MUNIN-project a focus group interview with bridge ofﬁcers resulted in a list of 165 pieces of information that was considered necessary for providing situation awareness for a remote operator [32]. In addition to the standard information usually found on the ship bridge, the operator on shore would need information from sensor systems regarding observation of other ships or objects, safety and emergency information (water ingress and ﬁreﬁghting), anchor handling, cargo and stability, security information, engine parameters, voice communication with other vessels, to mention some. The bridge ofﬁcers also requested information about the ships motions. This may reflect the information the seafarers need in order to maintain their ship sense. Sensors will need to detect and convey motions like heave, roll and slamming to an onshore operator. Porathe et al. (2014) suggested this could be solved with some kind of a ships equivalent to an airplane gyroscope [32]. The information design to assist the operators’ sensemaking in an onshore control centre will be a challenge. The knowledge that onboard is achieved by bodily sensing will probably need to be acquired from displays, at the risk of exceeding mental workload capacity of the operator. However, not all organizations share the belief that automation is the future direction of the maritime sector. As long as there is no clear strategic or ﬁnancial advantage the ship owners may not be pushing for autonomy [6]. Maritime accidents and incidents involving the cooperation between human and technology is occurring in the current reality, where the bridge is manned and have autonomous functions. Further research on embodied sensemaking and ship sense will give valuable insights to how seafarers make sense of their environment and might contribute to bridge design that facilitate sensemaking.

References 1. AGCS, Safety and Shipping Review 2017. An annual review of trends and developments in shipping losses and safety (2017). http://www.agcs.allianz.com/ 2. Porathe, T., et al.: At least as safe as manned shipping? Autonomous shipping, safety and “human error”. European Safety and Reliability Conference. Taylor & Francis Group (2018) 3. Sjøfartsdirektoratet (Norwegian Maritime Authority). Fokus på riskiko 2018 (2018). www. sdir.no 4. Dhillon, B.S.: Human Reliability and Error in Transportation Systems. Springer Science & Business Media, London (2007)

Sensemaking on the Bridge 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.

81

Reason, J.: Human Error. Cambridge University Press, Cambridge (1990) Oltedal, H.A., Lützhöft, M.: Managing Maritime Safety. Routledge, London (2018) Woods, D.D., et al.: Behind Human Error, 2nd edn. Ashgate Publising Ltd., UK (2010) Salvendy, G.: Handbook of Human Factors and Ergonomics, 4th edn. John Wiley & Sons Inc., New Jersey (2012) Grech, M., et al.: Human Factors in the Maritime Domain. CRC Press, Boca Raton (2008) Weick, K.E.: Sensemaking in Organizations. Sage, Thousand Oaks (1995) Kilskar, S.S. et al.: Sensemaking and resilience in safety-critical situations: a literature review. In: European Safety and Reliability Conference. Taylor & Francis Group (2018) Maitlis, S., Christianson, M.: Sensemaking in organizations: taking stock and moving forward. Acad. Manag. Ann. 8(1), 57–125 (2014) Snook, S.A.: Friendly ﬁre: The accidental shootdown of U.S. Black Hawks over Northern Iraq, pp. 1–257. Princeton University Press (2000) Ware, C.: Information Visualization, 3rd edn., pp. 1–30. Morgan Kaufmann, Boston (2013) Sandberg, J., Tsoukas, H.: Making sense of the sensemaking perspective: its constituents, limitations, and opportunities for further development. J. Organ. Behav. 36, S6–S32 (2015) Maitlis, S., Sonenshein, S.: Sensemaking in crisis and change: inspiration and insights from Weick (1988). J. Manag. Stud. 47(3), 551–580 (2010) Weick, K.E.: Enacted sensemaking in crisis situations. J. Manag. Stud. 25(4), 305–317 (1988) Wicks, D.: Institutionalized mindsets of invulnerability: differentiated institutional ﬁelds and the antecedents of organizational crisis. Organ. Stud. 22(4), 659–692 (2001) Kayes, D.C.: The 1996 Mount Everest climbing disaster: the breakdown of learning in teams. Hum. Relat. 57(10), 1263–1284 (2004) Dunbar, R.L.M., Garud, R.: Distributed knowledge and indeterminate meaning: the case of the Columbia shuttle flight. Organ. Stud. 30(4), 397–421 (2009) Vendelø, M.T.: Disasters in the sensemaking perspective: the Præstø Fjord accident. In: Disaster Research, pp. 190–202. Routledge (2015) Weick, K.E.: Reflections on enacted sensemaking in the Bhopal disaster. J. Manag. Stud. 47(3), 537–550 (2010) Weick, K.E.: The vulnerable system: an analysis of the tenerife air disaster. J. Manag. 16(3), 571–593 (1990) Weick, K.E.: The collapse of sensemaking in organizations: the mann gulch disaster. Administrative Science Quarterly pp. 628–652 (1993) Weick, K.E.,Sutcliffe,K.M.: Hospitals as cultures of entrapment: a reanalysis of the Bristol Royal Inﬁrmary. Calif. Manag. Rev. 45(2), 73–84 (2003) Fahim, M., Rezanejad, A.: An introduction to embodied cognition. Int. J. Lang. Linguist. 2(4), 283–289 (2014) Niedenthal, P.M.: Embodying emotion. Science 316(5827), 1002–1005 (2007) Cunliffe, A., Coupland, C.: From hero to villain to hero: making experience sensible through embodied narrative sensemaking. Hum. Relat. 65(1), 63–88 (2012) Roberts, B.: Recasting Odysseus: embodied sensemaking among seafaring leaders. Aust. J. Maritime Ocean Affairs 10(1), 19–34 (2018) Prison, J., Dahlman, J., Lundh, M.: Ship sense-striving for harmony in ship manoeuvring. WMU J. Maritime Affairs 12(1), 115–127 (2013) Prison, J., Lützhöft, M., Porathe, T.: Ship sense - What is it and how does one get it? RINA, Human Factors in Ship Design. Safety and Operation, London (2009) Porathe, T., Prison, J., Man, Y.: Situation awareness in remote control centres for unmanned ships. Human Factors in Ship Design & Operation, 26–27 February, London, UK (2014)

Examining Cloud Computing Applications from the Perspectives of Privacy and Uniﬁed Theory of Acceptance and Use of Technology Tihomir Orehovački1(&), Darko Etinger1, and Snježana Babić2 1

Faculty of Informatics, Juraj Dobrila University of Pula, Zagrebačka 30, 52100 Pula, Croatia {tihomir.orehovacki,darko.etinger}@unipu.hr 2 Polytechnic of Rijeka, Trpimirova 2/V, 51 000 Rijeka, Croatia [email protected]

Abstract. Recent advances in the ﬁeld of e-learning are focused on constructivist teaching methods that encourage student-centered educational process in which teacher plays the role of a facilitator and mediator. The employment of cloud computing applications enables students to, regardless of space and time, collaborate, learn from each other, exchange ideas and different digital resources, and build knowledge within set educational goals. One of the key prerequisites for successful integration of novel technologies in higher education processes is their adoption by students. This paper examines an interplay among UTAUT model constituents and following facets of privacy: concerns, risks, and control. The validity and reliability of the proposed research framework and associated hypotheses were examined by means of the partial least squares (PLS) structural equation modelling (SEM) technique. Implications for both researchers and practitioners are presented and discussed. Keywords: Cloud computing applications Educational settings Google Drive Uniﬁed Theory of Acceptance and Use of Technology Privacy PLS-SEM

1 Introduction Constructivism is one of the main learning models in e-learning pedagogy and suggests that learning is an active process, that knowledge can be socially constructed, and that the interpretation of knowledge is dependent on the prior knowledge [1]. Cloud computing technologies (such as Google Apps, Microsoft Ofﬁce 365, Zoho Docs, DropBox, iCloud, etc.) have the educational potential predicated on constructivism and cooperative learning [2]. The use of cloud applications in the learning process enables students to, regardless of space and time, retrieve and share information, ideas and different digital resources, build new knowledge, facilitate group dialogue, solve a problem, collaborate, simultaneously respond through a shared ﬁle in the cloud, and promote open communication within set educational goals [2]. Based on the results of a qualitative study, Yadegaridehkordi et al. [3] have found that the signiﬁcant characteristics of cloud-based collaborative learning applications are easy monitoring and © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 82–87, 2019. https://doi.org/10.1007/978-3-030-02053-8_13

Examining Cloud Computing Applications

83

assessment, time control and saving, cost saving, accessibility, ease of use, and easy connection to other applications. While cloud computing applications present a great opportunity for educational institutions, their use raises concern about a variety of privacy threats, as a result of placing a large amount of students’, teachers’ and institutional data into the hands of third party service providers. Namely, cloud computing applications, as multi-domain environments, employ various mechanisms, interfaces, and semantics for each domain which can use different security, privacy, and trust requirements [4]. The adoption of cloud computing by students is one of the key prerequisites for the successful integration of cloud technologies in the teaching-learning process in higher education institutions. The literature indicates that the most frequent factors influencing users’ adoption of cloud based learning system in higher education are: security, trust, usefulness, ease of use, beneﬁts, performance expectancy, social influence, effort expectancy, and facilitating conditions [5]. In that respect, Kayali et al. [6] have recommended the extension of the Uniﬁed Theory of Acceptance and Use of Technology (UTAUT) model [7] for the purpose of examining the adoption of cloud based elearning. As a follow up, this paper explores an interplay among UTAUT model constituents and the following facets of privacy: concerns, risks, and control. The remainder of the paper is structured as follows. Theoretical foundation of our work is briefly described in the following section. Employed research model and hypotheses are introduced in the third section. Findings are reported in the fourth section. Concluding remarks are provided in the last section.

2 Background to the Research As an advantageous model for assessing the success of introducing new technologies in the ﬁeld, the UTAUT model [7] helps us understand the acceptance of predictors with the aim of proactively designing interventions targeting population of users who may be less inclined to adopt and use new technologies. Based on UTAUT model, Hashim and Hassan [8] have found that performance expectancy is the most important factor influencing behavioral intention with respect to cloud computing, followed by effort expectancy, social influence, security, and trust. Furthermore, authors have determined that behavioral intention and facilitating conditions have a strong impact on the use of cloud computing services. Empirical results obtained by Li and Chang [9] have shown that user’s attitude towards cloud computing, subjective norm, perceived behavioral control, and perceived usefulness directly affect their behavioral intentions while perceived ease of use, transferability of computer skills, vendor reputation, perceived risk, privacy concerns, security questions, and concerns about vendor lock-in have indirect impacts on the adoption of cloud computing. By extending TAM model, Arpaci [10] has shown that perceived usefulness, subjective norm, and trust have a signiﬁcant positive effect on the attitude towards the use of mobile cloud storage services, which is a signiﬁcant predictor of behavioral intention in the same respect. According to Arpaci et al. [11], security and privacy have a strong impact on student attitudes related to the use of cloud services in educational environments, which has a signiﬁcant impact on the behavioral intention of using educational services in the cloud. Alotaibi [12]

84

T. Orehovački et al.

discovered that perceived risk has a signiﬁcant negative effect on perceived ease of use and perceived usefulness, but that it does not affects behavioral intention. Drawing on the Theory of Reasoned Action (TRA), the Privacy-Trust-Intention model and costbeneﬁt theories, Gashami et al. [13] uncovered that users’ intentions to use Softwareas-a-Service (SaaS) are negatively affected by privacy concerns, positively influenced by overall perceived beneﬁts, but that there is no direct signiﬁcant effect of privacy concerns on their intentions to use SaaS. In the context of professional cloud environments, Lang et al. [14] have found that perceived privacy is positively related to intention to expand cloud service thus explaining 23% of its variance. Given that to our knowledge no existing empirical research addresses facets of privacy with respect to the adoption of cloud computing applications, we initiated a study in that respect.

3 Research Model and Hypotheses The research model is composed of seven constructs aimed for explaining users’ behavioral intentions with respect to cloud computing applications used in educational settings. Behavioral intention measures the degree to which users are willing to continue to use cloud computing application and recommend it to others. Performance expectancy refers to the extent to which the employment of cloud computing application enhances users’ performance in managing artefacts. Effort expectancy denotes the degree to which is easy for users to become proﬁcient in interaction with cloud computing application. Social influence measures the extent to which majority of people that are important to the user think that he/she should employ cloud computing application. Privacy risk refers to the degree to which users believe is risky to provide cloud computing application with their personal data. Privacy concerns denotes the extent to which users are concerned about the privacy of their data and artefacts stored on cloud computing application. Privacy control measures the degree to which users think they have control over who has access to and is using their personal data. In that respect, following six hypotheses that constitute the proposed research model were deﬁned: H1. H2. H3. H4. H5. H6.

Performance expectancy positively influences Behavioral intention. Effort expectancy positively influences Behavioral intention. Social influence positively influences Behavioral intention. Privacy risk positively influences Privacy concerns. Privacy control negatively influences Privacy concerns. Privacy concerns negatively influences Behavioral intention.

4 Findings Details on participant demographics and employed research design can be, together with results of examining the psychometric features of attributes meant for measuring aspects of perceived security and privacy, found in [15] and [16].

Examining Cloud Computing Applications

85

The research model was examined by means of the Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 3.2.7 [17] software. The analysis was composed of two-step approach: the assessment of the measurement model which was followed by the assessment of the structural model. The assessment of reflective outer models involved determining indicator reliability (squared standardized outer loadings), internal consistency reliability (composite reliability), convergent validity (average variance extracted - AVE), and discriminant validity (Fornell-Larcker criterion, cross-loadings, and heterotrait-monotrait - HTMT ratio of correlations) [18]. To establish the reliability of the items and the convergent validity of the proposed constructs, a conﬁrmatory factor analysis was employed. As the item loadings for each construct were above 0.708 [18], meaning that the average variance extracted is higher than 0.5, we can conclude that the convergent validity is achieved. Six items that have not met the aforementioned requirements were removed from the measurement model. Internal consistency was explored by means of two indicators: Cronbach’s alpha (CA) and composite reliability (CR). Since values of CA and CR were in range from 0.851 to 0.928 and from 0.910 to 0.949, respectively, requirement related to internal consistency of scales was met. Discriminant validity was tested by comparing each construct’s average variance extracted (AVE) with its squared correlations with other constructs in the model [18]. The results of these comparisons revealed that each latent variable accounts for more variance in its associated indicator variables than it shares with other constructs in the same model. To further test the discriminant validity of the constructs, the heterotrait-monotait (HTMT) ratio of correlations was employed [18]. As the conservative threshold of 0.85 was not met, we can conclude that the discriminant validity of the proposed seven constructs is achieved. Slight caution should be taken in regard to the relationship between privacy risk and privacy concerns, even though the HTMT ratio of correlations indicates that there are no severe problems detected using this method. After establishing the reliability for the indicators and the convergent and discriminant validity of the constructs, we examined the structural model. The results of the PLS analysis for the proposed hypotheses are shown in Fig. 1. The structural model shows a signiﬁcant relationship between all constructs. Therefore, all hypotheses are supported. By analyzing the direct, indirect and total effects, it is evident that the strongest relationship is the one between privacy risk and privacy concerns (b = 0.668, p < 0.001), followed by the relationship between performance expectancy and behavioral intention (b = 0.472, p < 0.001) and the relationship between social influence and behavioral intention (b = 0.200, p < 0.001). Weak signiﬁcant relationships are detected as follows: between effort expectancy and behavioral intention (b = 0.169, p < 0.01), between privacy control and privacy concerns (b = −0.163, p < 0.01), and between privacy concerns and behavioral intention (b = −0.126, p < 0.01). The model explains 56.2% of the variance in privacy concerns, and 50.5% of the variance in behavioral intention. The effect size f2 coefﬁcients were calculated for the relationships between constructs. The results show strong effect size coefﬁcients between privacy risk and privacy concerns (f2 = 0.846), and moderate effect size between performance expectancy and behavioral intention (f2 = 0.276). The all other effect size coefﬁcients can be interpreted as weak, according to the guidelines provided by [18].

86

T. Orehovački et al.

Fig. 1. Research model and PLS results

5 Conclusion The objective of this paper was to examine an interplay between constructs that constitute UTAUT model and constructs that represent various facets of privacy with respect to cloud computing applications. For that purpose, an empirical study was carried out during which students completed scenario-based interaction with Google Drive that served as a representative sample of cloud computing applications and afterwards ﬁlled the post-use questionnaire. We in particular found that, in the context of cloud computing applications, performance expectancy, effort expectancy, social influence, and privacy concerns signiﬁcantly affect behavioral intention whereas privacy risk and privacy control signiﬁcantly contribute to privacy concerns. Reported ﬁndings can be used by researchers as a background for future studies in the ﬁeld as well as by practitioners as guidelines they should consider when designing and evaluating cloud computing applications.

References 1. Hung, D.: Theories of learning and computer-mediated instructional technologies. Educ. Media Int. 38(4), 281–287 (2001) 2. Denton, D.W.: Enhancing instruction through constructivism, cooperative learning, and cloud computing. TechTrends 56(4), 34–41 (2012) 3. Yadegaridehkordi, E., Iahad, N.A., Ahmad, N.: User perceptions of the technology characteristics in a cloud-based collaborative learning environment: a qualitative study. Int. J. Technol. Enhanc. Learn. 7(1), 75–90 (2015) 4. Takabi, H., Joshi, J.B., Ahn, G.J.: Security and privacy challenges in cloud computing environments. IEEE Secur. Priv. 8(6), 24–31 (2010) 5. Gambo, Y.: Review on factors influencing user acceptance of cloud-based learning system in higher education. Int. J. Comput. Appl. 7(3) (2017). https://rspublication.com/ijca/2017/ june17/6.pdf

Examining Cloud Computing Applications

87

6. Kayali, M.H., Saﬁe, N., Mukhtar, M.: Adoption of cloud based E-learning: a systematic literature review of adoption factors and theories. J. Eng. Appl. Sci. 11(8), 1839–1845 (2016) 7. Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D.: User acceptance of information technology: toward a uniﬁed view. MIS Q. 27(3), 425–478 (2003) 8. Hashim, H.S., Hassan, Z.B.: Factors that influence the users’ adoption of cloud computing services at Iraqi Universities: an empirical study. Aust. J. Basic Appl. Sci. 9(27), 379–390 (2015) 9. Li, Y., Chang, K. C.: Antecedents to individual adoption of cloud computing. In: Web-Based Services: Concepts, Methodologies, Tools, and Applications, pp. 1088–1110. IGI Global (2016) 10. Arpaci, I.: Understanding and predicting students’ intention to use mobile cloud storage services. Comput. Hum. Behav. 58, 150–157 (2016) 11. Arpaci, I., Kilicer, K., Bardakci, S.: Effects of security and privacy concerns on educational use of cloud services. Comput. Hum. Behav. 45, 93–98 (2015) 12. Alotaibi, M.B.: Exploring users’ attitudes and intentions toward the adoption of cloud computing in Saudi Arabia: an empirical investigation. J. Comput. Sci. 10(11), 2315–2329 (2014) 13. Gashami, J.P.G., Chang, Y., Rho, J.J., Park, M.C.: Privacy concerns and beneﬁts in SaaS adoption by individual users: a trade-off approach. Inf. Dev. 32(4), 837–852 (2016) 14. Lang, M., Wiesche, M., Krcmar, H.: Perceived Control and Privacy in a Professional Cloud Environment. In: Proceedings of the 51st Hawaii International Confer (HICSS), pp. 3668– 3677, Big Island, Hawaii (2018) 15. Orehovački, T., Etinger, D., Babić, S.: Perceived security and privacy of cloud computing applications used in educational ecosystem. In: Proceedings of the 40th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO 2017), pp. 717–722. IEEE, Opatija (2017) 16. Orehovački, T., Babić, S., Etinger, D.: Identifying relevance of security, privacy, trust, and adoption dimensions concerning cloud computing applications employed in educational settings. In International Conference on Applied Human Factors and Ergonomics, pp. 308– 320, Springer, Cham (2017) 17. Ringle, C.M., Wende, S., Becker, J.M.: SmartPLS 3, Boenningstedt: SmartPLS GmbH (2015). http://www.smartpls.com 18. Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M.: A primer on partial least squares structural equation modeling (PLS-SEM). Sage Publications Inc., CA, Los Angeles (2017)

Rob’Autism Project: Being Active in Social Interactions: The Robot-Extension Paradigm Rénald Gaboriau1(&), Sophie Sakka1, Didier Acier2, and Dimitri Delacroix3 1

Ecole Centrale Nantes, LS2N, 1 rue de la Noë, 44300 Nantes, France {renald.gaboriau,sophie.sakka}@ls2n.fr 2 UFR Psychologie, Université de Nantes, Chemin de la Censive du Tertre, 44312 Nantes, France [email protected] 3 UFR Sciences du Langage, Université de Limoges, 39 rue Camille Guérin, 87036 Limoges, France [email protected]

Abstract. Rob’Autism project is a new robotic approach based on the robotic mediation to enhance social interactions of teenagers with Autistic Spectrum Disorder (ASD). The program aims at giving every participant (and according to the characteristics of each) the means to be active and become a social contributor. Rob’Autism introduces the robot-extension paradigm: the robot is used as a prosthesis in communication which makes the participants of the program socially active, as they program the robot to act on the world (non-self). This paper focuses on the social interaction work performed during the program and excerpts from a pragmatic analysis of the interactions between the participants. Keywords: Robot extension Autistic Spectrum Disorder Active participation Pragmatic analysis Rob’Autism

1 Introduction The use of robots as a therapeutic aid for people with Autism Spectrum Disorder (ASD) is a topic which tends to develop [1, 2], especially since the release of various robotic platforms that can now be bought. Indeed, using robots in ASD support is a promising method to reduce anxiety and promote the development of certain skills that are necessary for social interactions: joint attention [3, 4], imitation [5, 6], emotion recognition [7, 8], etc. In all existing approaches, the companion robot paradigm is used: the robot is programmed to present some pre-established behavior or some social scenario. Studies generally show positives effects, but in some cases, the changes are not always signiﬁcant, and the generalization of the acquired skills still remains a problem [9, 10]. Moreover, there is a strong skepticism among professionals who doubt the real contribution of these approaches [11]. So far, most studies focused more on children’s deﬁcits and less on their skills even if the researchers have a real consideration of the subjects’ characteristics to be able to help them better. But in the robot companion approach, the subjects can only conform © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 88–94, 2019. https://doi.org/10.1007/978-3-030-02053-8_14

Rob’Autism Project: Being Active in Social Interactions

89

to what is proposed to them: they do not have the possibilities to rearrange it; in other words, their behavior is passive. They are simple users. As a consequence, this way of using robot does not really offer them a real area of creation or expression. This article presents an alternative approach to the companion robot: the extension robot. The user is treated as an active participant who builds his relationship to the world and to the others based on his own characteristics. The idea is to have a group of 6 teenagers with ASD aged 11–16 learn how to program a robot NAO (SoftBank Robotics) and use it as an extension to interact with others. The ﬁrst observations show that this approach favors the tendency to go towards the others and interact with them. A real collaboration with reciprocal contributions can appear and a faculty to change of perspective (theory of mind) is developing. To illustrate this process, this paper presents excerpts from the pragmatic analysis of social interactions between teenagers who participated in this program.

2 Methodology 2.1

Participants

The Rob’Autism project exists since 2014; every year one group of six teenagers follows the program. Therefore, 24 subjects divided into 4 groups of 6 teenagers diagnosed with ASD, aged to 11–16, participated in the experiment (21 boys, 3 girls). There is no selection on the speciﬁc type of autism: many interindividual differences exist between the participants. They speak but their level of speech is very different. They all have some abilities for reading and writing, sometimes very light. Some have echoes or conversational skills difﬁculties. Their behavior can be swings, selfaggression, violence, resistant to changes (all of them), a little bit autonomous in transports, mutism, and so on. Some are schooled (special classes or classical classes), others are not. All are familiar with computers, with different levels. 2.2

Organization

Every year, 6 young people are welcomed to attend 20 one-hour sessions, which alternate 10 robot programming and 10 non-robotic sessions. The sessions always take place in the same rooms, at the same times, with the same people. For robot programming, 3 robots are used with 2 teenagers per robot, using the same computer. The working tables are in the center of the room in such disposition that each group can see the two other ones. No negative judgment is made on the participants’ productions. During the sessions, a show is prepared to be presented publicly at the end of the program [12, 13].

90

R. Gaboriau et al.

3 Results Videos of the sessions are analyzed by the program supervisor and a neutral person. A pragmatic analysis of social interactions is carried out, which aims at examining the interactions without being deluded by what is immediately observed. Interaction is the result of an implicit construction [14]. We present thereafter short extracts of this analysis. They illustrate what happens in all the groups. The ﬁrst excerpt takes place during the ﬁrst robotic session, which is the second session of the program [15]. It is the ﬁrst time they use the robot. The second extract was chosen because new behaviors had appeared. In the following examples, the participants are 6 boys called them Pn (n = 1…6). Context 1. This is the 2d session [15]. Teenagers have formed their binomial group. The robot specialist has just introduced the robot and its use. The ﬁrst exercise is to make the robot speak. Binomial group 1 (P3 and P6). P3 makes the robot say: “shit”. Then P6 makes the robot say: “Hello, I’m going to kill you”. Every time, everyone laughs in the room, except P2 and P4. P3 and P6 laugh without looking at each other. They look at the computer screen. Using the robot, P6 asks the supervisor: “Hey, nux nux, how can we put some music?” The supervisor replies that they have to install a ﬁle for that. So P3 and P6 continue their tests. They write mainly phrases from movies. Binomial group 2 (P1 and P2). They never look at each other. They look at the screen or at the robot. P2 makes the robot say “shit” too and laugh. Then he writes a sentence to P6: “Do you want to play with me?” But P6 does not answer. P2, after pressing the button several times to turn on the robot, takes the keyboard and makes the robot count from 1 to 10. Binomial group 3 (P4 and P5). They never look at each other. P4 stays away from the computer. The robot of binomial group 1 has just stated: “Hello, I’m going to kill you” So P5 makes the robot say the same thing. He gets up, jumps, looks at the ceiling. When P5 talks to the supervisor, P4 makes the robot say: “I can speak” Subsequently, P5 takes ideas from others, and P4 tells the robot that he needs to update because he has old operating systems. Context 2. This is the 16th session [15]. The binomial groups have changed at each session. During this session, each binomial group record robot movements matching the recorded voices. Everyone showed his productions in turn and applauded those of others. We are in the 52nd minute of the session. Binomial group 1 (P1 and P2, together again). P1 gives the keyboard to P2 and says to him: “You can count now” When P2 writes a number (“40”), P1 adds disturbing elements (“15”) and laughs. Then he makes the robot say: “P5, do you want to watch the cartoon NAO robot?” P5 answers that he does not know this cartoon. Binomial group 2 (P4 and P6). P6 wants his robot to say to P2: “Do you want to count up to 1000?”, but P4 adds a disturbing element and the robot say: “Do you want to count up to 1000 asses?” Everybody laughs. P4 and P6 look at each other and laugh.

Rob’Autism Project: Being Active in Social Interactions

91

Binomial group 3 (P3 and P5). P3 makes the robot say to P2: “Can you count with decimals?” P6 looks at 34 and smiles at him. He says to him: “Oh Yeah!” At the end of the session, everybody counts and applauds, and P2 makes the robot say: “See you next Wednesday!”

4 Discussion A Non-Distressing Situation. The social situation proposed in the robotic sessions is new for the teenagers. It does not seem to put them in trouble. None of them shows anguish (context 1). Participants know who and what they will ﬁnd there. Over times, the pairs can be changed without posing any particular problem. Everyone manages to separate and then regroup in a new way easily (context 2). These changes allow experiencing in (dis)connections with the rest of the organization of the session. Real Technicians. From observation, they are efﬁcient in the use of the computer tool. To use the robot and its software (Choregraph) supposes a real technical ability. It is not a simple manipulation of object: it is necessary to bind the means and the goals, to choose the elements they need and to order them. As shown in context 1, teenagers seize the robot very quickly. The Robot as an Object of Satisfaction. The robot is also a strong investment object among the teenagers. We noticed that they do not seek immediate satisfaction (by seizing the robot from the outset, for example). They show that they do not allow themselves anything and everything. They wait for the robotician’s explanation. P2 tried to turn on and off the robot several times but was able to expire the realization of this intention, and even changed during the session of wish by wanting to make speak the robot. It is in the use of the robot that they ﬁnd some pleasure: a displacement of wishes could take place. The robot allows a certain freedom of speech. With the robot, they can indeed allow themselves to say or do (almost) everything they want. Teenagers do not speak as usual. Through the robot, they allow themselves different words. What is prohibited in everyday life is possible with the robot. Slang words are undoubtedly enabled by the fact that the interaction situation through the robot is a manufactured situation, a “staging” controlled by the user. Indeed, the teenagers make the robot act according to their wish. And the public plays the game: they listen to the robot, not to the user himself. Over time, the user has the role of a technician. There are many technical operations to perform to achieve efﬁciency. But from the spectators’ side (the rest of the group who watches and listens to the robot), they are forced to enter the proposed situation without really being able to change it. This greatly reduces the complexity of the interaction. Also, the audience returning a positive judgement on what has been implemented by the user allows him existing as an individual in this artiﬁcial micro-society. Reciprocal Social Interactions. The reciprocal social interactions take place gradually. Context 1 shows that the robot and the computer are very quickly the object of a joint attention. However, coping with each other seems difﬁcult. Binomial groups 2 and 3 do not really seem to be able to collaborate. They are together, but they do not try to bring novelty to each other’s proposals. The binomial group 1, which seems to be able

92

R. Gaboriau et al.

to maintain contacts in the relationship and collaborate, does not seem to be able to build a real cooperation with reciprocal contributions. They can share an emotion, but they do not look at each other and they cannot jointly build an exchange. Even with the adult, additional pieces of knowledge can be provided, but they do not contribute to progressive exchanges. The use of the robot does not open at ﬁrst to more composition with the other. We can think that the “nux nux” forces the other to answer the question without starting a game of mutual contribution. That is what happens: once the answer is given, the dialog ends. This difﬁculty concerning the joint and complementary construction of an exchange is found in the analysis of spontaneous speech acts and speech acts produced through the robot. What linguists call the illocutionary force (social meaning: expressive, directive) seemed well identiﬁed, but did not open a link of composition. The exchange is never continued. The propositional contents (representations) are not enriched, but simply sometimes identical. Nothing indicates that these elements brought are really lived in connection with the rest of the session in cohesion. For its part, P2 tends to reduce the tool to a speciﬁc production without being able to enter an interaction. He seems to be trying to reproduce with the robot what he does elsewhere, that is, to count. The situation of P5 may be different from the others. Unlike the others, he can quite jointly build an exchange, be in a group (he is the only one to refer to the group). However, staying focused, respecting the given framework is difﬁcult for him. Very quickly, he turns to the others, gets up, and goes to see the other binomial groups: he is a little scattered. He often has to be brought back to the present situation. He tends to adhere to others’ proposals, responds as echo. Context 2 shows that complementary contributions are starting to appear. The interactions are less focused on an element that remains isolated. Information can now be integrated into a broader exchange that makes sense. The contribution of the other begins to be taken into account. So, the interaction situation becomes more complex. P2’s counts are now associated with the entire situation. P5 begins to take P1’s solicitations into consideration. A kind of “big dialog” is taking place. And it is not just the robot that is being answered, but the user himself. The consideration of the other, of his intention of communication, seems more important. A certain game with intentions is even emerging. One interpretation of the mental states of the other seems thus present (theory of mind). It is not just a sharing of emotions anymore. The other can now have an intention. Now it seems possible to lend him the capacity to be able to seek a certain proﬁt, and to frustrate this search or to make him endure some relaxation (thanks to the introduction of a game between what is identical and what is different: “40, 15”, “1000 asses”, “decimal numbers”). A change of perspective is therefore at work. A “decentration”, that is to say the capacity to be able to change one’s point of view and take some distance, is therefore at work. All these elements seem to conﬁrm the interest of this project in that it favors reciprocal social interactions.

Rob’Autism Project: Being Active in Social Interactions

93

5 Additional Notes Two elements seem important to complete this analysis. Firstly, it is not the exchange with the robot that is valued during the sessions. In the Rob’Autism project, the robot is a tool, nothing else (not an interlocutor as in the companion robot paradigm). And it is indeed as a tool that teenagers use it. Since the beginning of the project, participants have never seen it as anything other than a machine, their machine, that helps them interact with others. Secondly, the idea of active participation of teenagers implies that they are assigned assumed roles and responsibilities. It is a kind of procuration that is immediately offered to them. So we place importance on the recognition of their contribution to the preparation for the show. It is also a way of linking their individual productions to the whole project. We conceive them as a service rendered to the common realization. The show presented to a wider audience further broadens this social recognition. Many participants use this experience to talk about it outside or ﬁnd a career orientation.

6 Conclusion This paper focuses on the effects of using a robot as an extension on the capability to perform social interactions. Our approach relies on the active role and the creativity of the participants to promote their identiﬁcation as social actors. Pragmatic analyses of social interactions during the sessions show that it is effective. Participants show a great interest for the robot and a real mastery of this tool. Not only reciprocal social interactions are taking place but a better consideration of the others appears. Future work includes more precise evaluation of speciﬁc observed effects and a better understanding of the object robot to better use it in the therapy context.

References 1. Huijnen, C.A.G.J., Lexis, M.A.S., de Witte, P.: Robots as new tools in therapy and education for children with Autism. Int. J. Neurorehabilitation 4 (2017). https://doi.org/10.4172/23760281.1000278 2. Pennisi, P., Tonacci, A., Tartarisco, T., Billeci, L., Ruta, L., Gangemi, S., Pioggia, G.: Autism and social robotics: a systematic review. Autism Res. 9, 165–183 (2016) 3. Charron, N., Lewis, L., Craig, M.: A robotic therapy case study: developing attention skills with a student on the Autism Spectrum. J. Educ. Technol. Syst. 46, 137–148 (2017) 4. Chevalier, P., Martin, J.-C., Isableu, B., Bazile, C., Lacob, D.-O., Tapus, A.: Joint attention using human-robot interaction: impact of sensory preferences of children with autism. In: 25th IEEE International Symposium on Robot and Human Interactive Communication (ROMAN), New-York, pp. 849—854 (2016) 5. Pennazio, V.: Social robotics to help children with Autism in their interactions through imitation. Res. Educ. Media 9, 10–16 (2017)

94

R. Gaboriau et al.

6. Petric, F., Mikli, D., Cepanec, M., Cvitanovi, P., Kova, Z.: Functional Imitation Task in the Context of Robot-assisted Autism Spectrum Disorder Diagnostics: Preliminary investigations. In: 26th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), Lisbon, pp 1471—1478 (2017) 7. Salvador, M.J., Silver, S., Mahoor, M.H.: An emotion recognition comparative study of autistic and typically-developing children using the zeno robot. In: Proceedings of the 35th Chinese Control Conference, 27–29 July, Chengdu, pp. 7064–7069 (2016) 8. Yun, S.S., Choi, J., Park, S.K., Bong, G.Y., Yoo, H.: Social skills training for children with autism spectrum disorder using a robotic behavioral intervention system. Autism Res. 7, 1306–1323 (2017) 9. Begum, M., Serna, R.W., Yanco, H.A.: Are robots ready to deliver autism interventions? a comprehensive review. Int. J. Soc. Robot. 8, 157–181 (2016) 10. Lehoux, M.-C., Gayadeen, S.: L’efﬁcacité du robot dans les interventions ciblant les habiletés sociales des enfants présentant un TSA. http://www.crditedmcq.qc.ca/download. asp?18807 11. Conti, D., Nuovo, D., Buono, S., Di Nuovo, A.: Robots in education and care of children with developmental disabilities: a study on acceptance by experienced and future professionals. Int. J. Soc. Robot. 9, 51–62 (2017) 12. Gaboriau, R., Sakka, S.: Le robot comme médiateur thérapeutique: une expérience auprès des jeunes autistes. Tétralogiques 22, 249–262 (2017) 13. Sakka S., Gaboriau, R., Picard, J., Redois, E., Parchantour, G., Sarfaty, L., Navarro, S., Barreau, A.: Rob’Autism: how to change autistic social skills in 20 weeks. In: International Workshop on Mediacal and Social Robots (2016) 14. Boullier, D.: L’impossible fraternité des ondes. LARES, Rennes (1985) 15. Sakka, S., Gaboriau, R.: A robotic puppet master application to asd therapeutic support. Int. J. Mech. Aerosp. Ind. Mechatron. Manuf. Eng. 11, 1483–1491 (2017)

Thermal Comfort Assessment: A Study Towards Workers’ Satisfaction in Metal Industry Norma de Melo Pinto1 and Kazuo Hatakeyama2(&) 1

2

Ergonmics Consulting Ofﬁce, Vitoria, ES, Brazil [email protected] Enterprise Consulting Ofﬁce, Salvador, BA, Brazil [email protected]

Abstract. The acceptable thermal comfort in work environment is mandatory for workers undergoing heavy load of work. This paper focuses on two unique research groups working on research related to thermal comfort. The ﬁrst research group focuses on research related to environmental chambers held within controlled environment, while the other group is dedicated to studying the research ﬁeld in which the researcher does not interfere with environmental variables, personal and subjective, and does not determine the activities carried on or clothing. This paper is based on research focusing on the ISO standard for working under speciﬁc national legislation on the subject of thermal comfort. In industries, work which is associated with heat or thermal energy presents added physical risks for various professional activities. This paper focuses on environment involves parameters affecting the gains or losses of heat of the body. Keywords: Thermal comfort Predicted Mean Vote Predicted percentage of dissatisﬁed Acceptability of thermal environment

1 Introduction Comfort and thermal stress are associated with the subjective human sensations conditioned to physical, physiologic and psychological factors. Thermal stress may be derived from a psychophysiological state to which an individual undergoes when exposed to environmental situations of extreme low temperatures: cold stress, or extreme heat: heating stress. A thermal stress is a subjective response which leads an individual to express personal thermal dissatisfaction resulting from changes in the body caused by environmental excessive high or low temperatures. Engineers and architects devoted their research work to the ﬁeld of thermal comfort have created the common awareness to the effect of thermal discomfort on any plant or facility operation and operators risks and overall process sustainability, and economical feasibility due to environmental thermal comfort effect.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 95–100, 2019. https://doi.org/10.1007/978-3-030-02053-8_15

96

N. de Melo Pinto and K. Hatakeyama

2 Objective This study aims at examining thermal comfort in metal industry while operators performing standard activities. Measure, analyse and compare the environmental and personal thermal variables as compared to the utilized standards determined by the existing legislation and regulations.

3 Background Araujo et al. [1] and Xavier et al. [2] published surveys on environment thermal comfort at schools, as well as researchers [3] covered studies at industry plants. Ruas [4] covered studies focusing on the environment thermal comfort and carried out a survey to know whether the environments are thermally adequate for the individuals and their work activities therein. Usually, the environment thermal comfort research and studies divided into two groups: The ﬁrst group of researchers developed a survey with the use of acclimatized chambers placed inside the environment and the thermal variables which kept under the surveyor’s control. They maintained control either to the environmental or the individual personal variables adjusting them to the thermal sensations and preferences of a better part of the individuals pertaining to the researching group. The second group of researchers carried out ﬁeld studies along which the researchers does not interfere in the environmental and individual personal variables. This establishes neither the activities developed by the individuals pertaining to the research group nor the clothes that they are to use. The comparison of both the environment variables and that of the individual personal ones to the standardized ones takes into consideration the sensations and the thermal preferences of employees in the plant facilities. They pertain to the studies carried on, the environmental thermal features as well as the responders´ perceiving emphasized therein. It is required that a speciﬁc labour legislation, created in Brazil, for thermal comfort and thermal stress, once the existing one are based on the international legislation. Along the time, researchers have been focusing their studies in the theme thermal comfort, among them [5] who have determined an equation and the thermal comfort zone. Whereas [6] worked out the Predicted Mean Vote (PMV) as well as the Persons´ Percentage of Disliking (PPD), while [1] have studied the parameters for thermal comfort in the school buildings. On the other hand, [7] searched about both thermal stress and thermal comfort in industry plants; in addition, [8] searched into the adapting and sustainable thermal comfort; [9] have conﬁrmed the applicability of the ISO 7730 only in laboratorial experiments, whereas [3] assessed the perceptive sensation percentage of the employees working in steel making plants. Furthermore, Gouvea [10] assessed thermal comfort in the environment of a manufacturing plant, and Gemelli [11] assessed the radiance and acoustics thermal comfort of a school environment. Other researchers [12] studied the methods for assessment of thermal comfort in damping and heating areas of Brazil; and [13] has improved the method of thermal comfort adjustments and, at last, [14] gathered within

Thermal Comfort Assessment

97

only one binding the prior and the current norms related to thermal comfort and thermal stress. However, the landmark on this subject has been the results from the studies of [6] as regarding to the changes of thermal heating in an individual body as an effect of his being conﬁned in a work environment. Such studies guided the researchers to interpretations in accordance with the International Standards of [15], as well as the [16] and [17]. The latest interpretation for ISO 7730 [17] ratiﬁes the studies of (1970) and (1972) [6] and aggregates therein the studies of [8] about an individual spotted discomfort. However, this research deals with environments shown to be homogenous and without no spotted discomfort. To that purpose, [17] deﬁnes the individual´s imaging and expressed satisfaction or dissatisfaction with the heating, cooling or pleasant environment. However, when referring to thermal comfort it is required that measurements made in accordance with the standard norms so that the parameters compared and analysed. This study is about the thermal comfort and the factors to be analysed shall be those of the PMV and PPD adopted by the International Standard [17].

4 Survey This paper embodies the second group or research focus, as it aims to knowledge the conditions and parameters to determine the rates of an environment thermal comfort in the ﬁeld of an industry of iron forging products plant facilities. It analyses the environment and the individual personal temperature variances, as well as the subjective parameters for the environment works where the heating in excess is or is not involved, in order to go in search of new knowledge to improve and update the existing literature on thermal comforts. Nevertheless, the purpose of this study is not to analyse the thermal discomfort caused by cooling or heating in excess at places not suitable for the human body. Proceeded to assess the conditions and perceptions of the thermal environment, made influence of variables and the comfort with subjective parameters in metal industry. The ﬁeld research carried out and forecast environmental variables measures with application of questionnaire forms for the collection of personal variables of the individuals, as well as their subjective parameters along the following stages [18]: 1. Collection of environmental variables from the research apparatus “comfortmeter Sensu” scheduled to register the measure every 30 min; 2. Collection from the questionnaires indicated votes for thermal sensations and preferences of the employees observing the garment features wore; 3. Calculation of the air related speed (Sar) and the temperature of the mean radiation. 4. To forecast the metabolic rate in accordance with metabolic rate for speciﬁc professional activities of ISO 8996 (2004); 5. To forecast the garments thermal insulation in accordance with ISO 9920 (2007); 6. To apply the PMV and PPD indexes in accordance with [6] as ruled by [17]; 7. To compare the PMV sensations and the real PPD dissatisﬁed according to the [17]. 8. Analysis of the variables standards;

98

N. de Melo Pinto and K. Hatakeyama

9. Analysis of the simple regression so as to detect if the PMV model has close correlation with the industry; 10. Multiple regression to ﬁnd the analysis equation formula for multiple regression; 11. Non-linear adjustment analysis.

5 Results and Discussion The thermal sensations reported by the respondents highlighted that 75% of the employees felt thermal discomfort due to the heat. In the measures of 54 out of 72, the values of sensations were above 0.5. Only 5 of the measurements indicated cooling discomfort. In the studies of Fanger [6], the values of Sens between – 0.5 to +0.5 present sensation of thermal comfort; the data indicates that 25% of the measures are in the standard of thermal comfort for moderate environment. The choice for more refreshing environments such as (+1 < Pref < + 3) represents 40.27% of the interviewed employees which correspond to 29 measurements; while the 19.44% of them reportedly stated to feel thermally comfortable and, the 12,50% stated to prefer that the thermal environment is not altered. 5.1

Descriptive Statistics of the Study Variables

With the data of the Table 1 shown below, the discussion of the results of the descriptive statistics and the variables of the study shown in the Table 2.

Table 1. Results from 72 measures Icl Tar (° Sar Trm (° RH PMV PPD Sens. Pref. N M (clo) C) (m/s) C) (%) (%) °. (W/m2) 1 105.00 1.17 18.97 0.28 19.02 85.42 0.44 9.04 -0.67 −0.13 2 129.00 1.13 18.93 0.42 19.11 85.66 0.68 14.72 0.27 −0.20 3 123.00 1.09 19.26 0.42 19.55 87.83 0.58 12.05 0.33 −0.20 70 141.00 1.01 26.13 0.60 27.14 60.95 1.80 67.02 1.86 1.00 71 158.00 1.01 29.94 0.84 31.49 53.97 2.64 95.92 2.71 1.86 72 140.00 1.01 29.62 0.83 30.67 54.43 2.36 90.02 1.71 1.14 N°. = number of measures; M = Average metabolic rate; Icl = Average thermal isolation of clothing; Tar = Average internal air temperature; Sar = Average relative speed of the air; Trm = Average radiant temperature; RH = Average relative humidity of the air; PMV = Predicted Mean Vote; PPD = Persons’ Percentage of Disliking; Sens = Average sensation of real thermal comfort; Pref. = Average real thermal preference.

Thermal Comfort Assessment

99

Table 2. Descriptive statistics of the variables of study Variables M Lcl Tar Trm RH Var PMV PPD Sens. Pref.

N Average Median Minimum Maximum Variance 72 132.70 131.00 105.00 162.22 202.98 72 0.99 0.98 0.74 1.26 0.01 72 23.66 22.94 13.10 37.20 30.87 72 25.76 24.82 14.18 40.36 37.84 72 56.37 57.42 22.08 87.83 235.71 72 0.76 0.65 0.28 2.25 0.14 72 1.29 1.12 −0.42 3.00 0.84 72 44.47 32.84 5.59 99.98 1,090.46 72 1.26 1.38 −1.43 3.00 1.20 72 0.77 0.59 −−0.75 2.80 0.86

Standard Deviation 14.25 0.12 5.56 6.15 15.35 0.38 0.92 33.02 1.09 0.93

6 Conclusion This study surveyed existing environmental thermal comfort within an industry sectors working in the ﬁeld of steel by-products manufacturing activities, however, analysis over other issues such as the deriving thermal stress and whether existing or not salubrious conditions to the employees working therein, are to be the purpose of other forthcoming researches. As to the second speciﬁc goal of the survey, to verify whether the environmental conditions are in accordance with regulating standards of [17], the collected data along the 72 measurements showed to be for PMV and PPD. Results showed that for 73.62% of the measurements, the employees reported their sensations as follows: slightly hot; hot and very hot, which may be considered to be reason for some discomfort in the practice of labour activities, this in general, turns the persons dissatisﬁed and the work environment thermally discomfort. The third speciﬁc goal of this research was to verify whether the correlations among the thermal sensations reported by the employees were in conformity with the regulating standards, the collected data shows a correlation of 81% between the PMV and the thermal sensation. For the surveyed industry, the PMV acknowledges the reportedly stated sensations. Thus, the PMV model [6] is applicable to the thermal sensations and preferences of the employees’ participants of this survey.

References 1. Araujo, V.M.D.: Parameters of thermal comfort for users of school buildings in Brazilian Northeastern shore. Doctor’s thesis. FAU - USP. Sao Paulo (1996) 2. Xavier, A.A.P.: Thermal comfort conditions for high school students in Florianopolis’ region. 198 f. Master’s Dissertation – Department of Civil Construction, UFSC, Florianopolis, (1999) 3. Barbiero, M.: Evaluation of perceptions regarding to thermal environment in metal industry: the case study. 239 f. Master’s Dissertation – EE - UFRGS, Porto Alegre (2004)

100

N. de Melo Pinto and K. Hatakeyama

4. Ruas, A.C.: Evaluation of thermal comfort: contribution to practical application of international norms. 79 f. Master’s Dissertation PGFEC – UNICAMP, Campinas (2001) 5. Hougthen, F.C., Yaglou, C.P.: Determining lines to equal comfort, and determination of the comfort zone. In: ASHRAE Transactions, vol. 29 (1923) 6. Fanger, P.O.: Thermal Comfort: Analysis and Applications in Environmental Engineering. McGraw-Hill Book Company, New York (1970) 7. Hackenberg, A.M.: Comfort and thermal stress in industries: surveys performed in Joinville’s region, SC and Campinas, SP. 270 f. Doctor’s thesis ME. UNICAMP, Campinas, SP (2000) 8. Nicol, J.F., Humphreys, M.A.: Adaptive thermal comfort and sustainable thermal standards for buildings. In: Moving Thermal Comfort Standards into the 21 Century, 2001, Windsor, UK. Proceedings. Oxford Brookes University, Oxford, April 2001 9. Olesen, B.W.; Parsons, K.C. Introduction to thermal comfort standards and to the proposed new version of EN ISO 7730. Energy and Buildings, Amsterdam, vol. 34, no. 6, pp. 537– 548 (2002). Ler foneticamente 10. Gouvea, T.C.: Evaluation of thermal comfort: experience in the clothing industry. 164 f. Master’s Dissertation in EC - DCC, Fac. de EC, Architecture e Urbanism, Campinas (2004) 11. Gemelli, C.B.: Evaluation of thermal comfort, acoustics and lighting in the school building with sustainable strategy and bio-climates: the case of municipal school of fundamental learning Frei Paciﬁc. 175f. Master’s Dissertation in Civil Engineering – UFRGS, Porto Alegre (2009) 12. Andreasi, A.A.: Method for evaluation of thermal comfort in warm and humid region in Brazil. 204 f. Doctor’s thesis in Civil Engineering. UFSC, Florianopolis (2009) 13. De Dear, R.J.: The adaptive model of thermal comfort: Macquarie University’s ASHRAE RP-884 Project (2004) 14. Lamberts, R., Xavier, A.A.P.: Thermal comfort and thermal stress. Laboratory of Energy Efﬁciency in Buildings. Florianopolis, SC (2011) 15. International Organization for Standardization. Moderate thermal environments - Determination of the PMV and PPD indexes and speciﬁcation of the conditions for thermal comfort, ISO 7730. Geneva (1984) 16. International Organization for Standardization. Moderate thermal environments - Moderate thermal environments - Determination of the PMV and PPD indexes and speciﬁcation of the conditions for thermal comfort, ISO 7730. Geneva (1994) 17. International Organization for Standardization. Moderate thermal environments - Determination of the PMV and PPD indexes and speciﬁcation of the conditions of thermal comfort, ISO 7730. Geneva (2005) 18. Pinto, N.M.: Conditions and parameters to estimate thermal comfort in industrial environment in metallurgical sector. Master’s Dissertation in PPGEP, UTFPR, Ponta Grossa (2011)

Abductive Thinking, Conceptualization, and Design Synthesis Dingzhou Fei(&) Department of Psychology, Wuhan University, Wuhan 430072, China [email protected]

Abstract. Design synthesis is a process of cognitive development that aims to manage complexity or seek to avoid confusion. Design is always a comprehensive synthesis of market demand, technology trends and business needs. In the synthesis process, designers attempt to organize, manipulate, trim, and ﬁlter the collected data to form a cohesive conceptual construction system. The design synthesis reveals cohesion and continuity; the combination shows the improvement of the organization, the reduction of complexity and the formation of idealized clarity and conceptualization. However, this cognitive synthesis is often not so obvious or even completely hidden. This article attempts to deﬁne this reasoning process from the perspective of psychology and takes it seriously in its universal signiﬁcance in the entire design process. This paper investigates that the following claims: (1) There are three types of applicability of abductive reasoning for design synthesis including: identiﬁcation of implicit design targets, idealization of innovative design concepts, and diagnosis of violating design constraints or design axioms. These three components have a common basis: conceptualization and reconceptualization. They can be taken as sense making from chaos and uncertainty. (2) Synthesis is an abductive thinking process. Abductive reasoning related to insight and creative problem solving, and it is this creative problem solving that is at the heart of the design synthesis methods. (3) Conceptualization relates three speciﬁc sub-processes: prioritizing, judging, and forging. Conceptualization is changeable. Keywords: Abductive thinking

Conceptualization Design synthesis

1 Introduction Design synthesis is a cognitive development process whose purpose is to manage the complexity or seek clarity of confusion. Design is always a comprehensive synthesis of market requirements, technology trends, and business requirements. In the synthesis process, designers try to organize, manipulate, trim, and ﬁlter collected data to form a cohesive conceptualization-building system. The synthesis reveals cohesion and continuity; the synthesis shows the promotion of organization, reduction and idealization of clarity and the formation of conceptualization. However, this cognitive synthesis is often less obvious and even completely hidden [1]. In this regard, existing research does not provide a persuasive thinking model. The purpose of this paper is to focus on

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 101–104, 2019. https://doi.org/10.1007/978-3-030-02053-8_16

102

D. Fei

the goal of demonstrating that the model of abduction reasoning may be a suitable model. Cognitive psychology attempts to use the cognitive methodology to reveal the secrets behind the design process. In practice, we often witness these scenarios: If you compare the starting state (given data) with the ﬁnal state (concepts), it is not yet clear how to derive the latter from the former. It can be said that the more innovative the output, the harder it is to determine how this idea is based on conceptual changes or conceptual mapping. Design synthesis is a thinking process or perception. Through data manipulation, organization, trimming and ﬁltering, designers can generate information and knowledge. The design process is considered as the reasoning from the initial information and ideas to the ﬁnal concept [2]. The design synthesis is an important activity for concept generation and conceptual design blueprint. Most of current approaches treat design as a problem solving process, where functional analysis plays a key role in analyzing and synthesis the given problem, task, goal, etc. This is because that conceptual design starts from a set of chosen functional requirements However, if we treat design process as problem solving, then our focus may place on how to satisfy the requirements for design and then ignore reasoning patterns behind design processes, which actually matter to us. The cognitive psychology in this aspect can provide powerful insights into design synthesis using laws of reasoning. The deductive reasoning and inductive reasoning are two popular thinking processes, but they seem to be not powerful enough to cover important feature in design process, for instance, creativity in design. As we all know, abduction inference can greatly improve the creation of design creativity. This article attempts to deﬁne this reasoning process from the perspective of psychology and takes it seriously in its universal signiﬁcance in the entire design process.

2 Conceptualization Three types of applicability of abductive reasoning for design synthesis including: identiﬁcation of implicit design targets, ideation of innovative design concepts, and diagnosis of violating design constraints or design axioms [3]. These three components have a common basis: conceptualization and reconceptualization. They can be taken as sense making from chaos and uncertainty. The sense making is an action-oriented process that people automatically go through in order to integrate experiences into their understanding of the world around them. Common to all methods of synthesis is a “sense of getting it out” in order to identify and forge connections [4]. In essence, sense making is an internal personal process, and synthesis can be a collaborative external process. In general, Synthesis requires designers to establish connections between seemingly unrelated issues through the process of selective pruning and visual organization.

Abductive Thinking, Conceptualization, and Design Synthesis

103

3 Abductive Thinking: Throughout Design Process Synthesis is an abductive thinking process. It can be considered as “the step of adopting assumptions, which is put forward by the facts… a form of inference” [5]. Abduction can be considered as the best explanation for forming ideas.. According to previous experience abduction is a logical way of thinking about reasoning or “best guessing” leap. Consider the example When I did A, B happened: I did something like A before, but the situation is not exactly the same. I have seen something like B before, but the situation is not exactly the same. I was able to abduct C because B was the cause. Abductive reasoning is related to insight and creative problem solving, and it is this creative problem solving that is at the heart of the design synthesis methods [6]. Conceptualization relates three speciﬁc sub-processes: prioritizing, judging, and forging connections. The ﬁrst prioritizing ranks all available data and information by their importance, which is guessed by designers’ experiences and the importance is ‘might be” and is consistent with previous examples. The second judging process is typical abductive thinking: judging relevant data and information by guessing. This process of synthesis forces the deﬁnition of relevance, as the designer will pass the gathered data “through a large sieve” in order to determine what is most signiﬁcant in the current problem solving context. Synthesis methods, then, require a constant reassessment of the current state as compared to the unknown end state. The third forging connections is an abductive logical story, positing a hypothesis based on inference. The activity of deﬁning and forging connections actively produces knowledge, in that new elements (gleaned from prior experiences in life) are combined with existing elements. Each of the methods emphasizes prioritization, judging, and the forging of connections. These methods illustrate pragmatic approaches to design synthesis that can be applied in design problems of any discipline or subject matter. Abductive reasoning nicely links different seemingly unrelated design components and ideas. This is the most different place from deductive and inductive thinking. It is this way of thinking that provides greater flexibility for cooperation and integration.

4 Conceptualization Is Changeable These processes are impossible to carryout out in a time, and require iterations and reframing. Reframing is a method of shifting semantic perspective in order to see things in a new way [7]. The new frame “re-embeds” a product, system, or service in a new (and not necessarily logical) context, allowing the designer to explore associations and hidden links to and from the center of focus. These processes lead to changes in conceptual mapping [8]. A concept map is a graphical tool for organizing and representing knowledge. It serves as a kind of template or scaffold to help to organize knowledge and to structure it, even though the structure must be built up piece by piece with small units of interacting concept and propositional frameworks. A concept map is a formal representation of a mental model; a mental model “represents a possibility, or, to be precise, the structure and content of the model capture what is common to the different ways in which the possibilities could occur.

104

D. Fei

5 Conclusion Design synthesis is a thinking process or perception formation through data manipulation, organization, trimming and ﬁltering, designers can generate information and knowledge. All the design process is considered as the reasoning from the initial information and ideas to the ﬁnal concept. As a powerful psychological tool, design synthesis is an process of thinking, or abductive sense making process. Through efforts of data manipulation, organization, pruning, and ﬁltering, designers produce information and knowledge. All knowledge is able to express in sentences and propositions and then design processes is seen as reasoning from initial information and premises into ﬁnal conclusion, and the companied psychological processes subordinates into propositional reasoning. This paper try to deﬁne this reasoning processes in psychological terms and seriously considerate them a pervasive sense makings in whole design processes. This study addresses conceptualizations occurs all stages in design process. Of course, abductive reasoning does not exclude formal logic methods, like the work of Stephen [9]. Perhaps these two different methods can be combined to give a more comprehensive analysis of the design process. However, where formalization and axiomatization are difﬁcult to succeed, models based on abduction are attractive models.

References 1. Veen, J.: The Art and Science of Web Design. New Riders Press, Indianapolis (2000) 2. Kolko, J.: Information architecture and design strategy: the importance of synthesis during the process of design. In: Educational Conference Proceedings. IDSA, San Francisco (2007) 3. Fallman, D.: Design-oriented human-computer interaction. In: Human Factors in Computing Systems, the Proceedings of CHI (Association for Computing Machinery), pp. 225–232 (2003) 4. Klein, G., Moon, B., Hoffman, R.: Making sense of sensemaking 1: alternative perspectives. Intell. Syst. (IEEE) 21(4), 71 (2006) 5. Coyne, R.: Logic Models of Design. Pitman, London (1988) 6. Johnson-Laird, P.: The shape of problems. In: The Shape of Reason: Essays in Honour of Paolo Legrenzi, V Girotto (eds.), pp. 3–26. Psychology Press (2005) 7. De Bono, E.: Serious creativity. J. Qual. Particip. 18(5), 12 (1995) 8. Kolko, J.: Information architecture: synthesis techniques for the muddy middle of the design process. In: 23rd International Conference on the Beginning Design Student Proceedings, Savannah (2007) 9. Liu, A., Stephen, C.-Y.L.: Alternation of analysis and synthesis for concept generation. CIRP Ann. 63(1), 177–180 (2014)

Is Truth Contextual? The Browsing Purpose, the Availability of Comparable Material, and the Web Content Credibility Evaluation Katarzyna Abramczuk1(&), Michał Kąkol2, Radosław Nielek2, and Cezary Biele1

2

1 National Information Processing Institute, Al. Niepodległości 188B, 00-608 Warsaw, Poland {kabramaczuk,cbiele}@opi.org.pl Polish-Japanese Academy of Information Technology, ul. Koszykowa 86, 02-008 Warsaw, Poland {michal.kakol,nielek}@pjwstk.edu.pl

Abstract. This paper discusses to what extent the browsing context needs to be taken into account when considering web credibility ratings. We focus on the availability of other websites on the same topic and the browsing purpose as two factors that can potentially influence how people perceive credibility and how accurate their judgments are when contrasted with expert evaluations. We analyze data from an experimental study in which subjects were asked to rate a set of websites on various dimensions in different contexts. We conclude that more context influences how often extreme evaluations are used, which ratings correlate with the declared knowledge of the topic, and which websites features are brought up as important for formulating the ﬁnal judgment. Keywords: Web credibility Subjectivity Judgment context-dependence Credibility experiment Reliability World wide web

1 Introduction The importance of the Internet as a source of information in many crucial domains is bound to grow. Consequently, ﬁltering out unreliable information and focusing attention on resources that are credible is crucial. This, however can pose a serious challenge, given that evaluation of information is, for the most part, subjective, relative, and situational. Moreover, credibility is a multidimensional and complex concept. It is related but not synonymous with truth [2]. Yet, it can be quite precisely estimated by analysing content, contextual and meta-information [1, 3, 5]. As shown in [4] credibility evaluation is topic-dependent and it changes over time. In the current paper we study how it depends on the browsing context. Our analyses concentrate on three context-related topics. First, we speculate that people who have to analyse the content of the website or compare it with the content of other websites on the same topic become more critical. Second, we conjecture that the © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 105–110, 2019. https://doi.org/10.1007/978-3-030-02053-8_17

106

K. Abramczuk et al.

credibility evaluations depend on people’s topic-related knowledge. Third, we hypothesize that differences in the website viewing process promote focusing on different aspects of the tested websites.

2 Methodology This paper is based on a study conducted at the Polish Japanese Academy of Information Technology (PJAIT) in Warsaw, Poland. We chose Polish websites on seven topics including GMO food, marijuana, “sesja” pills for students, the Dukan diet, homeopathy, hormonal contraception, and aspartam. The websites were chosen in a way that ensured presence of both credible and non-credible content on each topic. We asked two groups of respondents to rate the selected websites: students of PJAIT and subject-matter experts. The experts were asked to rate the websites’ credibility on a 5-point scale. Each website was evaluated by two experts. In the case of a disagreement the experts were encouraged to discuss the problem and establish a common rating. The students were randomly assigned to one of three experimental conditions. In the Free Browse (FB) condition the participants browsed and evaluated 9 websites on 9 different topics1. In the Topic Browse (TB) condition they evaluated 9 websites on the same topic. The Topic Search (TS) condition used the same website sets as TB but the participants had to answer one question related to the subject of the browsed websites, for example: “Is marijuana addictive?”. We asked for evaluations of websites’ credibility, and 7 credibility related dimensions on a 5-point scale. The additional dimensions included website presentation, knowledge and intentions of the author, and clarity, completeness, truthfulness and objectivity of information provided. The participants also wrote textual comments on the reasons guiding each of their credibility ratings. Furthermore, we run a short survey that included participants’ declarations concerning topic related knowledge. After cleaning the data we were left with 624 student ratings (213 in FB, 267 in TB, and 144 in TS) for 58 different websites coming from 87 students.

3 Results We start by presenting our analyses of the distributions of ratings in different conditions. Next, we investigate whether individual topic related knowledge influences the quality of credibility evaluations. Finally, we provide a related qualitative analysis of the textual comments.

1

Websites on two additional topics were added in this condition: orthodontia and vitamin B17 in the context of cancer treatment. Ratings for these websites are not included in the following analyses to ensure comparability across conditions.

Is Truth Contextual? The Browsing Purpose

3.1

107

Ratings Distributions

The ratings of the websites on various dimensions were highly correlated and their means oscillated around the middle of the 5-point scale i.e. 3. They were higher than 3 (One-sample T-test) for credibility (E = 3.13, p = 0.01), presentation (E = 3.18, p < 0.01), intentions (E = 3.28, p < 0.01), clarity (E = 3.89, p < 0.01) and validity (E = 3.54, p < 0.01). Inspecting their distributions across the experimental conditions revealed differences with respect to two aspects: clarity (X2 = 13.40, p = 0.03)2 and completeness (X2 = 16.94, p = 0.03). As it turns out, the observed differences do not corroborate the assumption that people who have to analyse the content of the website or compare it with the content of other websites on the same topic become more critical. Rather their tendency to use extreme evaluations changes. This can be treated as an indicator of their certainty regarding the rating (See Fig. 1). For clarity we can see that both the presence of other websites on the same topic and the need to analyze the content decrease the prevalence of extreme evaluations, especially the highest ratings. Information that seems clear when unchallenged can be seen as less apparent in a context. For completeness the effect is reversed. Comparison with other websites on the same topic increases the number of extreme evaluations both the lowest and the highest. This is probably because the context provides a reference which makes this type of rating easier. To further test the assumption of the existence of differences in the propensity to use extreme evaluations in different conditions we run a generalized linear mixed model for binary data with extreme evaluation (1 or 5) as a dependent variable, experimental condition as an independent variable and random effects for subjects. This test showed signiﬁcant effect of TB condition for the ratings of completeness (z = 2.02, p = 0.04). Another indicator of the varying difﬁculty of evaluations in different experimental conditions is the prevalence of the indeterminate answers (“I don’t know”). In general it is highest for validity (0.22), objectivity (0.13) and completeness (0.11). Anova analysis reveals a varying number of indeterminate ratings of validity in different conditions (F = 3.11, p = 0.04). Rating validity is visibly easier in the TS condition. 3.2

Declared Knowledge

Next we tested whether the declared topic knowledge makes people more critical. Table 1 presents Spearman correlations between the declared knowledge on the topic of the website for different experimental conditions. The effect of asserted expertise is visible only in the two conditions in which other websites on the same topic were also presented. For TB we can see that more knowledgeable subjects were more critical of websites’ presentation and objectivity, and less critical of their clarity. For TS they gave more favourable ratings for credibility, intentions and validity. It is interesting to observe that the effect of knowledge on evaluations refers to more substantial dimensions in the case of TS than in the case of TB. This might be an indicator that in the case of TS the subjects paid more attention to the substantial dimensions. We will come back to this in the next section. 2

For this test the two lowest ratings were grouped together to ensure reliability.

108

K. Abramczuk et al.

Fig. 1. Distributions of ratings of clarity and completeness of information on the evaluated websites on a 5-point scale (1 - low values)

Obviously in general the expected relationship between the topic knowledge and credibility ratings is not linear. One would expect that more knowledgeable people are more critical of the non-credible content and rate the credible content more favourably than the less knowledgeable people. To test whether this is the case we used the expert evaluations of credibility. To acquire reliable estimates we recoded expert credibility ratings, lay credibility ratings and knowledge declarations to binary variables where the two highest ratings were classiﬁed as positive and the remaining ratings were classiﬁed as negative. We ran a generalized linear mixed model for binary data with binary credibility ratings as a dependent variable, binary expert credibility ratings and binary knowledge declarations as independent variables and random effects for subjects. Qualitatively we did observe the expected interaction, but it was not statistically signiﬁcant. The mean frequency of positive credibility evaluations of the non-credible content equals 0.35 for the less knowledgeable subjects and 0.30 for the more knowledgeable subjects. The mean frequency of positive credibility evaluations of the

Is Truth Contextual? The Browsing Purpose

109

Table 1. Spearman correlations between declared topic knowledge and ratings for different experimental conditions. Dimension Credibility

FB 0.02 (p = 0.74) N = 202 Presentation −0.06 (p = 0.42) N = 209 Knowledge −0.01 (p = 0.85) N = 195 Intentions −0.08 (p = 0.25) N = 204 Clarity 0.11 (p = 0.12) N = 208 Completeness 0.03 (p = 0.64) N = 190 Validity 0.07 (p = 0.34) N = 167 Objectivity −0.04 (p = 0.62) N = 192

TB −0.05 (p = 0.45) N = 253 −0.13 (p = 0.04)* N = 163 −0.04 (p = 0.49) N = 245 0.03 (p = 0.65) N = 247 0.18 (p < 0.01)* N = 258 −0.02 (p = 0.72) N = 234 0.03 (p = 0.68) N = 196 −0.17 (p = 0.01)* N = 222

TS 0.18 (p = 0.04)* N = 135 −0.04 (p = 0.67) N = 141 0.15 (p = 0.1) N = 133 0.2 (p = 0.02)* N = 136 0.09 (p = 0.27) N = 140 0.14 (p = 0.11) N = 132 0.23 (p = 0.01)* N = 121 0.17 (p = 0.06) N = 128

credible content equals 0.48 for the less knowledgeable subjects and 0.54 for the more knowledgeable subjects. The main effect of the actual credibility (i.e. expert rating) on lay credibility ratings was pronounced and signiﬁcant (z = 2.24, p = 0.02) indicating that lay evaluations are generally in line with the expert evaluations. 3.3

Textual Comments Classiﬁcation

The analysis of the textual comments corroborates our hypothesis. We categorized users’ comments into three categories: (a) comments referring to easily accessible website features such as: website type, presence/absence of advertisements, internet domain and website presentation (b) comments referring to features requiring deeper analysis of the website content e.g.: information richness, the amount of effort put into preparing the website (is it a professional website), and the inferred intentions of the authors (c) comments referring to relation of the website to the personal experience and attitudes e.g.: whether the subject has had a previous contact with the presented ideas or whether s/he has strong beliefs regarding the origin of the information presented (e.g. church, government). The percentages of comments of a particular type in different conditions are presented in Table 2. Due to the low sample size we were unable to conduct quantitative analyses of the differences between experimental groups. Qualitative analyses, however, suggests that TS experimental condition invited readers to conduct a deeper analysis of the websites and avoid being guided by the individual experience and attitudes, comparing to the other two experimental conditions. Another interesting result is the relatively low percentage of comments ‘requiring deep analysis’ in the FB experimental condition which suggests that users scan the websites superﬁcially when doing that without any particular goal.

110

K. Abramczuk et al. Table 2. Percentage of different comment types in the three experimental conditions. Dimension Easily accessible Requiring deep analysis Individual experience

FB 66% 23% 11%

TB 54% 23% 14%

TS 62% 34% 4%

4 Summary Our results indicate that subjective and situational factors are of relatively small importance for credibility evaluations, but there are some speciﬁc cases in which they matter. For example participants in the FB condition used extreme evaluations more often than the students that saw a set of websites on the same topic. It seems that more context leads to more balanced opinions. This effect is stronger when people are asked to answer domain related questions. Secondly, topic related knowledge does affect how people perceive website content but only when their browsing is limited to one topic. Furthermore, it correlates with the more substantial dimensions of the evaluation when browsing is purposeful i.e. requires gathering some new information. Finally, we can formulate a qualitative conﬁrmation of the hypothesis that people pay attention to different website characteristics depending on their browsing focus and context. These preliminary ﬁndings suggest that a further study of contextual determinants of website perception is worth pursuing. Acknowledgments. This work is partially supported by Polish National Science Centre grant 2015/19/B/ST6/03179.

References 1. Wierzbicki, A.: Understanding and measuring credibility. In: Web Content Credibility. Springer, Cham (2018) 2. Liu, X., Nielek, R., Adamska, P., Wierzbicki, A., Aberer, K.: Towards a highly effective and robust Web credibility evaluation system. Decis. Support Syst. 79, 99–108 (2015) 3. Wawer, A., Nielek, R., Wierzbicki, A.: Predicting webpage credibility using linguistic features. In: Proceedings of the 23rd International Conference on World Wide Web, pp. 1135–1140. ACM, April, 2014 4. Nielek, R., Wawer, A., Jankowski-Lorek, M., Wierzbicki, A.: Temporal, cultural and thematic aspects of web credibility. In: International Conference on Social Informatics, pp. 419–428. Springer, Cham, November, 2013 5. Kakol, M., Nielek, R., Wierzbicki, A.: Understanding and predicting web content credibility using the content credibility corpus. Inf. Process. Manag. 53(5), 1043–1061 (2017)

All Doors Lead to the Kitchen – Sustainability and Wellbeing Challenges in a Shared Centrepiece of Living Soﬁe Andersson(&) and Ulrike Rahe Department of Industrial and Materials Science, Division of Design & Human Factors, Chalmers University of Technology, 412 96 Gothenburg, Sweden {sofiean,ulrike.rahe}@chalmers.se

Abstract. The kitchen ﬁgures a central place in the home where a signiﬁcant share of a household’s resource consumption takes place. Sharing the kitchen between multiple households has potential to bring positive sustainability effects due to more efﬁcient use of both material resources and energy. The concept of shared kitchens has, however, thus far had a limited diffusion. This paper explores the potential of shared kitchens as a future sustainable living environment by studying user experiences from a Living Lab setting. It builds the base for an overarching larger European collaboration on how future shared kitchens should be designed in order to support everyday practices while optimising the conditions for achieving positive impact on both sustainability and wellbeing. Findings are presented from ﬁve focus areas concerning different use contexts: (1) accessing, (2) cooking, (3) living and socialising, (4) storing, and (5) cleaning. Keywords: Kitchen

Sharing Sustainability Wellbeing Living lab

1 Introduction The average household size in the European Union is decreasing, with single households accounting for one third of all households, thereby being the most common group as well as the one that has increased the most during the last decade [1]. Smaller households are in general less resource-efﬁcient than larger households where space, energy, furnishings and transportation are shared between several household members [2, 3]. Although energy efﬁciency of buildings and appliances has improved signiﬁcantly over the last 30 years, a growing demand for living space and appliances has been found to counterbalance energy efﬁciency improvements from technological developments [4]. Despite increasing incomes and consumption in Western countries, wellbeing levels have been found to remain static and even decline, which is a paradox that indicates potential to reduce consumption without compromising levels of wellbeing [5]. An opportunity to step away from the individualistic and materialistic path towards wellbeing is offered by the concept of sharing. Sharing is an ancient, fundamental human behaviour [6] that, despite offering resource saving potentials [7] as well as social and © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 111–116, 2019. https://doi.org/10.1007/978-3-030-02053-8_18

112

S. Andersson and U. Rahe

economic beneﬁts, has thus far had a limited diffusion [8, 9]. It is therefore interesting to explore ways to create sustainable living solutions that could appeal to a wider group of people, allowing for compromises between private and shared spaces [10]. The kitchen ﬁgures a central place in the home that is becoming an increasingly multifunctional living space [11]. Thereby, it serves as a good starting point for evaluating the potential of sharing as a future form of sustainable living.

2 Research Approach and Methodology 2.1

Research Aim

This paper aims to explore the potential of shared kitchens as a future sustainable living environment by studying user experiences from a Living Lab setting. The following research questions are posed: How should a shared kitchen be designed in order to optimise the conditions for both sustainability and wellbeing? How should Living Labs proceed in experimenting with shared spaces as a future form of sustainable living? 2.2

Research Context

The study was performed in a Living Lab in Sweden, which besides from being a research environment for future sustainable living is also a residential building containing 29 dwellings for students and researchers. Five of these are private apartments and the remaining 24 are private rooms organised into four architecturally identical ‘clusters’. In each cluster, six private rooms surround a spacious common area containing a shared kitchen, two bathrooms, a living room section, a hallway and a balcony. The private rooms measure only 3.6 3.6 3.6 m and contain a sleeping loft, a small bathroom without a shower and a small ‘kitchenette’ without cooking possibilities. The primary interest of this study is the shared cluster kitchen, which consists of two rows of kitchens mirrored to each other, together forming a long island placed centrally in the common area. The furnishings and appliances of the two kitchen rows are identical, apart from one base cabinet which has been replaced with a dishwasher on one side of the kitchen. Furthermore, the kitchen contains two fridge/freezer columns with a small top cabinet, two ovens with microwave function and cabinets above, two induction cooktops covered by a large hood, base cabinets with drawers and two sink units. The dining area is an extension of the countertop, which forms a bar table for up to six people at the farthest end of the kitchen, facing a large window with the balcony outside. 2.3

Interviews

Seven Living Lab residents signed up for an interview, representing three of the four clusters and one of the private apartments. For this paper, only the six interviews with cluster residents have been analysed. All interviews were held in the homes of the interviewees and were audio recorded, with permission from the participants.

All Doors Lead to the Kitchen

113

The interviews were of a semi structured character, following a prepared template but adding follow-up questions whenever needed. First, participants were given floor plans of the cluster living space and encouraged to talk about how they used it, mark out areas that they felt positive or negative about and describe any changes they would like to make. The focus was then turned more speciﬁcally to the kitchen area with questions regarding its use, layout, design, placement, appliances, storage possibilities, support for sustainable behaviour and sharing aspects. The ﬁnal part of the interviews had a forward-looking perspective on future dwellings, sharing and sustainability. Interview recordings were transcribed and imported to the software NVivo, where it was coded into different themes. Quotes included in the paper have been translated from Swedish to English by the ﬁrst author.

3 Findings 3.1

Accessing

Several interviewees had previous experiences from shared kitchens, which were often placed at the bottom of a corridor. This kind of ‘corridor kitchen’ was described as a rather anonymous area where “you never know if anyone stole your food or not”. One interviewee described her previous corridor kitchen as “quite disgusting since it hadn’t been cleaned properly for… well a very long time”. In contrast, the central position of the cluster kitchen seemed to promote responsibility among the cluster residents due to its visual and physical presence: “Since all doors open to the kitchen it is easy to, if someone does not clean the dishes, you can say “hey, deal with this!”, while if it had been a kitchen in some corridor you don’t really know… so as a shared kitchen it works very well.”

It was also found that a central location may, on the other hand, result in a lack of calm spaces in the kitchen. The many doors accessing the common area limit the possibilities for utilising the space properly and one interviewee explained that: “…what makes this apartment difﬁcult is that you have compromises all the time […] It becomes a hallway and a living room, it becomes a hallway and a kitchen.”

3.2

Cooking

Although common meals were found to be a rare occurrence in the clusters, several interviewees appreciated the possibility to cook at the same time without necessarily sharing the same dish: “Since everyone has different needs and lunch boxes, it would be such a big project to try to cook together. But at the same time, it’s very nice to cook and eat together even though you somehow make your own thing.”

The cluster kitchen was described as having enough space for three or four persons to cook at the same time, even though a large share of the workspace was barely used, due to a high fridge/freezer column “cutting it off” from the rest of the workspace. With the two sides of the kitchen mirrored to each other, cluster residents were able to cook

114

S. Andersson and U. Rahe

face-to-face with another roommate instead of just facing a wall. One interviewee explained how this had made cooking and daily chores such as washing dishes more fun, which in his case had contributed to a habit of cooking at home more often. The uneven distribution caused by having one dishwasher only resulted in a disrupted workflow. The cluster residents constantly had to walk around the long kitchen island since all dishes ended up on that side, the reason why one of the interviewees mentioned that “the kitchen doesn’t work at all, it’s a roundabout”. Cooking in the shared cluster kitchen was found to place high demands on the quality and performance of appliances due to frequent use and wear. In particular, two of the appliances were given several complaints. The ﬁrst one was the hood: “The hood doesn’t work, it cannot be turned off or on. It’s sometimes on and sometimes off and it doesn’t suck out any air at all so it gets very, very smelly, so we usually open the door to the balcony, which is not so energy efﬁcient.”

The second one was the combined oven and microwave, which was explained to function poorly both as an oven and as a microwave, although one interviewee appreciated the space efﬁciency provided by its multifunctionality. In one of the clusters, both ovens had broken so that one did only function as an oven and the other one only as a microwave. An interviewee from that cluster thought that: “The oven is a bad solution, you should not mix the microwave with the oven, it just gets really messy because it boils over in the microwave and then burns in the oven.”

3.3

Living and Socialising

The possibilities for socialising was frequently mentioned as the very best aspect of the shared cluster kitchen, which was described as the centre of the cluster and the place where you most frequently meet your roommates. The dining area, located at the end of the kitchen island where it faces a large window with the balcony outside, seemed to create a pleasant environment where the residents, apart from eating, sometimes also sit down and have a chat, surf on their computers or study. However, taking the form of a bar table ﬁxed to the cooking area, the interviewees found potential for improvements both in terms of comfort and flexibility. Several mentioned that they wanted a “real” kitchen table of normal height, where they would have the possibility to sit down and with more than six people at the same time. 3.4

Storing

Separation and clarity was found to be a key issue when it comes to storage of food, waste and kitchenware. Most importantly, as one of the interviewees stated, there needs to be enough storage space. For kitchenware and pantry goods, there seemed to be more than enough storage provided by the many base cabinets with drawers. Cooling and freezer space was, however, explained to be less abundant in the shared cluster kitchen. One interviewee explained that as a student, it would be good from an economic point of view to be able to store more food in the fridge and freezer. Since the participants mainly cooked their own food, they did not share groceries to a large extent, apart from some spices, cooking oil, sauces and baking goods. The main

All Doors Lead to the Kitchen

115

part of their groceries was kept in separate drawers and on separate shelves in the fridge. For kitchenware, it was perceived as a positive thing to be able to share equipment within the kitchen but also noted that things wear out faster, causing a resistance to share items you care for with others. Regarding waste storage, one interviewee explained that since everyone cook their own food, the cluster kitchen is subject to a large amount of waste: “In a family you cook food together but here we cook three separate meals on each side per day, which means that we normally produce more waste.”

The storage space originally dedicated to waste in the cluster kitchen was restricted to one drawer containing three separate bins in each sink cabinet, comparable to the conﬁguration of a standard Swedish kitchen. It became clear that this storage space was inadequate, both in terms of the amount and the opportunity to separate fractions. Waste separation was consequently described as a challenge by all interviewees. In one cluster, waste storage was expanded to other kitchen drawers as well and in another cluster, waste ended up in the hallway once the standard drawer was ﬁlled up. Among the cluster residents, waste sorting was explained to be performed to varying degree and it was a subject that occasionally caused some irritation. Some of them had stopped sorting some categories of waste since they moved there just because it was too big of a project: “Normally, I would have sorted soft plastics and everything but there is no space so I can’t take it.”

3.5

Cleaning

In the Living Lab, there is a cleaning service taking care of the bathrooms and floors in the common area of the clusters, which mainly leaves the task of cleaning up the kitchen to the residents. In the three clusters included in the study, cleaning was not organised by the residents according to any system. In order for it to work, one interviewee explained that: “I think the most important thing is that everyone has a similar sense of responsibility towards the common areas, because if everyone would accept that it’s messy, or could live with having it that way, then it would be okay, but as long as someone deviates from it – that’s when it turns into a problem.”

The fact that you need to take the responsibility to clean up after yourself was mainly seen a positive thing among the interviewees, even though it may demand a little more effort compared to a single household kitchen where leaving the mess for a while longer doesn’t affect anyone else. However, as one of the interviewees said: “you can’t expect it to be pedantic in a shared kitchen”.

116

S. Andersson and U. Rahe

4 Discussion and Conclusions A central position of the kitchen in relation to other (private) living spaces was found to support socialising but also to promote a sense of responsibility in its usage. However, the kitchen also needs to include calm spaces to support wellbeing. Regarding cooking, it can be concluded from this study that a shared kitchen requires appliances of high quality and a good amount of workspace supporting several people to cook at the same time. When designing a shared kitchen, there is an opportunity to focus on the social aspects of cooking and bring people together rather than keeping them apart at the kitchen’s different ends. Storage space need to be spacious enough with clearly delimitated sections, especially in order to support a well-functioning system for waste recycling. Having a common sense of responsibility among all users was regarded as key for shared products as well as shared tasks such as handling waste, washing dishes and cleaning up. A suggestion for future research in Living Labs is to investigate requirements placed on shared kitchens used by a wider variety of households. Topics that need further consideration are how to achieve higher flexibility and adaptability for different use cases and how to improve the balance between centrality and calm spaces in the kitchen. Acknowledgments. This work is part of the European collaboration project The Circular Kitchen, supported by the Climate-KIC initiative of the EIT, Area of Focus Urban Transition.

References 1. Eurostat: Household composition statistics. https://ec.europa.eu/eurostat/statistics-explained/ index.php/Household_composition_statistics 2. Keilman, N.: Biodiversity: the threat of small households. Nature 421(6922), 489–490 (2003) 3. Klocker, N., Gibson, C., Borger, E.: Living together but apart: material geographies of everyday sustainability in extended family households. Environ. Plann. A. 44(9), 2240–2259 (2012) 4. Gram-Hanssen, K.: Efﬁcient technologies or user behaviour, which is the more important when reducing households’ energy consumption? Energ. Efﬁ. 6(3), 447–457 (2013) 5. Jackson, T.: Where is the “wellbeing dividend”? nature, structure and consumption inequalities. Local Environ. 13(8), 703–723 (2008) 6. Belk, R.: Sharing. J. Consum. Res. 36(5), 715–734 (2010) 7. Leismann, K., Schmitt, M., Rohn, H., Baedeker, C.: Collaborative consumption: towards a resource-saving consumption culture. Resources 2(3), 184–203 (2013) 8. Jarvis, H.: Saving space, sharing time: integrated infrastructures of daily life in cohousing. Environ. Plan. A. 43(3), 560–577 (2011) 9. Mont, O.: Institutionalisation of sustainable consumption patterns based on shared use. Ecol. Econ. 50(1–2), 135–153 (2004) 10. Hagbert, P.: “It’s just a matter of adjustment”: residents’ perceptions and the potential for low-impact home practices. Hous. Theory Soc. 33(3), 288–304 (2016) 11. Hand, M., Shove, E., Southerton, D.: Home extensions in the United Kingdom: space, time, and practice. Environ. Plan. D: Soc. Space 25(4), 668–681 (2007)

User as Customer: Touchpoints and Journey Map Camila Bascur(&), Cristian Rusu, and Daniela Quiñones Pontiﬁcia Universidad Católica de Valparaíso, Av. Brasil No. 2241, 2340000 Valparaíso, Chile [email protected], {cristian.rusu, daniela.quinones}@pucv.cl

Abstract. Customer eXperience (CX) is a concept traditionally related to marketing and Service Sciences. Lately there is an increasing interest on CX from the Human-Computer Interaction (HCI) community. CX extends the User eXperience (UX) concept: it examines the whole customer journey and experiences with several systems, products or services that a company offers, instead of focusing in a single one. CX is one of the most important factors when it comes to maintaining a competitive advantage. The paper examines a speciﬁc case study, when a user wants to buy a ticket through a virtual travel agency. We identify the phases of the lifecycle of the consumer’s journey, the existing points of contact between customer and company through systems, products, services (touchpoints), and the emotions that could predominate in each of the stages. Keywords: Customer Experience (CX) Customer Journey Map Touchpoints

Customer lifecycle

1 Introduction Customer eXperience (CX) is one of the most important factors when it comes to maintaining a company competitive advantage with its peers. Customers have greater power and influence in companies and they have the means to make them valid, such as websites or social networks, among others. They know more about the products, services, competitors and price offered by companies and they are looking for unique and memorable experiences that accompany the delivery of systems, products and services [1]. The relationship between companies and customers is covered by various stages ranging from the initial awareness of a potential customer, through the shopping experience, until after the use of the product and/or service [2]. CX is a concept traditionally related to marketing and Service Sciences. Lately there is an increasing interest on CX from the Human-Computer Interaction (HCI) community. CX extends the User eXperience (UX) concept. The ISO 9241 standard deﬁnes UX as “The perceptions and responses of people, as a result of the anticipated use, of a product, system or service” [3]. The paper examines a case study, focusing on a user who wants to buy a ticket through a virtual travel agency. We identify the phases of the lifecycle of the consumer’s journey, the existing points of contact between customer and company through © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 117–122, 2019. https://doi.org/10.1007/978-3-030-02053-8_19

118

C. Bascur et al.

systems, products, services (touchpoints), and the emotions that could predominate in each of the stages. Section 2 briefly analyzes the CX concept. Section 3 describes the case study, highlighting the touchpoints and the experiences that may occur. Finally Sect. 4 presents conclusions and future work.

2 The Customer eXperience Concept CX is a concept that encompasses both customers and companies, and refers to the physical and emotional experiences that occur through the interactions with the products, systems and/or services of a brand, occurring from the ﬁrst direct and conscious contact until the post-consumer stage [4]. CX is strictly personal and involves the customer’s participation at different levels [5–7], transcending the moment of the purchase. Lemon and Voerhoef deﬁne CX as “a multidimensional construction that focuses on the cognitive, emotional, behavioral, sensory and social responses of the client to the company’s offers throughout the customer’s purchase process” [8]. CX generates a bond between the customer and the product, providing a memorable experience for the customer, which adds value to the product [9]. Schmitt suggests the existence of ﬁve types of CX as a basis for a general analysis of experiential marketing [10]. They are present in different magnitudes throughout the customer “journey”, generating perceptions that can be both positive and negative: • Sense: derives from the ﬁve senses of the human being (sight, hearing, smell, taste and touch), and can be exploited in such a way as to affect the user who wants to acquire something. • Feeling: mainly stimulates the senses and internal emotions of customers, establishing a strong connection between customers and suppliers of products/services making customers feel attracted to the brand or product and have positive emotional responses. • Thought: is responsible for integrating experiences with the habits of a group of people in a creative way, allowing them to understand the experience and increase the interest in the experience provided by the marketing operators. • Act: integrates behavioral options, such as, the interaction a client has with a given product/service; these behaviors leave a lasting impression or become a direct unconscious response. • Relative: transcends individual and personal emotions, associating the ideal self with other people or cultures. To better understand what CX means we have to consider all stages that the customer crosses, and the different types of experience. The stages and types of experiences are covered by what is known as the customer’s lifecycle, which is made up of four phases [11, 12]: • Acquisition phase: begins when the consumer (future customer) feels the need to acquire a new product/service and develops knowledge about the organization that offers it. The consumer is in charge of evaluating all the options available in the

User as Customer: Touchpoints and Journey Map

119

market and to see which of these is able to better meet their needs. Finally, the consumer becomes a customer. • Loyalty phase: the maximum value is reached when the product/service is used, the client is fully aware of the brand. • Reactivation phase: occurs when the customer decreases the intensity of use of the product/service, this leads to the customer’s distance to the brand. It is where customers could be attracted by the competition. • Winning back phase: reconquers users who used to be customers. The main objective is to generate value to the former customer through actions that the company considers relevant.

3 The Customer Journey Map: A Case Study The points of contact between customer and company, through systems, products, services, are called touchpoints. They correspond to any interaction (including encounters where there is no physical interaction) that might alter the way that your customer feels about your product, brand, business or service [13]. Touchpoints are of vital importance as they are the means by which a company discloses information to its former and future clients that is relevant to take into account when wanting to acquire a new product/service, thus generating interaction between suppliers and customers. The interaction can be active, passive, direct or indirect and perceived by any of the senses that the human being possesses. Often companies carry out a Customer Journey Map (CJM), which is a strategic management tool widely used today to understand CX in an organization [14]. Its objective is to diagram the beginning-to-end steps that the client follows when interacting with a certain organization. It identiﬁes both positive and negative aspects, and ﬁnds critical points in which improvements are necessary [15]. The diagram’s points represent perceptions in a certain stage. Positive aspects are those that produce certain satisfaction in the customer. Negative aspects generate in the concern or dissatisfaction. Critical aspects (highlighted in the diagram) require particular attention by the companies. Neutral aspects are the combination of the above-mentioned aspects. Figure 1 describes the CJM of a traveler that purchases plane tickets through a virtual travel agency. It identiﬁes CX stages, touchpoints, and perceptions. Table 1 presents in each of its columns the touchpoints that deﬁne the CJM for each of the stages that the traveler goes through. Table 2 analyzes the type of experiences that may occur during the customer’s journey with their respective touchpoint, which is represented with a letter in image 1; it should be noted that for this case of study there are no touchpoints that are related to the type of experience relative.

120

C. Bascur et al.

Fig. 1. The customer journey map for an airline passenger who purchase the ticket from a virtual travel agency. Table 1. Touchpoints. Trip preparation Company web sites Price comparison Word of mouth Email

Arrival at the airport Word of mouth

Control and wait Events

Flight

Display Sponsor ship

Display

Catalogues Promos

Gift

Destination arrival Discount Newsletter Loyalty program

Sales person Packaging

Table 2. Experiences through the customer’s journey. Sense (A) Search of destination (B) Investigation of destination (C) Online booking (F) Check in (H) Search of boarding gate (J) Flight information (K) Search of departure gate (N) Information of the area

Feeling (D) Booking information (E) Reception of ticket (O) Search for a transfer

Thought (G) Luggage reception (P) Mail reception

Act (I) Security control (L) Boarding (M) Service on board

Relative

User as Customer: Touchpoints and Journey Map

121

4 Conclusions and Future Work Today more and more companies are interested in the way they offer their products and services to consumers, so the concept of CX is increasingly used. It refers to all the interactions that the consumer has with an organization. These interactions occur within the lifecycle of the client, across several phases during the customer’s journey. The case study that we present allows analyzing the basic concepts of CX: phases, touchpoints, perceptions, and CJM. Many organizations have to improve CX, as one of the most important issue that deﬁnes customer’s loyalty. CX is a complex and very personal phenomenon. Therefore, CX studies are time and cost consuming. HCI studies usually focuses on UX, dealing with a single product, system or service. CX offer a holistic view instead. That is why we think CX should be an HCI topic, not only a Service Science one. As future work, we intend to focus on how traditional UX evaluation methods could also serve to evaluate certain CX aspects. Acknowledgments. This work was supported by the School of Informatics Engineering of the Pontiﬁcia Universidad Católica de Valparaíso – Chile. Camila Bascur has been granted the “INFPUCV” Graduate Scholarship.

References 1. Maital, S.: The Experience Economy: Work is Theatre & Every Business a Stage. MIT Sloan Management Review, Cambridge (1999) 2. Jantsch, J.: Duct Tape Marketing: The World’s Most Practical Small Business Marketing Guide. Thomas Nelson Inc., Nashville (2011) 3. ISO 9241-210: Ergonomics of human-system interaction — Part 210: Human-centered design for interactive systems. International Organization for Standardization, Geneva (2010) 4. Laming, C., Mason, K.: Customer experience—an analysis of the concept and its performance in airline brands. Res. Transp. Bus. Manag. 10, 15–25 (2014) 5. Vanharanta, H., Kantola, J., Seikola, S.: Customers’ conscious experience in a coffee shop. Procedia Manuf. 3, 618–625 (2015) 6. LaSalle, D., Britton, T.A.: Priceless: Turning Ordinary Products into Extraordinary Experiences. Harvard Business School Press, Brighton (2003) 7. Schmitt, B.: Experiential marketing. J. Mark. Manag. 15(1–3), 53–67 (1999) 8. Lemon, K.N., Verhoef, P.C.: Understanding customer experience throughout the customer journey. J. Mark. 80(6), 69–96 (2016) 9. Chen, S.C., Lin, C.P.: The impact of customer experience and perceived value on sustainable social relationship in blogs: an empirical study. Technol. Forecast. Soc. Change 96, 40–50 (2015) 10. Schmitt, B.H.: Experiential Marketing: How to Get Customers to Sense, Feel, Think, Act, and Relate. The Free Press, New York (1999) 11. Una visión diferente. https://jaimeporta.com/2016/01/22/el-ciclo-de-vida-del-cliente/ 12. InformaBTL. https://www.informabtl.com/cual-es-el-ciclo-de-vida-del-cliente/

122

C. Bascur et al.

13. The CEO Refresher. http://www.refresher.com/alrpmtouchpoint.html 14. Rosenbaum, M.S., Otalora, M.L., Ramírez, G.C.: How to create a realistic customer journey map. Bus. Horiz. 60(1), 143–150 (2017) 15. Javier Mejias. https://javiermegias.com/blog/2013/04/customer-journey-map-mapaexperiencia-cliente/

Enhancing User Experience with Embodied Cognition SuKyoung Kim(&) Sapporo City University, Geijutsu-no-mori 1, Minima-ku, Sapporo 005-0864, Japan [email protected]

Abstract. The emerging design disciplines and research are rapidly transforming, and not only designer-driven, but also user-driven are prominent values in the design process. This paper presents a considerable design discipline towards enhanced user experience focused on embodied cognition, by distinguishing the following deﬁnitions in the concept model: human-being, subjective innate ﬁlter, and experience. In the approach, the suggestion in embodied cognition were considered that people’s cognitive processes are tightly related to the interaction between their body and the physical environment for enhancing user experience in their daily lives. By verifying the relationship between subjective innate ﬁlter and relevant variables, it clariﬁes how to enhance user experience with embodied cognition in design. Keywords: Design theory Research methodology Inclusive User experience

Evaluation

1 Introduction How to enhance user experience with embodied cognition? Clarifying the process of getting knowledge could be the initial point to answer this question. There can be no doubt about the fact that all human knowledge begins with experience. Humans experience external features through their eyes, ears, nose, and so on. However, as I. Kant said that although all our knowledge begins “with” experience, it does not follow that it arises “from” experience [1]. Even though human knowledge is a compound of that which human receive through impressions, we don’t distinguish from that raw material until long practice has made us attentive to it and rendered us capable of separating one from the other [1], and the accumulative knowledge is in relation to intuition. The unanswered question is that how does an individual’s experience affect or is affected in relation to knowledge and intuition? Rethinking what design should “be” and “do” is one of the most prominent approaches to enhancing user experiences with embodied cognition as the strong intuitions.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 123–129, 2019. https://doi.org/10.1007/978-3-030-02053-8_20

124

S. Kim

2 Re-thinking the Deﬁnitions in Design The author has designed and developed the concept model to clarify the relationship between the perceived information obtained from the external feature and the individual’s subjective evaluation results, such as emotion and decisions. Initially, the concept model was designed aiming to deﬁne Kansei information, which is to explain how individuals change or modify their decisions beyond rational decision process. Figure 1 which is the ﬁrst phase concept model, shows the process of how human beings perceive and evaluate external features subjectively. Based on the 3 steps evolving concept models, it is possible to re-think the deﬁnitions in design, and clarify what is design for and what is does.

Fig. 1. The concept models in the ﬁrst phase explaining a subjective innate ﬁlter considering the variety of subjectivity (left) the second phases explaining the modiﬁed-targets of experience (right)

2.1

User as “one individual”

The term Kansei, which is similar to the term “affective” in English, has been developed in order to design feelings into products [2]. The author created the concept model aiming to clarify the decision-making process in design, and also to verify the relationship between perceived information and the outputs, i.e., reaction, and decisionmaking. Figure 1 (left) shows the ﬁrst phase of the concept model which is based on the methodologies in previous Kansei studies [3]. During the ﬁrst phase, the author postulates individual as a subjective innate ﬁlter considering the variety of subjectivity in her studies in the dissertation for PhD degree, and this subjective innate ﬁlter separate individuals as “one individual”. By subjective innate ﬁlter, inner and outer “can be separated”, and it “can explain” the reason why the subjective evaluations are different even with as the same stimuli as given all individuals. Figure 1 (right) shows the evolving concept model in the second phase presenting how experiences influence on subjectivity, focused on not only subjective innate ﬁlter but also understanding. This evolving concept model aimed to explain the reason why an individual decision is changed time to time both in subjective reactions and rational thinking process [4]. The second phase concept model convinces of that why an individual decision arrives at a different conclusion with the same situation and information, e.g., facts and data. It clariﬁes the scope of “experience” which influences

Enhancing User Experience with Embodied Cognition

125

both “subjective innate ﬁlter” and “understanding”. It changes the ability of “one individual” to perceive the intended meaning of the situation or information. Figure 2 shows the third phase of the evolving concept model focused on the process of “differences” of perception, cognition, and recognition in the affective process [4, 5]. The two types of outcomes influence the perceptual and conceptual fluencies as familiarity, preference, aesthetic feeling, and so on; and the two interact to impact user evaluations of a new product again through individual experience [4, 5]. In the latest concept model, external feature is explained with “a separating substance”. As to clarify the difference between cross and inter-modal perception, “external feature” is disassembled into ﬁve elements of scent, taste, sound, geometric, haptic. When “an individual” senses “an apple”, ﬁve different variables perceive with ﬁve different sensors. And the perceived information can become an uncountable number of data depending on the pair of the elements of “the external feature” and “the ﬁve sensors”.

Fig. 2. The concept model in the third phase explaining the relationship of perception, cognition, recognition in the affective process

2.2

What Is Design

Design is a part of our daily lives. It means that designers do not only have to develop new products with which people pursue their lives but also have to decide on the kind of life and society which these products will support [6], i.e. the so-called transformative design. To rethink the notion of transformative design for the not so far future, it should be understood that not focus on what design is, per se, but focus on what design will be expected by users. In other words, user research is about understanding users and their needs, and user experience design is about designing based on the user’s interactions with a product from moment to moment [7]. Therefore, clarifying the evaluation process of “one individual” starts from an external feature perceived and

126

S. Kim

arrives at results, such as reactions and decisions. A user’s experience is the cumulative effect of many factors, some that can be controlled and some that cannot be controlled [8]. However, as for the design, the external features elements are “designed” depending on the intention of the designer.

3 Empirical Researches Two empirical researches were conducted to verify the hypothesis with the concept models. Experiment 1 veriﬁes whether there is more effective information in the evaluation of multisensory process, and some evaluation value, such as sweetness, influences broadly in the evaluation of multisensory process. Experiment 1 consists of threefold phases (1) stimuli-screening for the main experiment (2) stimuli-producing for the main experiment (3) and, evaluation. The ﬁrst phase veriﬁed the subjective sensory giving of the visual and olfactory stimuli, such as sweetness, sourness, lightness, and so on. In the second phase, screened visual and olfactory stimuli were combined to form the experimental stimuli according to the visual-olfactory congruence degree. In the ﬁnal phase, the subjects evaluated the prepared visual-olfactory stimuli which had the same traits as the stimuli-screening. The results present that (1) Positive-feeling-triggered smell as the quality of having weight, affects the evaluation words sweetness and sourness. Also, Positive-feeling-triggered smell as the quality of having weight, affects the evaluation word preference only when paired with light visual stimuli. (2) Positivefeeling-triggered sweet smell is more effectible than sweet feeling visual stimuli. Positive-feeling-triggered sweet smell has negative effects on the evaluation of sweetness. It is reasonable to assume that positive-feeling -triggered sweet smell is more effectible stimuli except as the same value as sweetness (3) Negative-feelingtriggered sweet smell affects in sweetness and preference. Through the research, it is veriﬁed that negative feelings from visual information is modiﬁable with positive olfactory information in the case of LIGHTNESS and SWEETNESS conditions. Furthermore, negative feeling from olfactory information affects sweetness feelings. Experiment 2 was conducted in Japan and Finland to compare the cognitive differences in “letters”. Letters are important symbol when using a language. What might be just a letter in some languages could be considered to be a symbol in other languages. For example, ä is a letter in Finn, whereas it is considered to be a symbol in Japanese. Experiment 2 aims to make clear the perception category, whether the stimulation displayed is a letter or a symbol using Japanese characters. Confusing Japanese letters in Hiragana and Katakana, were used as stimuli. Finn and Japanese nationalities participated in the experiment to investigate the effect of language familiarity on the relationship between perception category and sureness. It was hypothesized that the modulation of language familiarity is attributed to the perceptual fluency of reading and writing languages. The results of the effect of language familiarity present that well-established biases of native speakers reduce misunderstanding in letter perception; well-established biases intervene in intuitive aesthetic evaluation process to reduce those aesthetic values. The ﬁndings support “proper” propositions of design focused on the target user, whether it is for native speakers or

Enhancing User Experience with Embodied Cognition

127

foreigners when using letters. Knowledge is an important factor which helps understand the outer world. However, knowledge-based understanding categorizes what the symbol means before the individual percepts what and how its shape and character trigger feelings. The aim of the second experiment was to clarify the influence of understanding in language.

4 Consideration and Conclusion The aim of the present paper is twofold: (1) to clarify the validity of the suggested methodological approach to enhancing user experience, also (2) to verify the reliability of the methodological approach in design doing. Firstly, prominent terms in design were re-deﬁned to build a conceptual framework for integrating user experience into objects focused on the relationship between perception and cognition in individual subjective evaluation to designed external stimuli, e.g., objects and services. Secondly, two practical experiments were conducted to conﬁrm the validity of the suggested conceptual framework. From the results of Experiment 1, it is veriﬁed that there is the more effective information i.e., olfactory in the evaluation of multisensory process, and some evaluation value such as sweetness, influences broadly in the evaluation of multisensory process. The results present that sweetness is relevant not only to taste but also to other perception such as olfactory and visual as an embodied cognition. Sweetness is a basic taste most commonly perceived when eating foods rich in sugars. Newborn human infants also demonstrate preferences for high sugar concentrations in breast milk [9, 10]. It is a convincing explanation that in the natural settings that human primate ancestors evolved in, sweetness intensity should indicate energy density, while bitterness tends to indicate toxicity [11–14]. And, sweet tastes are regarded as a pleasurable experience, except in excess. It is assumed that sweetness did not show any statistical signiﬁcance in the evaluation with sweet stimuli. Further research will be focused on these issues. The aim of the second experiment was to clarify the influence of understanding in language. Knowledge is an important factor which helps understand the outer world. However, knowledge-based understanding categorizes what the symbol means before the individual percepts what and how its shape and character trigger feelings. The ﬁndings of these two experiments show that, (1) less visualized information has more influence in the evaluation of multisensory process than that of the fully visualized information; (2) less certainty leads to stronger sureness in the evaluation of knowledge-based evaluation than fully certain information does. The ﬁndings from the results present that experiential knowledge is related to embodied cognition, and those relationships allow individuals to function effectively [14]. Because, practice is a signiﬁcant method of knowledge creation, creating skill and knowledge for the practitioner, and deepening existing skills and knowledge. Because they are properties or attributes of the individuals who experience and embody them, however, they need not be shared with the external world or the larger community to be effective [14]. While embodied cognition suggests that people’s cognitive processes are tightly related to the interaction between their body and the physical environment [15, 16],

128

S. Kim

growing research has centered on the role of bodily sensory-motor mechanisms in understanding languages [17], user experiences with external features, such as products [18], and movement-based interactions [19]. What is the feeling, and how does the feeling influence on an embodied cognition? The ﬁndings of the researches could hold the answer to this question and present the role of embodied cognition in transforming the design era. It frames an enhanced-evaluation method in the unexplored potential of user experience design. Acknowledgments. Each experiment was supported by KAKENHI and Sapporo City University.

References 1. Kant, I.: Critique of Pure Reason. Penguin Classics, London (2008). Revised edition 37 (a), 23 (b) 2. Schütte, S.: Designing Feelings into Products: Integrating Kansei Engineering Methodology in Product Development. Linköping Studies in Science and Technology Thesis, no. 946 (2002) 3. Kim, S., Cho, Y., Niki, K., Yamanadka, T.: Integrating affective values to sustainable behavior focused on Kansei engineering. Int. J. Sustain. Eng. 9(6), 378–389 (2012) 4. Kim, S., Cho, Y.: Towards enhanced affective design: rethinking the notion of design. In: The Proceeding of International Conference on Design Engineering and Product Innovation (2017) 5. Kim, S.: Design for Transformation by Minding the Gap between Visual and Olfactory Perception. World Design Summit 2017 (Montreal Canada 10.16-20) (2017) 6. Hummels, C., Frens, J.: The reflective transformative design process. In: Proceeding of CHI EA 2009: Extended Abstracts on Human Factors in Computing System, pp. 2655–2258 (2009) 7. Hassenzahl, M.: User experience and experience design. In: The Encyclopedia of HumanComputer Interaction, 2nd edn. (2013). Chapter 3 8. Bule, L.: The User Experience Team of One: A Research and Design Survival Guide Rosenfeld Media, 1st edn., 9 July 2013 9. Desor, A., Maller, O., Turner, E.: Taste acceptance of sugars by human infants. J. Comp. Physiol. Psychol. 84(3), 496–501 (1973) 10. Schiffman, S.: Taste and smell in disease (Second of two parts). N. Engl. J. Med. 308(22), 1337–1343 (1983) 11. Altman, S.: The monkey and the ﬁg: a Socratic dialogue on evolutionary themes. Am. Sci. 77, 256–263 (1989) 12. Johns, T.: With Bitter Herbs They Shall Eat It: Chemical ecology and the origins of human diet and medicine. University of Arizona Press, Tucson (1990) 13. Logue, W.: The Psychology of Eating and Drinking. W.H. Freeman, New York (1986) 14. Friedman, K.: Conversation, Discourse, and Knowledge. She Ji J. Des. Econ. Innov. 3(2), 75–82 (2017) 15. Gibson, J.: The Ecological Approach to Visual Perception. Houghton Mifflin, Boston (1979)

Enhancing User Experience with Embodied Cognition

129

16. Wilson, M.: Six views of embodied cognition. Psychon. Bull. Rev. 9, 625–636 (2002) 17. Gibbs, W., Wilson, L.: Bodily action and metaphorical meaning. Style 36, 524–540 (2002) 18. Van Rompay, T., Hekkert, P., Saakes, D., Russo, B.: Grounding abstract object characteristics in embodied interactions. Acta Psychol. 119, 315–351 (2005) 19. Loke, L., Robertson, T.: Moving and making strange: an embodied approach to movementbased interaction design. ACM Trans. Comput. Hum. Interact. 20, 7 (2013)

A Study on the Effect of Human Factor for Atypical Design in the Architectural Design Studio Yungil Lee(&) Department of Architecture, Hoseo University, 20 Hoseo-ro 79beon-gil, Baebang-eup, Asan-si, Chungcheongnam-do, Republic of Korea [email protected]

Abstract. This study evaluates the effect of human factor on atypical building design. The research method involved 77 students in authentic design courses who proposed atypical design for playgrounds before and after deploying various human-ﬁgured images. They then scored their experiences based on using the human-ﬁgured images. Statistical analysis of those scores discloses that deploying human-ﬁgured images helps students evaluate the validity and functionality of design solutions, ﬁnd various functions and improve safety and convenience. These results illustrate the usability of human factor simulation in the atypical design process. The ﬁndings of this study can contribute to developing a computational tool for education in atypical architecture design. Keywords: Human factors Design studio Evaluation

Human behavior Atypical architectural design

1 Introduction Human factor is the most important standard for evaluating the quality of design alternatives. It is sometimes ignored in the design process, which places more emphasis on atypical shapes. Recently, atypically shaped buildings have become a worldwide trend in architectural design. Due to community requests for new architectural features, a lot of unusual shapes created by parametric design methods are appearing in the city. From a pedagogical aspect, a design process based on the parametric design method is good because it presents the possibility to create new shapes. However, the student should consider the human behaviors guided by the designed physical shapes. Conventionally, human factor is known as a worthwhile tool for designing a built environment. Unfortunately, it is hard to ﬁnd studies on the effects of human factor on the digital design process. This study evaluates the effect of human factor on atypical building design. The research method involved 77 students in authentic design courses who proposed atypical design for playgrounds before and after deploying various human-ﬁgured images. They then scored their experiences based on using the human-ﬁgured images. Statistical analysis of those scores discloses that deploying human-ﬁgured images helps students evaluate the validity and functionality of design solutions, ﬁnd various functions and © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 130–134, 2019. https://doi.org/10.1007/978-3-030-02053-8_21

A Study on the Effect of Human Factor for Atypical Design

131

improve safety and convenience. These results illustrate the usability of human factor simulation in the atypical design process. The ﬁndings of this study can contribute to developing a computational tool for education in atypical architecture design.

2 Methods A total of 77 students in the spring semesters of 2016 and 2017 participated in a computer-aided architectural design course called Digital Design Lab 1. They were all third-year university students at H University in South Korea majoring in architecture, which is a ﬁve-year professional program. One of the purposes of this course was to develop students’ ability to generate new architectural forms using a digital design tool. Thus, the instructor asked the students to design a playground that would allow children to participate in plenty of activities through various types of spaces and sculptures. To design the playground, the students used a digital modeling tool which allowed them to model atypical shapes freely. The commercial digital modeling tool they used was Rhino 3D. In one week, the students designed a playground and made a digital model. During this process, they designed physical shapes for children based on the students’ own previous experiences and knowledge. In the following week, the instructor asked the students to put the human images illustrating human behavior into the designs according to the physical shapes of the playground as a human behavior evaluation process. Finally, students were asked to upgrade the designed playground according to the result of the evaluation (Fig. 1).

Fig. 1. The design results of atypically shaped playgrounds using the human-ﬁgured shapes.

After completing the playground design tasks, the students completed the survey questionnaire. They compared the two design results—before and after the human behavior evaluation—and answered a questionnaire. The questionnaire used a seven-step

132

Y. Lee

Likert scale (0 = not at all; 7 = very much). The assessment criteria, which indicated creative problem ﬁnding, included (1) conﬁdence in evaluating the usability and value of the design results, (2) exploring novel and unexpected forms and spaces, (3) ﬁnding varying functions of the design solutions, and (4) examining the reliability of the safety and comfort of the playground users. The criteria stemmed from existing creativity theories that were modiﬁed to match the contexts of the given design tasks [1, 2].

3 Results A paired-samples t-test was conducted to compare the students’ self-evaluation scores on creative problem ﬁnding before and after the human behavior evaluation. (1) In the comparison of student conﬁdence in evaluating the usability and value of the design results, the t-test results indicated that when the students designed the playground after evaluation, they had more conﬁdence in evaluating the usability and value of the design results (M = 3.93, SD = 1.26) than when the students designed the playground before evaluation (M = 4.51, SD = 1.04), t(77) = –4.228, p < .000. (2) Analysis of the results also indicated that when the students designed the playground after evaluation, they found more varying functions of design solutions (M = 4.41, SD = 1.21) compared to when the students designed the playground before evaluation (M = 4.87, SD = 1.00); t(77) = –3.604, p < .001. (3) In the comparison of the reliability of the playground users’ safety and comfort, analysis of the results indicated that when the students designed the playground after evaluation, they found more reliability regarding the safety and comfort of the playground users (M = 3.90, SD = 1.36) than when the students designed the playground before evaluation (M = 4.41, SD = 1.26); t(77) = –3.090, p < .003. However, considering criteria (4), the results indicate that when the students designed the playground after evaluation, they didn’t ﬁnd more novel and unexpected forms and spaces (M = 4.42, SD = 1.196) than when they designed the playground before evaluation (M = 4.74, SD = 1.151); t(77) = -1.918, p < .059. Table 1 presents the results of the statistical analyses in detail.

Table 1. The results of the paired-samples t-test on creative problem ﬁnding before and after the human behavior evaluation. M(SD) Before evaluation (n = 77) 3.93 (1.260)

Conﬁdence in evaluating the usability and value of the design results Exploring novel and unexpected forms 4.42 (1.196) and spaces Finding varying functions of the design 4.41 (1.217) solutions Examining the reliability of the safety and 3.90 (1.368) comfort of the playground users *p < .05, **p < .01.

After evaluation (n = 77) 4.51 (1.046)

Mean P difference

–0.584

0.000**

4.74 (1.151)

–0.311

0.059

4.87 (1.004)

–0.454

0.001**

4.41 (1.260)

–0.506

0.003**

A Study on the Effect of Human Factor for Atypical Design

133

4 Conclusion and Discussion This present study investigated the relationship between a human behavior evaluation based on behavioral descriptions and creative problem ﬁnding. According to the results of the analysis, the human representations in the atypical architectural design process enhance the students’ conﬁdence in evaluating the usability and value of the design results. With human behavioral representation, students can ﬁnd more various functions of design solutions. As well, human behavioral representation has a positive effect on students’ reliability concerning the safety and comfort of the playground users. Taken together, the results indicate that the representation of user behavior in atypical architectural design has an overall positive effect. In other words, human representation influences architects’ problem ﬁnding and relevant design development not only in the ﬁeld of conventional design but in atypical design as well [3]. However, the presence of human behaviors could not facilitate the ﬁnding of more novel and unexpected forms and spaces. In the previous research, Hong and Lee show that reductive human shapes and limited behaviors restrict diverse conceptions of the relationship between the spaces and the occupants. As well, traditional human representations are mostly used to deduce numerical appropriateness, such as scale assessment. This result could be caused by the students’ familiarity with these conventional usages of human representation and because they only used several types of human images. Despite the overall positive effect of human representation in atypical design, the representation of various human behaviors is needed to enhance the ﬁndings of more novel and unexpected forms and spaces [4, 5]. The present study’s limitations and suggestions for future research are as follows. First, it is necessary to study the differences in user’s behavioral expressions between atypical design and general design. Second, it is necessary to study the difference between dynamic user behavioral representation and static user behavioral representation. As well, the speciﬁc effects of user behavior were not investigated. Despite these limitations, the present study reveals the hypothetical effects of behavioral representation on designers’ creative ﬁndings in atypical architectural design. Acknowledgments. This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2018R1A2B6005827).

References 1. Amabile, T.M.: Creativity in Contexts. Westview Press, Boulder (1996) 2. Boden, M.A.: Creativity and Arts: Three Roads to Surprise. Oxford University Press, New York (2012)

134

Y. Lee

3. Paletz, S.B.F., Peng, K.: Problem ﬁnding and contradiction: examining the relationship between naïve dialectical thinking ethnicity, and creativity. Creativ. Res. J. 21, 139–151 (2009) 4. Hong, S., Lee, Y.: A study on the effectiveness of using human behavior representation on creative motivation in architectural design process. Archit. Inst. Korea 30, 93–101 (2014) 5. Imrie, R.: Architects’ conceptions of the human body. Environ. Plan. D Soc. Space 21, 47–65 (2003)

E-material Creating and Formatting Application Kristine Mackare(&), Anita Jansone, and Maksims Žigunovs Faculty of Science and Engineering, Liepaja University, Liela iela 14, Liepaja, Latvia [email protected]

Abstract. Technologies are in daily use, also for educational reasons and getting knowledge. Reading from screens and from papers are different. People are in need to adapt but it is slow process. It can make some different problems that affect people life quality. There is a need for solution. We offer one of the possible solutions: E-material formatting application, what improve comfort of using e-material in learning and study process and decrease near work load by adapting on individual needs. It is based on developed recommendations for user-centric and adaptive educational e-material creation and formatting. Possible technical solutions have been found. Keywords: Application

E-material E-study Formatting Technologies

1 Introduction Technologies are the part of our everyday life. Use of it in e-study and for getting knowledge is prevalent and will increase. People can’t stop this modern progress. As information is presented by pictures, video, and audio formats very often, still the massive amount of information comes from the textual material. Such smart and digital devices like computers (both desktop computers and laptop computers), tablets, smartphones, book readers, etc. and internet directly or indirectly are used in both formal and non-formal education [1]. All near work, especially viewing texts and images, reading, learning both print media and computer screen, are making effects on vision. As it is known, reading from printed materials and screen is different. Reading research shown that screen reading is much difﬁculty [2] and readers need to use different reading model [3]. People are adapting, but the evaluation process isn’t as fast as technology progress. The visual system needs help. We can’t stop development, but we can try to help achieve more comfortable and healthy screen reading. There is a need for user-centric and adaptive educational e-materials. Users ask for the more individual approach for screen work which should be applied for natural and comfortable perception and with visual processes, thus helping the learning process and facilitating memorization.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 135–140, 2019. https://doi.org/10.1007/978-3-030-02053-8_22

136

K. Mackare et al.

One possible solution is the customized e-material creating and formatting application for using on MOODLE type pages. It is a central goal of research. Methodology: As the research content of several parts methodology are full and content of literature research, several e-surveys, statistical data research, research on possible technical solutions. To achieve the goal, ﬁve research steps have been created: 1. 2. 3. 4. 5.

stage stage stage stage stage

– – – – –

development of formatting recommendation for e-materials research on the e-material automatic formatting need research on the possible technical solution (4 methods) development of application based on concept requirements approbation.

For this moment, steps from 1 to 3 have been completed, and steps 4 and 5 remain.

2 Recommendations for Methodologies Guidelines To successfully participate in e-learning and e-studies, users need both excellent eskills and well-designed e-learning materials: high-quality content, comfortable, easyto-understand and comprehensible text, suitable formatting parameters of e-materials. First, there is a need to develop recommendations for methodologies guidelines of ematerial formatting parameters. There have been viewed and analysed more than 100 different literature sources. Based on literature research the survey was made to carry out research on users’ habits and preferences of varying text formatting parameters - font style, size, spacing, colour of text and background, for e-study materials. It consists of several parts. Practical part was more analysed. It brings to conclusions which help to develop recommendations. Most respondents chose a larger font size, even very young people. Not all the results from the survey are the same as it was found in a literature review. Also, there are made some hypotheses about gender differences in preference [4, 5] (Table 1). In this connection, some recommendations for guidelines was made for target group without reading difﬁculties and without any signiﬁcant vision problems [4, 5]. Also, predictable theoretical recommendations for the group over 40 years old have been made. Recommendations based on survey, literature review and vision science. These recommendations must be checked experimentally in the target group by combined complexes in the context of e-material to conﬁrm their effectiveness.

E-material Creating and Formatting Application

137

Table 1. Recommendations for guidelines Target group by age 7–15

16–39

40+a

Formatting parameters Font Body text size Arial 12–18 pt Verdana TNR Arial Verdana Georgia Arial

Headings size 14–20 pt

Line spacing 1,15

14 pt

16 pt

1,5

14–16 pt

16– 18pt

1,5

Verdana Georgia

Background and text colour Black on white Dark grey on white White on black Black on white Dark grey on white Dark green on white Black on white Dark green on white Very dark grey on white

a

Theoretical This work deals with part of the target group – children (7–15 y.o.) and adults (16–38 y.o.).

3 The Concept for Application Prototype Second, the concept for application prototype had made. It based on developed recommendations for e-material formatting and by users’ individual factors and need. The idea involves such component as an idea of the tool by itself: what it must do and how, what information must contain in the database, and vision of design. It must give a clear understanding of the application to the programmer what must be done. The app must work for both – e-material creators and e-material users (Fig. 1). Concept consist of: 1. The aim of application a. Target group - for whom app is intended to work and use b. Get information about users or target group c. Analyse of information d. Provide advices e. E-material formatting option (by recommendation) 2. What application should do a. Get basic user information (age, gender, work or education scope, vision problems or refraction correction) b. If user wants more individual advices – get additional users information (about reading and learning difﬁculties and complains, vision problems or eye diseases, speciﬁc working conditions, other health conditions or diagnoses, etc.) c. Get basic information about target group d. Made information analyses (followed by scheme) e. Provide possible advice for formatting, based on analyses

138

K. Mackare et al.

Fig. 1. Sequence diagram for application concept [6]

f. Made e-material formatting by advice (font type, size, spacing, text and background colour) g. Possibility to cancel or change changes in formatting h. Additional formatting tools, if user wants to do it manually or in addition (zoom, colour masks) 3. How application should work a. Fast and qualitatively b. Get information and make analyse c. Suggest formatting advice d. Make formatting for e-materials and text e. Need to be possible format popular document formats - word, pdf, excel, etc. 4. Data base a. Users proﬁles b. Users answers c. Suggested advices for e-material formatting d. User choice - ask the question whether to use the choice that was used in the previous connection session e. Data about users’ comfort responses after use of suggested formatting

E-material Creating and Formatting Application

139

5. Interface, design, architecture a. Must have an individual user proﬁle and login b. User information - survey-generated questions – option c. During the analysis, there should be a statement that an analysis is being made (moving image with text?) d. A suggestion window with a visual example of the proposed formatting format, as it will look like, if applied e. Possibility to save advised format f. Button for additional formatting g. At the end of the work - assessment of subjective feelings (to be written and executed shortly) 6. Description of the case a. Registration b. Authorization c. Document opening d. Uploading e. Create your own criteria f. Rate the formatting g. Save

4 The Possible Technical Solution Research on the possible technical solution for making application prototype has been done. Four method evaluation descriptions (pros and cons) and four method developed solutions have been tested. Four methods chose for analyses: Visual Basic, Macros, XML, OLE. The most appropriate solutions have been found to make app working correctly as formatting device for different text document formats. The XML approach has much more opportunity for future modiﬁcations and researches. This method is recognized as more advantage methods than others. XML Pros and cons: + Every object data type management. + Deep changes in data linking. + Non-OS binded. + Non-environment oriented. + Options for previously not known object creation. – A lot of programming. – Complicated structure. – No debugging. For technical solution: The XML approach means that MS Word document should be renamed with *.zip extension. Then the archive is being extracted and then application makes changes in MS Word content XML ﬁles. It will be used for development of application prototype.

140

K. Mackare et al.

5 Conclusion Technologies use increase. It can make some different problems, especially in vision and eye health, that affect people life quality. There is a need for the solution. Based on all information, research and survey, recommendations for different target groups were made. Well formatted e-materials could improve users reading comfort and vision health, that can be followed by educational and life quality increase. The concept for application prototype had been made, and the most appropriate technical solutions have been found. It will be used for development of application prototype. Last part is to develop a prototype of the application to analyses user and user needs, offer the best possible e-material formatting, and brings it to life. It is the next part of the research.

References 1. Mackare, K., Jansone, A.: Habits of using internet and digital devices in education. In: Proceeding of the International Scientiﬁc Conference, 25–26 May 2018, vol. 5. Society. Integration. Education (2018) 2. Khan, M., Khushdil, : Comprehensive study on the basis of eye blink, suggesting length of text line, considering typographical variables the way how to improve reading from computer screen. Adv. Internet of Things 3(1), 9–20 (2013) 3. Nielsen, J.: Designing Web Usability: The Practice of Simplicity, p. 420. New Riders Publishing, Indianapolis (2000) 4. Mackare, K., Jansone, A.: Research of guidelines for designing E-study materials. In: Proceedings of the 11th International Scientiﬁc and Practical Conference, 15–17 June 2017, vol. 2. Environment. Technology. Resources (2017) 5. Mackare, K., Jansone, A.: Recommended formatting parameters for E-study materials. In: 3rd International Conference on Lifelong Education and Leadership, ICLEL 2017, vol. 1. IJLEL (2018) 6. Zīverte, A.: Elektronisku teksta dokumentu pielāgošanas metodes, Master thesis, Liepaja University (2018)

The User Experience of 3D Scanning Tangible Cultural Heritage Artifacts Chee Weng Khong(&) and Muhammad Asyraf Mhd Pauzi Faculty of Creative Multimedia, Multimedia University, Jalan Multimedia, 63100 Cyberjaya, Malaysia {cwkhong,asyraf.pauzi}@mmu.edu.my

Abstract. Many culture and heritage (C&H) institutions in Malaysia, private and public alike, are aware of the potential beneﬁts to acquire and engage in the three-dimensional (3D) preservation of its tangible cultural assets. Many have heard about 3D scanning but few have had the opportunity to experience it. The acquisition and archival of digitized cultural assets provide unprecedented prospects to use and access cultural material. However, studies on the user experience (UX) for 3D scanning and its documentation activities are still few and far between. A successful 3D scanning of C&H assets is based upon the accomplishment of the entire scanning process. This paper presents the usercentered evaluation of the 3D scanning process involving the user, scanning device, the environment and the artifact. As a result, the study reveals the core UX issues and proposes user-centered recommendations to address them. Keywords: User experience 3D scanning White light scanner User-centered

Culture heritage

1 Introduction In the past two decades, three-dimensional (3D) scanners have become more portable and its use in culture and heritage (C&H) activities have been widely reported [1–3]. Most of the 3D scanning carried out for C&H activities around the world have focused on the preservation of historical artifacts and the built environment [4]. Digital preservation by way of using 3D scanning approaches have contributed positively towards the C&H domain [5, 6] including archaeological landscapes [7], rock art [8] and monuments [9]. These preservation activities have illustrated the immense value and potential of 3D scanning technology in heritage studies and its management [10]. A digital archive of C&H assets comprising of high quality 3D data is a great asset to heritage institutions and museums, especially for restoration activities and degradation monitoring [11]. Though the 3D scanning of C&H artifacts have been going on for a while now its user experience involving the scanning operator’s experience is largely unknown. One of the reasons may be due to the limited awareness, knowledge, studies and disclosure about the scanning process itself. This paper attempts to reveal the user experience of 3D scanning C&H artifacts by addressing the scope of the scanning process and the required engagement to accomplish the whole scanning cycle. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 141–147, 2019. https://doi.org/10.1007/978-3-030-02053-8_23

142

C. W. Khong and M. A. M. Pauzi

2 3D Scanning Artifacts The Malaysian National Heritage Act 2005 [12] pointed out that a historical artifact “means any artifact or object which holds a value in terms of religion, tradition, art and history including art crafts similar to sculpture and such.” Any acquired 3D scanned data of an artifact is essential to its preservation [5, 6]. The 3D scanned data literally holds actual data, which means that it provides the opportunity for viewing, measuring, reproducing by way of ‘reverse engineering’ damaged surfaces and textures on tangible artifacts, and for archiving. According to UNESCO (2017) [13] heritage is characterized into four categories comprising of tangible cultural heritage, intangible cultural heritage, underwater cultural heritage and natural heritage. This study considers the stages of 3D scanning of tangible, movable C&H artifacts. 2.1

3D Scanning

3D scanning consists of multiple technologies that are applied concurrently to measure the geometric properties of an object. Boehler and Marbs [14] deﬁnes a 3D scanner as “any device that collects 3D coordinates of a given region of an object’s surface automatically and in a systematic pattern at a high rate of achieving the results in near real time”. The acquired 3D data can be used to construct a 3D model for various purposes such as archaeology [15], manufacturing industry [16], fashion and clothing industry [17], healthcare [18], entertainment industry [19], food industry, and cultural heritage [20, 21]. The purpose of 3D scanning is to produce point cloud geometric data of the objects’ surface that is archived for later use, such as to construct a 3D model [22]. A 3D scanner is selected based on eight primary factors [14, 23]. They are scan speed, scan resolution, range limit, ﬁeld of view, surface registration, type of imaging camera, size and weight, and scanning software. Awareness about these factors is important as they affect the scanning process and the resulting scan quality. As 3D scanners slowly move towards mainstream application there is a tendency for ﬁrst time users to overlook these important factors. To date there is no one scanner that ﬁts all scanning purposes, and it is important to choose the right scanner for the job. 2.2

3D Scanning Process

Several scanning processes or workflows have been published [3, 24, 25]. A closer look at the studies highlight three similar stages comprising of pre-scanning, scanning and post-scanning. However, no in-depth description or considerations were disclosed. A recent study by Asyraf [23] identiﬁes a more holistic, integrated and updated list of processes involved. Figure 1 shows a list of user considerations throughout the 3D scanning process based upon. The pre-scanning stage usually involves planning and management activities prior to scanning. This includes site observation, timeline, costing, personnel, and any related background investigations to determine the type of scanning device needed. The second stage of the workflow involves the 3D documentation and capture (scanning) of the artifact. Depending on the physical dimensions, accessibility of the artifact or how complex the shape and form of the artifact is,

3D Scanning Tangible Cultural Heritage Artifacts

143

the operator will need to scan the artifact from various points of view and angle. The acquired raw digital data is only viewable in the scanning software and specialized third-party software. The post-scanning stage is labor and computing intensive as it involves cleaning up and repairing the captured 3D cloud points. The end process involves the creation of a suitable archiving format where completed ﬁles are exported to the desired format, such as .STL, .WRL, .OBJ, .AMF or .3MF. Stage/Level Pre-scan

User Consideration Access to artifact Surveillance Artifact condition Artifact material Artifact dimension Location Lighting External Factors (weather, temperature, humidity, etc)

Stage/Level Scan

User Consideration Set-up/Preparation Supporting equipment Scanner position Artifact position Operator position Scan pattern

Stage/Level Post-Scan

User Consideration Editing Alignment Global registration Fusion Texturing Exporting Archiving

Fig. 1. User considerations for scanning tangible movable C&H artifact.

3 Evaluation Method 3.1

Procedure

The activity of 3D scanning is very much a physical one where, at the core, it involves the operator (user), the scanning device, the environment and the artifact. The scanning experience (UX) revolves around the scanning process and the operator’s direct and indirect interaction with the scanning device, the environment and the artifact. Up to the point of this study there are no known local museums that have 3D scanning or that are involved in the archiving of 3D scanned data. However, there are units at local museums dealing with artifacts through conventional methods. A think-aloud approach [26, 27] was considered involving museum employees, familiarizing them with the scanner and issues while going through the stages of the scanning process, and providing ﬁnal remarks via a post-questionnaire. 3.2

Participants and Apparatus

This study involved eight (8) users (six men, two women, average age: 31.3), all are right handed and are employees from local museums actively involved in C&H artifacts. All participants requested strict anonymity, no pictures to be taken and not to disclose their afﬁliations, before agreeing to participate in this study. All participants have heard about 3D scanning but none have used a 3D scanner before (novice users). The scanning was conducted using an Artec MHT handheld structured-light 3D scanner together with a workstation laptop running on Intel i7 CPU, 16 GB RAM and onboard nVidia Quadro graphics card. The artifact, scanner, laptop and other peripherals required for scanning were loosely placed on a conference table under common ofﬁce lighting overhead (white fluorescent). Participants were seated during the session.

144

C. W. Khong and M. A. M. Pauzi

The measures obtained from the study includes UX issues identiﬁed during the think-aloud session over the three stages (pre-scanning, scanning, post-scanning), spontaneous verbal comments, and responses from the post-questionnaire. All UX issues were processed and coded, identifying common issues among users and matching issues into wider UX and usability principles.

4 Results The participants’ overall impression of the entire scanning experience, from beginning to the end of a scanning process, was positive with 37.5% ﬁnding it ‘very good’ while the rest found it ‘somewhat good’ at 62.5%. There were no negative responses. A total of 112 verbal comments were obtained from the whole session and were categorized into UX issues and other comments. We then analyzed each UX issue to identify uniqueness before mapping to design and usability principles to help produce eight categories in which the identiﬁed issues come under. The eight categories are accessibility, inspection, environment, physical setup, control, visualization, editing, and archiving. These are running categorizations to help generalize and make sense of the participants’ responses based on our experience in the scanning process. They are not any kind of a formal classiﬁcation. Accessibility relates to approaching, addressing and corresponding with the museum and related government departments. At times the acquisition, validation and documentation processes at the museum, prior to the actual viewing, access to or the release of a tangible movable artifact is required. Inspection is largely the visual and physical examination of the artifact to ascertain its surface quality, surface reflectance, type of material, texture, color, weight, complexity of its form or shape, and overall size. Environment is to determine the location for scanning, the ambient conditions, weather, lighting, clothing, humidity, wind, etc. The physical setup largely relate to the preparation of scanning equipment and the need for frames, supports, jigs and scaffolds with regards to the environment and artifact. Control and visualization is about the use, handling and execution of the 3D scanner and scanning software. Editing is time consuming and laborious, involving the clean-up and editing of the cloud points in the scanned data. Archiving is the ﬁnal step dealing with the creation, export and archiving of the 3D data.

5 Conclusion 3D scanning is a task-oriented, hands-on activity that requires planning and management of resources. To successfully scan an artifact in C&H means accomplishing all stages of the scanning process. This study identiﬁed the UX elements throughout the scanning process based on the think-aloud protocol and a questionnaire. There are particular UX issues that have been highlighted and we have identiﬁed the following user-centered recommendations:

3D Scanning Tangible Cultural Heritage Artifacts

145

• Accessibility: Ensure that users are aware of the various gatekeepers, stakeholders and paperwork needed to acquire the artifact. • Inspection: Provide a checklist of items relating to the artifact, environment and equipment to help users identify and consider for action. • Environment: Provide a user guide for in/outdoor scanning conditions to help users optimize and plan their tasks. • Physical setup: Provide users with an illustrated guide for setting up the common equipment and support props prior to scanning. • Control: Ensure users use both hands to hold the scanner and follow a 2–3 min scanner handling time to reduce fatigue. • Visualization: Increase the screen/monitor size of the scanning software by providing a larger monitor. • Editing: Provide a step-by-step guide to clean-up, align, register, fuse, patch and render the scanned data. • Archiving: Ensure users are aware of the stakeholder’s requirements and purpose for archiving the 3D scanned data to as to produce the correct 3D ﬁle format. Overall results from the study indicate that participants are excited and positive about 3D scanning, but it comes with its challenges. We believe this study provides a platform for further UX and usability considerations for the 3D scanning of C&H artifacts and its preservation.

References 1. Fontana, R., Greco, M., Materazzi, M., Pampaloni, E., Pezzati, L., Rocchini, C., Scopigno, R.: Three-dimensional modelling of statues: the Minerva of Arezzo. J. Cult. Herit. 3(4), 325– 331 (2002) 2. Li, J., Guo, Y., Zhu, J., Lin, X., Xin, Y., Duan, K., Tang, Q.: Large depth-of-view portable three-dimensional laser scanner and its segmental calibration for robot vision. Opt. Lasers Eng. 45(11), 1077–1087 (2007) 3. Cheng, H.M.: The workflows of 3D digitizing heritage monuments. In: Laser Scanner Technology. InTech (2012). https://doi.org/10.5772/32812 4. Wachowiak, M.J., Karas, B.V.: 3D scanning and replication for museum and cultural heritage applications. J. Am. Inst. Conserv. 48(2), 141–158 (2009) 5. De Reu, J., Plets, G., Verhoeven, G., De Smedt, P., Bats, M., Cherretté, B., Maeyer, W.D., Deconynck, J., Herremans, D., Laloo, P., De Clercq, W., Van Meirvenne, M.: Towards a three-dimensional cost-effective registration of the archaeological heritage. J. Archaeol. Sci. 40(2), 1108–1121 (2013) 6. Dellepiane, M., Dell’Unto, N., Callieri, M., Lindgren, S., Scopigno, R.: Archeological excavation monitoring using dense stereo matching techniques. J. Cult. Herit. 14(3), 201– 210 (2013) 7. Verhoeven, G., Taelman, D., Vermeulen, F.: Computer vision-based orthophoto mapping of complex archaeological sites: the ancient quarry of Pitaranha (Portugal-Spain). Archaeometry 54(6), 1114–1129 (2012)

146

C. W. Khong and M. A. M. Pauzi

8. Plets, G., Verhoeven, G., Cheremisin, D., Plets, R., Bourgeois, J., Stichelbaut, B., Gheyle, W., De Reu, J.: The deteriorating preservation of the Altai rock art: assessing threedimensional image-based modelling. J. Aust. Rock Art Res. Assoc. (AURA) 29(2), 139 (2012) 9. Koutsoudis, A., Vidmar, B., Ioannakis, G., Arnaoutoglou, F., Pavlidis, G., Chamzas, C.: Multi-image 3D reconstruction data evaluation. J. Cult. Herit. 15(1), 73–79 (2014) 10. De Reu, J., De Smedt, P., Herremans, D., Van Meirvenne, M., Laloo, P., De Clercq, W.: On introducing an image-based 3D reconstruction method in archaeological excavation practice. J. Archaeol. Sci. 41, 251–262 (2014) 11. Barone, S., Paoli, A., Razionale, A.V.: 3D reconstruction and restoration monitoring of sculptural artworks by a multi-sensor framework. Sensors 12(12), 16785–16801 (2012) 12. National Heritage Act 2005: Act 645 (Malaysia), The Commissioner of Law Revision Malaysia under the Authority of the Revision of Laws Act 1968, Percetakan Nasional Malaysia Berhad (2006) 13. United Nations Educational, Scientiﬁc and Cultural Organization (UNESCO): Deﬁnition of the cultural heritage. http://www.unesco.org/new/en/culture/themes/illicit-trafﬁcking-ofcultural-property/unesco-database-of-national-cultural-heritage-laws/frequently-askedquestions/deﬁnition-of-the-cultural-heritage/ 14. Boehler, W., Marbs, A.: 3D scanning instruments. In: Proceedings of the CIPA WG 6 International Workshop, Corfu, 1–2 September 2002, pp. 9–12 (2002) 15. Boehler, W., Vicent, M.B., Heinz, G., Marbs, A., Müller, H.: High quality scanning and modeling of monuments and artifacts. In: Proceedings of FIG Working Week, Athens, 22–27 May 2004 16. Wong, K.V., Hernandez, A.: A review of additive manufacturing. ISRN Mech. Eng. 2012, 10 (2012). https://doi.org/10.5402/2012/208760 17. Istook, C.L., Hwang, S.J.: 3D body scanning systems with application to the apparel industry. J. Fash. Mark. Manag. Int. J. 5(2), 120–132 (2001). https://doi.org/10.1108/ eum0000000007283 18. Rengier, F., Mehndiratta, A., von Tengg-Kobligk, H. et al.: Int. J. Comput. Assist. Radiol. Surg. 5, 335 (2010). https://doi.org/10.1007/s11548-010-0476-x 19. Apuzzo, N.D.: 3D Human Body Scanning Technologies Overview, Trends, Applications, Imagina, pp. 1–49, Monaco (2011) 20. Levoy, M.: Proceedings of the Second International Conference on 3-D Digital Imaging and Modeling, pp. 2–11 (1999) 21. Addison, A.C.: Safeguarding heritage’s endangered digital record. In: New Heritage: New Media and Cultural Heritage, p. 27 (2007) 22. Yu, F., Lu, Z., Luo, H., Wang, P.: Three-Dimensional Model Analysis and Processing. Springer, New York (2010) 23. Asyraf, M., Pauzi, P.: Digital Preservation of Malaysian Historical Artefact Using 3D Scanner: A Case Study of Mah Meri Mask (Master Thesis), Multimedia University, Siti Hasmah Digital Library, Malaysia (2018) 24. Li, R., Luo, T., Zha, H.: 3D digitization and its applications in cultural heritage. In: Digital Heritage, pp. 381–388 (2010) 25. Heritage, English. 3D Laser Scanning for Heritage: Advice and guidance to users on laser scanning in archaeology and architecture. English Heritage (2007)

3D Scanning Tangible Cultural Heritage Artifacts

147

26. Birns, J.H., Joffre, K.A., Leclerc, J.F., Paulsen, C.A.: Getting the whole picture: collecting usability data using two methods - concurrent think aloud and retrospective probing. In: Proceedings of UPA Conference, Orlando, 8–12 July 2002 27. Petrie, H., Precious, J.: Measuring user experience of websites: think aloud protocols and an emotion word prompt list. In: CHI 2010 Extended Abstracts on Human Factors in Computing Systems, Atlanta, 10–15 April 2010. https://doi.org/10.1145/1753846.1754037

International Museums and Transcultural Impact on Gulf States: The Louvre Abu Dhabi as a Case Study Mohamed El Amrousi, Mohamed Elhakeem(&), and Evan Paleologos Abu Dhabi University, Abu Dhabi, United Arab Emirates [email protected]

Abstract. The study of emerging contemporary museums in the United Arab Emirates suggests that new forms of architecture are emerging that deal with the past and present at the same time. The opening of the Louvre Abu Dhabi in 2017 and the number of visitors that visited the museum highlights the emergence of new spaces of gathering in Al-Saadiyat Island as a new cultural hub in Abu Dhabi. Traditional Museums in the United Arab Emirates are mostly heritage forts that have been restored and are conservative in their exhibitions. The Louvre Abu Dhabi on the other hand represents an experimental transdisciplinary design that includes continuity and discontinuity from tradition. This paper studies the Louvre Abu Dhabi, and we simulate the complex flow pattern around the Louvre using the 2D-hydrodynamic Finite Element Surface Water Modeling System. Keywords: The Louvre Abu Dhabi Cultural hub 2D-hydrodynamic surface water modeling

1 Introduction Abu Dhabi’s urban expansion plan as part of its 2030 Vision includes a gentriﬁcation plan for its shorefronts as well as fostering cultural spaces that cater to its multi-ethnic expatriate community. Much of the history of Abu Dhabi as exhibited in its traditional museums focuses on Bedouin culture, and crafts from the pre-petroleum era including ﬁshing and pearl diving. Such sanitized exhibitions exclude many social groups that have existed for centuries and represent a rich collective memory of the city. The recent blossoming of construction activities has provoked profound changes within the culture itself, affected by globalization provoking new forms of architecture that is none traditional yet considers the socio-cultural context. The Louvre Abu Dhabi by Jean Nouvel is built in Arabian (Persian) Gulf and surrounded by water. This strikes a contrast between its iconic domical structure, its interplay with light, shade and shadow and an irreversible relationship with Abu Dhabi’s shorefronts. This collision of contradictory currents represents an emerging pluralistic design that encompasses multiplicity under the penumbra of a broader spectrum of Islamic Arts. It also signiﬁes Abu Dhabi’s interest to become part of the global cities network through transcultural forms redesigned to create a new space of human gathering under the 180 m perforated steel © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 148–153, 2019. https://doi.org/10.1007/978-3-030-02053-8_24

International Museums and Transcultural Impact on Gulf States

149

dome [1]. The Louvre’s dominant design element – a single overarching dome creates a textured assemblage of abstract forms allows light to inﬁltrate a space of gathering beneath it. White cubical structures create a collage of forms and a pervasive network of pathways that end in vistas of Abu Dhabi’s waterfront (Fig. 1). This paper studies the Louvre Abu Dhabi design in relation to water and transcultural impact. We also simulated the complex water flow pattern around the Louvre using the 2Dhydrodynamic Finite Element Surface Water Modeling System.

Fig. 1. The Louvre Abu Dhabi.

2 The Louvre Abu Dhabi Design and Context The opening of the Louvre Abu Dhabi in 2017 and the number of visitors that visited the museum highlights the emergence of new spaces of gathering in Al-Saadiyat Island-the new cultural hub of Abu Dhabi. The Louvre Abu Dhabi reintroduces the geometry of Islamic Arabesques to a massive domical structure creating an important icon, and manifests the importance of redeﬁning traditional forms beyond the postmodern interpretation. The spaces of gathering created represent the pluralistic cultural understanding of spaces in the living city. This multi-ethnic composition is described in the ﬁctional novel Arabia [2]. The multi-cultural presence in Dubai and Abu Dhabi manifested through a strong presence of an Indo-Persiannate community that display their popular culture [2]. Today this is visible in Downtown Abu Dhabi and Diera in Dubai in the variety of South East Asian/Indian restaurants, merchandise and movie advertisements in the living city. In Contrast, traditional museums in the UAE such as Al-Fahidi Fort/Dubai National Museum reconstruct a past through an ethnographic curation of manikins that display life during the pre-petroleum era (Fig. 2). Representation of certain crafts as the past that no longer exists including pearl-diving, traditional dhows and exclusion of social groups creates melancholic exhibitions. In contrast, spaces of the Louvre Abu Dhabi can be viewed as an expression of pluralism that offers a new understanding of arts, architecture in relation to waterfronts. They also manifest the importance of Gulf cities as nodes of interaction between East and West. The Louvre Abu Dhabi displays artifacts from different cultures, not arranged in a chronological or ethnographic order rather they are displayed based on their artistic value and set against various architectural backdrops. Here the architecturally inverted space brings together cultural products of diverse communities, it transcends beyond the boundaries of time and space to encompass products, arts and artifacts (Fig. 3).

150

M. El Amrousi et al.

Fig. 2. Dubai National Museum – maritime exhibition.

Jean Nouvel’s spaces seek to offer new interpretations of culture through a regeneration of Islamic ornaments that juxtapose a contrast between artifacts, the interplay with light, shade and shadow created by the perforated dome and water that constantly interacts with art and architecture [3]. This collision of contradictory currents reflects the pluralistic design that encompasses multiplicity under the penumbra of a broader spectrum-the dome. The Louvre Abu Dhabi represents an experimental transdisciplinary design that includes continuity and discontinuity from tradition. Its form and relationship to water create a new context for the understanding of arts, architecture in relation to Abu Dhabi. This is further exempliﬁed through the interlacing screens of the dome that give reference to the ornately carved screens known as mashrabiyya, or jail screens that were used to veil private spaces [4]. These perforated devices used geometries, swirling foliage and arabesques to position the viewer to look out at the external world while retaining the discretion of space. The Louvre Abu Dhabi displays new forms that reinforce the conscious awareness of how Islamic architecture addressed the juncture between the interior space and the outside world. Such design is also an iconic referential to communal identity, reinforced through the variation of light patterns displayed on the walls and open space under the dome. The inclusivity and diversity of forms and fragments of Islamic art symbolized by the replication of a traditional Islamic medina under a perforated dome in the Louvre meditate on larger issues of the perception, utilization, and fabrication of visual forms [5]. The scale of the perforated dome also highlights the technologies that avail today and manifest the resources and emerging interest in culture and the exhibition of arts.

Fig. 3. The Louvre Abu Dhabi - arts and exhibition context.

International Museums and Transcultural Impact on Gulf States

151

3 Model Description Numerous, commercially available hydrodynamic models (either 1D, 2D, or 3D) exist for simulating flow around structures [6]. Flow around structures can not be easily simulated using 1D-hydrodynamic models. 2D-hydrodynamic models have been deemed more appropriate for simulating flow conditions around structures because they can provide spatially varied information regarding the flow around the structures [7]. In addition, 2D models require less data than 3D models for model calibration and veriﬁcation [7]. In this study, we chose the 2D Finite Element Surface Water Modelling System (FESWMS) to simulate the flow conditions around the Louvre Abu Dhabi museum. FESWMS is part of the commercially available Surface water Modelling System (SMS) software package (version 12.1) with a graphical interface that combines a series of hydrodynamic/sediment codes developed by the Federal Highway Administration [8]. FESWMS solves the differential forms of the continuity and the momentum equations in the stream wise and transverse directions using the Galerkin method of weighted residuals providing water depth and depth-averaged velocity magnitude in x and y directions at each node in the grid [8]. FESWMS inputs for model calibration are the Manning’s coefﬁcient of roughness, n, and the eddy viscosity, m. The Manning’s n is an empirical coefﬁcient that accounts for the total flow resistance from interactions with the boundary, while the eddy viscosity m accounts for flow resistance due to the internal shear stresses, or the Reynolds’ stresses of the fluid incorporating the added energy dissipation due to turbulence in the flow [6]. The eddy viscosity is not a physical property of the fluid, but rather reflective of the turbulent nature of the flow.

4 Application of the 2D Hydrodynamic Model to the Louvre Abu Dhabi The Louvre Abu Dhabi museum was built on the Arabian (Persian) Gulf and is surrounded with a number of offshore breakwaters to damp the wave energy around the museum and control foundations scour (Fig. 1). Due to the complex geometry of study region where the museum was built, the 2D hydrodynamic model FESWMS was used to investigate the flow pattern around the Louvre Abu Dhabi. The model inputs are: the region bathymetry, Manning’s coefﬁcient n = 0.025, eddy viscosity m = 0. 3 m2/s, flow rate around the breakwaters Q = 10 m3/s, water surface elevation at the downstream y = 4 m. Figure 4 shows the simulations of the water depth and the velocity vectors around the Louvre Abu Dhabi. It can be seen from the ﬁgure while complex flow circulation patterns developed around the museum, the water around the Louvre has almost a constant flow depth of 4 m. It should be pointed out here that these circulation patterns are important because they help in controlling sediment deposition, aquatic growth, and sulphide formation. Figure 5 shows the water velocity simulation, which is in the range of 0.03 to 0.2 m/s. Thus, the offshore breakwaters damped the wave energy effectively by reducing the velocity around the museum and hence, preventing foundations scour.

152

M. El Amrousi et al.

Fig. 4. Simulation of the flow depth and circulation pattern around the Louvre Abu Dhabi, UAE.

Fig. 5. Simulation of the flow velocity around the Louvre Abu Dhabi, UAE.

International Museums and Transcultural Impact on Gulf States

153

5 Conclusion The Louvre Abu Dhabi offers a fundamentally different spatial context and constructed spatial-social ambience, through an introduction of a pan-Islamic architectural vocabulary. In addition, we attempted so highlight the innovative design of the Louvre Abu Dhabi and its relation to Abu Dhabi’s shorefronts through a model that predicts the flow depth and velocity around the museum. The model predictions show that while complex flow circulation patterns developed around the museum, the water around the Louvre has almost a constant flow depth of 4 m. These circulation patterns are important because they help in controlling sediment deposition, aquatic growth, and sulphide formation. The water velocity around the museum was in the range of 0.03 to 0.2 m/s. Thus, the offshore breakwaters damped the wave energy effectively by reducing the velocity around the museum and hence, preventing foundations scour.

References 1. Oxford Business Group: Abu Dhabi 2015. Oxford Business Group, London (2015) 2. Raban, J.: Arabia Through the Looking Glass. Collins Harvill, London (1979) 3. Dalu, J.: Surface pattern and light. In: Michell, G. (ed.) Architecture of the Islamic World: Its history and Social Meaning, New York (1978) 4. Al-Asad, M.: Applications of geometry. In: Frishman, M., Khan, H. (eds.) The Mosque, History, Architectural Development and Regional Diversity, pp. 55–71. Thames and Hudson, London (1994) 5. Grabar, O.: The Formation of Islamic Art. Yale University, New Haven (1973) 6. Papanicolaou, A.N., Elhakeem, M., Krallis, G., Prakash, S., Edinger, J.: Sediment transport modeling review - current and future developments. J. Hydraulic Eng. 134(1), 1–14 (2008) 7. Elhakeem, M., Papanicolaou, A.N., Wilson, C.G.: Implementing streambank erosion control measures in meandering streams: design procedure enhanced with numerical modelling. Int. J. River Basin Manag. 15(3), 317–327 (2017) 8. Froehlich, D.: User’s Manual for FESWMS Flo2DH: Two-Dimensional Depth-Averaged Flow and Sediment Transport Model. Release 3 (2002)

Live Action Carnavalia: A Case Study of a Process for User Engagement Vladimir Barros(&) and Breno Carvalho(&) C.E.S.A.R School, Recife, Brazil [email protected], [email protected]

Abstract. The digital world is a reality for big companies looking through new alternatives to reach tanned and new users. The carnival is a celebration of Brazilian culture and motivated us to align the areas of interaction design, usability, and branding. The purpose of this article is to propose a new use of Facebook’s Live Action interaction tool, creating animated and interactive content that seeks to increase user engagement. In general, we use the inputs on interaction design of Norman [9], Saffer [3] and Rogers (2013), gestalt context with Gomes Filho [6] and games through Schell [8] and Salen & Zimmerman [2]. Through the case study of the game Carnavalia, it is approached how the process of game design and UX, applied to a mass event, can enhance the interaction of the tool bringing quantiﬁable and positive returns for product or brand exposure. Keywords: Live action Engagement

Interaction design Gamiﬁcation UX

1 Introduction According to the studies of some digital agencies of the Country, like the Digital Atratis1, the more tanners in a post, the greater it’s scope. This makes the numbers on a page interesting when it comes to defending the investment for companies looking for new markets, as the internet offers great brand visibility at low cost compared to the traditional or analog media investment model. Thinking about that, social networks like Facebook offer tools like live action, live posts, where the user can interact with other people, in real time, through pre-deﬁned reactions by the app. These are called reactions. As seen in a report on Nexo Jornal2, today, the Lives have six reactions (like, love, haha, wow, sad and grr) that limit the relationship of people within the tool, according to Bernardo [1]. Over time, people fail to follow or even dislike such content. It was from these observations, in lives actions, that we saw the need for an improvement in the tool to align some elements of user experience in a way that would potentiate 1

2

Site Atratis Digital. Available at: http://atratis.com.br/blog/alcance-das-publicacoes-no-facebookcomo-funciona-cada-uma-das-acoes. Accessed on 19 Mai 2018. 2 Site Nexo Jornal. Available at: https://www.nexojornal.com.br/expresso/2016/02/24/%E2%80% 98Rea%C3%A7%C3%B5es%E2%80%99-v%C3%A3o-se-somar-ao-%E2%80%98curtir%E2%80% 99.-Por-que-o-Facebook-mudou. Accessed on 22 Mai 2018.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 154–159, 2019. https://doi.org/10.1007/978-3-030-02053-8_25

Live Action Carnavalia

155

interactions. So comes the Carnavalia, digital project that uses Facebook’s live action with a strategic game design, making use of user interactions to gain empathy and engagement. Published in the fan page of the Jornal do Commercio System (SJCC), through a series of seven animated live animated actions in the pre-carnival period between January and February 2018.

2 Development For the development of the application, we follow the authors Salen and Zimmerman [2], therefore they constituted a description, that presents/displays a relation of the design, playfulness, and culture based: Rules, giving a logical structure to the game; Interaction, which structures the experimental, social and representational design in the game and with other players; and, Culture, means that contextualizes the game, projecting them and executing them. Another theoretical contribution to the creation of Carnavalia was the study of the design centred on the Saffer activity [3], focusing on observing the behavior around the basic tasks (like choosing the emoticon and interacting with other netizens). In it, users are still very important, however, they are their behaviors that are most essential to making the best decision. Another point was the use of gamiﬁcation to promote user participation during lives and achieve interactions. According to Deterding [4], gamiﬁcation is the addition of game elements in different contexts, such as health, education, to improve user experience and engagement. Still, on the digital engagement, we follow the mental script AIDA de Siqueira [5], to get the attention of the user, delivering interesting content, showing some beneﬁt that the character would have if he got his vote. In this way, the user would take the action for himself, enjoyed and shared the material. 2.1

User Analysis

As a main strategic part and following Saffer’s approaches [3], studying the user was the ﬁrst step in the project. In mid-April, 2017, analyzing how the public behaved within the Fan page of the Jornal do Commercio System, it was simpler to draw behavioral parameters that would help in the elaboration of the Carnavalia project. We apply the precepts of Saffer [3], in the post Sport x Santa (prototype of the ﬁrst live action), broadcast on April 28, 2017. It was begun to observe that the relations of the people inside the network, when there is some situation of the ﬁght, take place for a matter of representativeness. This prototype facilitated a lot in the creation of Carnavalia, allowing to reassess and to alter structures of the project. 2.2

Visual Empathy

In this aspect, we use the Gestalt laws of João Gomes Filho [6]: unity, segregation, uniﬁcation, and closure, basic precepts for design. Following the same reference, we use the softness and softness of geometric pieces with rounded edges, transmitting

156

V. Barros and B. Carvalho

fluidity to the organic forms, facilitating the visual reading without breaks and giving the idea of graphic continuity. The palette of colors counts on tones that were a tendency in the summer of the time: the purple and the yellow, seen in several reports like in the site of the Elo7, published by Oliveira [7]. The warm tones are predominant by the afﬁnity of the color with the locality, always worked in a gradation of less vibrant tones so as not to tire the vision of the user. 2.3

Feedback

From the book by Jesse Schell [8], we have seen that the quality of feedback can exert a great influence on how much the player understands and likes his game. Therefore, we apply screen transition feedbacks at the beginning, middle, and end of the animated live. Based on Norman’s contributions [9], we align the project with the principle of visibility, making the elements visible to ensure that the execution and evaluation gaps are objective. Therefore, we inserted a timer regressive during the live action, to cause apprehension in the public, which always waits for something at the end of that deﬁned time. We present the visually animated reactions to be easily identiﬁed and reinforced with a splash displayed on the screen each period of time (Fig. 1B).

Fig. 1. A. Screen with the presentation of the duel. B. Splash with reaction. C. Truck excited exposing brand during the carnavalia. D. Dialog screen. E. Winner/loser indicative screen. F. Winner screen.

The progressive counting of the reaction (Fig. 1D), released from 10 likes, the dance steps of the characters. This micro-interaction occurred through a colored button presented in three spaces in the indicator of the personage (Fig. 1D). Another question observed analysing the representativeness that each character had, was the possible direct contact of the public with his chosen one. This connection became even greater when we created chat boxes between the characters and the user, that, during the game, established simple messages usually in thanks of the vote.

Live Action Carnavalia

2.4

157

Dynamics of Motion Design

The motion design technique was worked out with the laws of Gestalt [6] to make the user the driving force in the campaign’s success. The dynamics of each character’s movements were related to pop culture since the whole process had an empathy and a different visual appeal. Classical animes such as Knights of the Zodiac (1986), Dragon Ball (1989), One Piece (1999), Chirriro (2003) and Naruto (2007) brought some inspirations, added to Brazilian carnival ﬁgures. According to Schell [8], we also prioritize user involvement, managing expectations through feedback, always trying to create a sense of ownership. This degree of empathy worked contributed signiﬁcantly to the project since it is through public engagement that the project has grown and achieved signiﬁcant results.

3 Carnavalia: The Interaction with the User For the online project, we chose to run a Carnavalia video stream using the OBS Studio tool, because Facebook would only mirror the image that would run inside our computer, programmed in JavaScript. Broadly speaking, we developed an animated live action that turned in a game and the user a player from the moment they chose a side to give their vote and convince other users to vote for their favorite character. After a certain amount of votes, they would unlock the steps that would leave the character closest to victory. These moves were limited to three for each character. In the initial screen, the SJCC Carnival campaign is presented. In the sequence, the screen of the opponents’ clash was presented, with the ﬁrst application of sponsoring brands, in a period of time sufﬁcient so that it did not bring an aversion to the user (Fig. 1A). In the screen of the duel (Fig. 1B), the characters placed in the corners of the scene, as well as the other elements: name, respective reaction, steps and the accounting of the votes. Also shown is the timer, set to one hour, timeout for engagement without possible aversions. Throughout the duel, the vote counts up to the number set by the planning to unlock the coup. It runs and feedback appears on the character’s informational tab. Another important feedback is the question of the dialog with the user (Fig. 1D). During the match, there are also small insertions of brands that pass quickly during the duel. They appeared in animations with trucks and drones (Fig. 1C) in a secondary way, but perceived in the general context. Thus it goes until we have the ﬁnal result, the winner celebrates with euphoric gestures and the loser follows with a sad aspect (Fig. 1E). The winner is set manually, who has the most votes, moves to the next stage.

158

V. Barros and B. Carvalho

4 Validation We used the Facebook Analytics tool to visualize the number of interactions and reach of the project. We achieved in the seven duels, with one hour each, a result of 414,936 impressions, 305,351 people reached and 2,474 interactions. After obtaining the results, we performed a comparison with a monthly average per post of the page numbers of the SJCC for three months. From January to March (precarnival, carnival, and post-carnival months) we captured the numbers of these three parameters to evaluate the success of the tool. Then we did it (Table 1).

Table 1. Comparison of numbers per post month and Carnavalia

A Compared to averages of whole months, Carnavalia was superior in the three quantitative parameters that we used as the basis for the defense of the project, and in some, more than doubled its value in numerical terms. We conclude that the numbers show that the project has achieved its goals, both in terms of engagement and the new proposal to use an animated live and gamiﬁed.

5 Conclusion We conclude that Carnavalia has brought a new alternative within a limited tool such as Facebook’s live action. This view has shown that a campaign’s success may be tied to the way it is presented and empathized with certain products. Understanding the user and presenting new structures that make their journey in the digital age more pleasant is an important point to understand other relationships, including commercial ones. We have achieved with Carnavalia to create and validate an environment for the insertion of the user more intensely, giving him a greater opportunity for control. The numbers reached by the project indicate that through studies and structured planning, we were able to create a campaign using design tools, where the social network was the channel to reach a number of people that conventional strategies do not reach.

Live Action Carnavalia

159

In this way, we combine great tools like UX, game design and culture to reach the most important target of the digital age: the user. The analysis of the project gave us new perspectives, including new plans and improvements for the tool. New studies will be developed to further engage the public since we can increase these numbers.

References 1. Bernardo, K.: ‘Reações’ vão se somar ao ‘curtir’. Por que o Facebook mudou. Site Nexo Jornal, Rio de Janeiro (2018). https://www.nexojornal.com.br/expresso/2016/02/24/%E2% 80%98Rea%C3%A7%C3%B5es%E2%80%99-v%C3%A3o-se-somar-ao%E2%80%98curtir %E2%80%99.-Por-que-o-Facebook-mudou. Accessed 22 May 2018 2. Salen, K., Zimmerman, E.: Rules of Play: Game Design Fundamentals. MIT Press, Cambridge (2003) 3. Saffer, D.: Designing for Interaction: Creating Innovative Applications and Devices. New Riders, San Francisco (2009) 4. Deterding, S., Dixon, D., Khaled, R., Nacke, L.E.: Gamiﬁcation: toward a deﬁnition. In: CHI 2011 Conference on Human Factors in Computing Systems, Vancouver, 7–12 May 2011. http://gamiﬁcation-research.org/wp-content/uploads/2011/04/02-Deterding-Khaled-NackeDixon.pdf . Accessed 22 May 2018 5. Siqueira, A.: AIDA (atenção, interesse, desejo e ação): entenda as 4 etapas do conceito. Site Resultados Digitais. Santa Catarina (2018). https://resultadosdigitais.com.br/blog/aida/. Accessed 25 Jan 2018 6. Gomes Filho, J.: Gestalt do objeto: sistema de leitura visual. Escrituras Editora, São Paulo (2009) 7. Oliveira, C.: Cores em alta para 2018: veja as principais tendências. Site Elo7, São Paulo (2018). https://blog.elo7.com.br/moda/cores-em-alta-para-2018.html. Accessed 25 Jan 2018 8. Schell, J.: The Art of Game Design: A Book of Lenses. Morgan Kaufmann Publishers, Burlington (2008) 9. Norman, D.A.: O design do dia-a-dia. Rocco, Rio de Janeiro (2016)

Incorporating Human Factors in In-Plant Milk Run System Planning Models Aleksandra Polak-Sopinska(&) Faculty of Management and Production Engineering, Lodz University of Technology, Wolczanska 215, 90-924 Lodz, Poland [email protected]

Abstract. The supply of materials to workstations in a manufacturing system can be realized in many different ways. In the last few years, the milk run based in-plant supply has been widespread. Researchers have developed many models for planning activities related to milk run systems and for enhancing the efﬁciency of such systems. The proposed models for planning milk run activities largely ignore workers’ characteristics or human factors, which leads to only partially realistic results. This paper contributes to the existing literature on milk run system planning models by literature review on integrating the human factor into milk run system planning and scheduling models and presenting the results of studies on physical intensity of an milk run operator’s work for selected milk run concepts. Keywords: Human factors Physical intensity of work Milk run system Logistics Efﬁciency

Energy expenditure

1 Introduction The just-in-time and just-in-sequence feeding of manufacturing machines and assembly stations is getting more and more important in production. The supply of materials to workstations in a production system can be realized in many different ways. In the last few years, the milk run (MR) based in-plant supply has been widespread [1]. In-plant MR is an alternative to the most common method of indoor transportation that is the forklift. The idea behind an in-plant MR system is to supply various goods form various locations in one run using a train consisting of a tugger and several carts. Usually, an MR operator works on a ﬁxed route and ﬁxed schedule shuttling between workstations in the production hall. The main tasks of an operator are to drive the train, load and unload materials [2, 3]. An MR operator’s job is labour- and time-intensive. The efﬁciency of an MR system depends to a large extent on the worker’s characteristics, warehouse/production line equipment design, location of facilities, characteristics of the trailers, routing of MR vehicles, scheduling and assignment of routes to technological resources, problems with queues [1]. Based on the literature review [1–4] and the author’s knowledge, it can be concluded that researchers have developed many models for planning activities related to MR systems and for enhancing the efﬁciency of such systems by recommending various warehouse layouts and types, different buffering and other objects that © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 160–166, 2019. https://doi.org/10.1007/978-3-030-02053-8_26

Incorporating Human Factors in In-Plant MR System Planning Models

161

make up an MR system, and different ways to route MR vehicles. The proposed models for planning MR activities largely ignore workers’ characteristics or human factors suggesting that they cannot be substantiated, which leads to only partially realistic results. Perhaps this is because ergonomic conditions may be difﬁcult to quantify [5]. However, it indicates that there is a discrepancy between the situations described in the literature and the situations observed in practice, which clearly suggests that there is a need to include the human factor into planning MR models. This paper contributes to the existing literature on MR system planning models by literature review on integrating the human factor into MR system planning and scheduling models and presenting the results of studies on physical intensity of an MR operator’s work for three MR concepts.

2 Literature Review This literature review provides a snapshot of a diversity of practices discussed in publications on including the human factor in in-plant MR planning models. It does not pretend to cover the entirety of the literature but rather offers an informed evaluation of purposefully selected literature with regard to MR, human factors and EE. The literature on the incorporation of human factors in the design of in-plant MR systems is generally scarce. A majority of studies only discuss part of the process carried out by an MR operator, namely, order picking in the warehouse. A seminal publication in this area is the article by Grosse et al. [6]. The researchers propose a conceptual framework for integrating human factors into the planning of models of order picking activities, and hypothesize that doing so improves the performance of an order picking system and workers’ well-being. The article can be useful in designing work conditions including MR route optimization, however, only in warehouses and supermarkets. Battini et al. [7] may also be referred to when trying to select an optimal MR route for an MR operator in a warehouse. This paper provides a model and an analysis of integration of human EE as an ergonomic parameter into the storage assignment problem using a bi-objective approach that considers both total order picking time and human EE. Garg et al. [8] proposed the metabolic rate prediction model which can be used to estimate metabolic rates of a wide variety of manual material handling jobs. The model is based on the assumption that a job can be divided into simple tasks and that the average metabolic EE rate of the job can be predicted by knowing the EE of the simple tasks and the duration of the job. The researchers proposed 27 equations to calculate the net metabolic cost of manual material handling tasks. As for the planning of an entire process of a MR operator that would take EE into account, no relevant literature has been found except the articles by Polak-Sopinska A. In [9], guidelines for the assessment of physical work intensity of an MR operator were presented. In [10], a case study was described. Droste et al. [11] in their approach to the planning of an in-plant MR system with regard to time (of material provision), proposed to use - MTM Logistics, whereas with regard to ergonomics (of milk-run operator) - Multiple Loads Tool (MLT otherwise known as MultipLa). Unfortunately, the researchers only presented a model approach to the planning of an in-plant MR system while failing to report any case studies or to

162

A. Polak-Sopinska

provide speciﬁc guidelines. No deﬁnite guidelines or procedures concerning the incorporation of the human factor in the planning of MR systems are given in the VDI 5586 standard, either. Physical ability of the worker is only to a limited extent considered in MR system planning software, e.g. RoutMan. Physical capabilities of a human were taken into account in the ergonomic evaluation of basic tugger train concepts. The results of the research were presented in [12]. The ergonomic evaluation was based on the measurement of the three orthogonal components of the acting force which needs to be applied by a tugger train operator. The above brief literature review indicates a need for further, in-depth research in this area in the future.

3 Milk Run Classiﬁcation Criteria and Concepts There are different concepts of MR, and therefore, it can be classiﬁed according to various criteria. If the general condition is taken under consideration, the following criteria can be applied [2, 3]: material source; handling unit; replenishment principle. If the organizational structure is considered, criteria such as [2]: a route, the assignment of a vehicle to a route, MR control principle, integration of loading process, integration of empty bins process can be used. According to [2], there are no fewer than 21 MR concepts. No two concepts are exactly the same. In this paper, only three concepts are taken into account. In the ﬁrst concept, an MR operator, according to the orders, manually loads small load carriers (SLCs) (bins, cartons) and smaller trolleys prepared in the manufacturing supermarket onto the MR train (SLC trolleys and cargo-liner trolleys), and, having done that, following a ﬁxed route, he drives the train to production workstations. While stopping at the stations, the operator manually unloads SLCs and smaller trolleys and collects empty bins and trolleys. At the end of each tour, he unloads the empty bins and trolleys. He operates according to a schedule while orders are generated using Kanban. Driving time takes up a small percentage of the work shift. The second concept differs from the ﬁrst one in that the MR operator manually loads and unloads both SLCs and large load carriers (LLCs) (pallets, roll containers, boxes) and the TT concept is a combination of trolley and taxi concepts. The third concept is similar to concept 2 but the operator loads and unloads only LLCs, and he loads both manually and using a forklift at the source.

4 Empirical Research into In-Plant Milk Run Operator’s Work The study was carried out at three assembly plants operating in Poland. In each company, the EE for 12 randomly selected MR routes for each of the three MR concepts was measured. The study was performed in the years 2015–2018. The EE for the tasks of the studied jobs were determined with the indirect calorimetry method based on pulmonary ventilation measurements [13]. The measurements were carried out for three workers (males) on each MR route. The pulmonary

Incorporating Human Factors in In-Plant MR System Planning Models

163

ventilation measurements were carried out with an EE meter (MWE-1) which automatically translates the recorded volume of ventilation into EE [13]. For the remaining activities, the EE was assessed based on the Lehmann’s estimation method [14]. Based on the EE per min. while performing basic job tasks and on their duration, the EE for the entire shift was calculated At all studied plants, one shift spanned 8 h. The details of the EE assessment and analysis are explained in [9] and are not repeated here. As for the ﬁrst concept, the tugger train comprised 5 trailers in each studied case. The SLCs weight ranged from 0.5 to 16.4 kg, the weight of small loaded trolleys from 45 to 80 kg, whereas of empty ones from 20 to 30 kg. It is difﬁcult to determine the total mass of loads transported during one cycle because not all materials were supplied in each cycle. To calculate the total mass of carried or pushed loads per shift per operator, the mass of each load was multiplied by the number of picks and the products were then added up. During each shift, the operators manually handled from 3098 to 12102 kg, 7345 kg on average. Each time, they carried the load over distances ranging from 0.5 to 5 m. They made an average of 15 stops (min. 9, max. 22). The duration of each cycle ranged from 30 to 40 min. The cycle was repeated from 11 to 15 per shift. The average energy expenditure per shift was 8605 kJ (standard deviation (SD) 422 kJ, min. 7826 kJ, max. 9502 kJ), which for males means very heavy work. In 23 cases, the work was very heavy above 8400 kJ/shift. The energy expenditure for manual handling tasks averaged at 7884 kJ (SD 425 kJ, min. 7034 kJ, max. 8646 kJ). In 9 cases, it exceeded the rate of 8400 kJ/shift allowable by Polish law. The tasks that required 20–30 kJ/min (heavy work per minute) always took longer than one half of the shift, 284 min. on average (min. 254 min., max. 340 min.). In 11 cases, the rate of 30 kJ/min for casual labor allowable by Polish law was exceeded. As regards the second concept, the tugger train system included from 4 to 5 trailers. The SLCs weighed from 1.8 to 18.5 kg, whereas the LLCs from 120 to 320 kg. During one shift, the operators manually handled from 33825 to 63090 kg, on average 53450 kg (they pushed from 27936 to 59186 kg and carried from 2680 to 5890 kg). Each time, they moved the load over a distance ranging from 1 to 6 m. They made an average of 4 stops (min. 3, max. 5). Each cycle spanned 15 to 25 min. The cycle was repeated from 15 to 30 times per shift. The average energy expenditure per shift equaled 8090 kJ (SD 197 kJ, min. 7842 kJ, max. 8450 kJ), which means heavy work for males. In 3 cases, the work was very heavy. The energy expenditure for manual handling tasks was on average 6965 (SD 189 kJ, min. 6720 kJ, max 7400 kJ) In none of the cases were the Polish legal norms exceeded. The time spent on tasks that required the EE of 20– 30 kJ/min equaled on average 248 min. (min. 238 min., max. 305 min.). As for the last concept, the tugger train was comprised of 5 or 6 trailers. The LLCs weighed from 122 to 450 kg. While the characteristics of workstations were being written up, it was discovered that for 13 routes, the legal limits concerning the allowable mass of carts with loads were exceeded. The EE measurements were performed only when the required changes had been implemented. The operators manually pushed from 44289 kg to 112320 kg per shift, an average of 81300 kg. Each time, they covered a distance ranging from 3 to 7 m. They made an average of 5 stops (min. 4, max. 6). The duration of each cycle ranged from 15 to 25 min. The cycle was repeated from 18 to 30 times per shift. The average EE per shift was 7467 k (SD 532 kJ, min. 6543 kJ, max. 8235 kJ), which for males constitutes heavy work. The EE for manual handling tasks

164

A. Polak-Sopinska

was on average 4734 kJ (SD 499 kJ, min. 4113 kJ, max. 5870 kJ). In none of the cases were the Polish legal norms exceeded. The time spent on tasks that required the EE of 20–30 kJ was on average 158, min. 75 min., max. 204 min. In all of the cases, the employer should provide wholesome beverages for the workers, and for 23 operators working according to the ﬁrst concept and 3 workers working according to the second concept, also refreshments. The operators should not be required to work overtime and should be provided with additional rest breaks depending on the rate of energy expenditure per shift (from 10 additional minutes up to as much as 20% of the total working time [9]). In all of the studied cases, the employer should require that workers aged 45+ undergo periodic medical examination every three years. Concept 1 was the heaviest in terms of the EE and it should be avoid by logistics planners. Concept 3 proved to be the lightest, which stemmed from the fact that the worker spent part of the shift operating a forklift. Nevertheless, it bears emphasizing that the pace of work was so big, that even during the forklift operation the work was moderately heavy for the worker. Therefore, the worker did not have the opportunity to recover while doing lighter tasks after heavy work. Summing up, in none of the cases, regardless of the studied concept, no work was moderately heavy or light per shift. In 26 cases out of 108 of the studied ones (24%), the work was very heavy and should not be performed by older workers, whereas in 11 cases (for the concept 1), it should not be performed by any male. The presented results of the study prove that in order to ensure a sustainable high level of efﬁciency and productivity, and to make certain that decision-making support models represent real world situations as accurately as possible, it is indispensable that human factors be incorporated into the design of work-intensive systems of in-plant MR.

5 Discussion In the course of the study, it was observed that few in-house logistics managers understand the need for integrating the human factor into the planning of MR systems. Investment to improve ergonomic conditions in logistics processes is, in general, frequently of secondary importance because many managers still hold the belief that it does not create added value. The situation could be amended if the correlation between the human factor and MR systems was studied further with regard to its impact on the level of fatigue, productivity, performance, number of errors, absenteeism, number of accidents and collisions, mature 45+ workers leaving work, etc. A lot can also be done by improving measurement and assessment methods for fatigue management systems [15]. EE measurement methods and equipment should be more intuitive and cheaper to implement; results should be easier to read, and their applied value should be easier to understand. Were these conditions met, logistics planners would be more likely to use them. Further research on incorporating the human factor into MR system planning, including the studies that have been conducted so far concerning the human factor in order picking planning models, could provide an excellent basis for developing a comprehensive approach to the design and planning of MR systems. In parallel, regulatory initiatives to expand legislation enforcing work safety in logistics should be

Incorporating Human Factors in In-Plant MR System Planning Models

165

undertaken [6]. For example, in 2017, important changes were introduced into Polish regulations concerning allowed values of EE, manual handling load weight, and the required force that a worker needs to exert to initiate object movement. Acknowledgments. The described research was carried by the research group IDEAT Industrial Diagnosis & Ergonomic Accessibility for Technology Excellence.

References 1. Bányai, T., Telek, P., Landschützer, C.: Milkrun based in-plant supply – an automotive approach. In: Jármai, K., Bolló, B. (eds.) Vehicle and Automotive Engineering 2. VAE 2018. Lecture Notes in Mechanical Engineering, pp. 170–185. Springer, Cham (2018) 2. Klenk, E., Galka, S.: Analysis of parameters influencing in-plant milk run design for production supply. In: 12th International Material Handling Research Colloquium. Technishe Universitat Munich, pp. 25–28 (2012) 3. Mácsay, V., Bányai, T.: Toyota production system in milkrun based in-plant supply. J. Prod. Eng. 20(1), 141–146 (2017) 4. Korytkowski, P., Karkoszka, R.: Simulation-based efﬁciency analysis of an in-plant milk-run operator under disturbances. Int. J. Adv. Manuf. Technol. 82, 827–837 (2016) 5. Rouwenhorst, B., Reuter, B., Stockrahm, V., Van Houtum, G.J., Mantel, R.J., Zijm, W.H.M: Warehouse design and control: framework and literature review. EJOR 122(3), 515–533 (2000) 6. Grosse, E.H., Glock, C.H., Jaber, M.J., Neumann, P.W.: Incorporating human factors in order picking planning models: framework and research opportunities. Int. J. Prod. Res. 53 (3), 695–717 (2015) 7. Battini, D., Glock, C.H., Grosse, E., Persona, A., Sgarbossa, F.: Human energy expenditure in order picking storage assignment: a bi-objective method. J. Comp. Ind. Eng. 94, 147–157 (2016) 8. Garg, A., Chafﬁn, D.B., Herrin, G.D.: Prediction of metabolic rates for manual materials handling jobs. Am. Ind. Hyg. Assoc. J. 39(8), 661–674 (1978) 9. Polak-Sopinska, A., Wrobel-Lachowska, M., Wisniewski, Z., Jalmuzna, I.: Physical work intensity of in-plant milk run operator. Part I - guidelines for assessment. In: Karwowski, W., Trzcielinski, S., Mrugalska, B., Di Nicolantonio, M., Rossi, E. (eds.) Advances in Manufacturing, Production Management and Process Control. AHFE 2018. Advances in Intelligent Systems and Computing, vol. 793, pp. 66–76. Springer, Cham (2019) 10. Polak-Sopinska, A.: Physical work intensity of in-plant milk run operator. Part II – case study. In: Karwowski, W., Trzcielinski, S., Mrugalska, B., Di Nicolantonio, M., Rossi, E. (eds.) Advances in Manufacturing, Production Management and Process Control. AHFE 2018. Advances in Intelligent Systems and Computing, vol. 793. Springer, Cham (2019) 11. Droste, M., Deuse, J.: A: planning approach for in-plant milk run processes to optimize material provision in assembly systems. In: ElMaraghy, H. (ed.) Enabling Manufacturing Competitiveness and Economic Sustainability. Springer, Heidelberg (2012) 12. Keuntje, C., Kelterborn, M., Günthner, W.A.: Considering ergonomics in the planning of tugger train systems for production supply. In: MATEC Web of Conferences, vol. 95 (2017)

166

A. Polak-Sopinska

13. Konarska, M., Kurkus-Rozowska, B., Krokosz, A., Furmanik, M.: Application of pulmonary ventilation measurements to assess energy expenditure during manual and massive muscular work. In: Proceedings of the 12th Congress of IEA, Toronto, Human Factor Association of Canada, pp. 316–317 (1994) 14. Lehmann, G.: Praktische Arbeitsphysiologie (Practical Work Physiology). Georg ThiemeVerlag, Stuttgart (1953) 15. Gajšek, B., Đukić, G., Opetuk, T., Cajner, H.: Human in manual order picking systems. In: Improving Working Conditions in Occupations with Multiple Disadvantages. Ofﬁce of the European Union, Luxembourg (2015)

Design and Evaluation of an Innovative Assisting Device for Improving Blood Circulation in Osteoarthritis Yan-Chun Lin(&) and Fong-Gong Wu Department of Industrial Design, National Cheng Kung University, No. 1, University Road, Tainan City 701, Taiwan [email protected], [email protected]

Abstract. In this study, an innovative assisting device is proposed to respond to Osteoarthritis (OA) prevention needs of aging society. Due to the modern lifestyle, people lack exercise and obesity problem. The risk of Osteoarthritis is increase recently. OA usually occurs in weight-bearing joints. Joint pain often worsens when more active and eases when resting; although as the disease progresses, may ﬁnd in pain most or even all of the time. Resolving pain and stiffness brings inconvenience to life and deterioration of body functions to OA patients is essential. This study explores in depth the impact of the physiological cycle of knee osteoarthritis on degeneration. Design a useful adjuvant therapy device, which can effectively improve the regulation of peripheral nerves and blood vessels, thereby increasing the mobility of elderly. Prevention in OA positively impacts health cost and elderly’s quality of life in old age. Keywords: Human factors Assisting device Design

Osteoarthritis prevention Elderly daily life

1 Introduction 1.1

Background

The lifestyle of modern people leads many degenerative diseases to occur in younger age. Osteoarthritis (OA) is the most common form of the degenerative joint disease and is a leading cause of disability in people over 50 years of age [1]. OA is characterized by the breakdown of cartilage in joints. Over time the articular cartilage can thin or form cracks. Pieces of cartilage may come loose and float inside the knee, further irritating the joint. After an extended period, the cartilage can become worn entirely away, and the bones begin to rub together [2]. OA usually ﬁnds in which joint is overuse and weight-bearing, such as hip, knee and ﬁngers. The primary symptoms include joint pain, stiffness and limitation of movement. Disease progression is usually slow but can ultimately lead to joint failure with pain and disability. Chronic pain may seriously affect health-related quality of life (HRQOL) and is regularly associated with secondary comorbidities such as anxiety, depression and sleep disturbance [3]. Radiographic evidence of knee osteoarthritis is present in approximately 30% of men and women over the age of 65. Approximately 80% of © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 167–172, 2019. https://doi.org/10.1007/978-3-030-02053-8_27

168

Y.-C. Lin and F.-G. Wu

those with OA will have limitations in movement, and 25% cannot perform their signiﬁcant activities of daily life [4]. At ﬁrst, OA occurred in the cartilage disintegration. Cartilage tissue without blood vessels, only by the surrounding blood vessels and joint synovial fluid to provide nutrients, so blood circulation is essential for the protection of cartilage [5]. Resulting in decreased bone density and synovium membrane cells cannot normally secrete synovial fluid, the decline in estrogen in women also indirectly leading to earlier degradation than men. Joint movement requires adequate synovial fluid to provide lubrication to have different angles of action. The reduction of synovial fluid also causes reduction of the joint capsule, thus lead to cartilage wear and tear serious. The above studies have shown that blood circulation has a signiﬁcant impact on patients with knee osteoarthritis. However, the existing aids, mainly focus on the mechanics to assist OA patients in reducing pain and walking better. Enhance the circulation of blood not only can increase synovial fluid production also can stimulate the cycle of bone metabolism in the acceleration. Stimulation of the peripheral nerves in different parts of the leg to increase the blood circulation of the knee, to ﬁnd more effective parts. Then design an innovative assisting device for that site to prevent knee osteoarthritis from occurring before mild symptoms occur. The device is to develop a prevention method to improve joint activity, considering the factors of human, motor patterns. 1.2

Purpose

Due to the aging process and improper use of the body, the tear and wear of cartilage cause the bones to start touching and rubbing to cause pain and inflammation. Maintaining the health of the joint cartilage before it has worn out and repairing the injured cartilage in time is the only way to avoid OA. When the synovial fluid in the joint capsule is not enough could cause the joint stiffness also painful [6]. Therefore, this study aims to explore the effects of human blood circulation and neuro stimulation on Osteoarthritis, consider the causes of human factors and improve blood circulation for OA. Expecting to design the innovative assisting device to slow down and prevent pain and lower body functions caused by OA. The purpose of the study divided into the following four points: (1) To explore the impact of OA patient in daily life activities, by semi-structured interview and questionnaire survey. (2) To explore stimulate which part of lower limb useful in OA. (3) To analyze the existing assisting devices classify the functions and stimulate ways, ﬁnd the lack of assisting devices. (4) Using Focus groups to design a new type of assisting device to improve the quality of life in OA patients (Fig. 1).

Design and Evaluation of an Innovative Assisting Device

169

Fig. 1. Research flow chart

2 Literature Review Osteoarthritis is a common problem in middle-aged and older people over the age of 50. Whether it is because of active sports or work injuries or an accidental fall, it is easy to let such symptoms come early. Due to the modern people’s eating habits and longterm sedentary ofﬁce style, lack of exercise coupled with obesity is a problem faced by most middle-aged and older people, which makes the knee joints intensify and the metabolism slows down. Non-systemic lesions of OA, the disease itself is a limited joint structure changes, only affected by the infringed cartilage [7], subchondral bone lesions and joint edge lesions briefly described as follows: (1) Changes in cartilage: The decrease of proteoglycans in cartilage components is the earliest change. This lesion causes softening and abrasion of the cartilage, especially at the site of weight bearing or stress concentration (mostly the inside of the knee). When the lesions intensify, the cartilage layer becomes thin, the surface is rough, and even local ulcers form. Finally, the cartilage is hardened, and the joints are exposed. The

170

Y.-C. Lin and F.-G. Wu

direct abrasion of hard bones causes difﬁculty in walking. Cartilage has no vascular nerves, no pain, only relying on peripheral blood vessels and synovial fluid to provide nutrients, cartilage wears faster than the repair rate, so the prevention and treatment of cartilage have not yet been more critical in the early stages. (2) Skeletal changes: Changes in bone degeneration can lead to bone hyperplasia, and the formation of osteophytes (bone spurs, bone hyperplasia). Osteophytes often occur under the edges of the joints or the cartilage, that is, the cartilage wear replaces the articular surface, and the bones become smooth, bright, hardened, and even ivory. The bone marrow cavity under the articular surface is subsidised and ﬁbrotic and causes a thin cystic sac and surrounded by new bone and ﬁbrous tissue. These cysts often occur below the ivory bone or at the attachment of the joint ligament and joint capsule, thus causing the bone structure to be abnormal and fragile. (3) Soft tissue changes: The soft tissues include non-skeletal tissues such as Muscle, Tendon, Ligament, and Bursa. In cases of mild or moderate inflammatory reactions, synovium and joint capsules become hyperplastic, ﬁbrosis, and cause cartilage or bone lesions. The function of blood circulation is to provide the nutrients needed by the organization and to take away the products after metabolism; the conduction and convection of heat transfer heat from the inside of the body to the skin [8]. Regarding bone quality, many studies have shown that blood plays a signiﬁcant role in the nutritional regulation of bone metabolism, which is related to the maintenance of the differentiation of articular cartilage and the secretion of synovial fluid by the synovial membrane. Increasing the blood flux of peripheral microvessels can reach a higher metabolic rate of the femur, which improves the cartilage metabolism of degenerative arthritis signiﬁcantly. Plasma contains essential to bone growth factors (similar to immune system T cell antibodies) that can help bone hyperplasia [9]. Regarding the temperature contributed by blood circulation, the core of temperature has a speciﬁc correlation with the peripheral nerve development of degenerative arthritis and the release of nerve conduction factor [10]. Increased blood flow has the beneﬁt of reducing muscle and nerve tension; increases in blood through the arteries through mechanical pressure, and increases muscle temperature through friction help improve the neurological perception of peripheral degeneration of bone [11]. Therefore, to explore the blood circulation of middle-aged and elderly will be quite useful in improving and preventing degenerative knee osteoarthritis.

3 Method 3.1

Research Method

OA patients observation Participants. A group of patients conﬁrmed to be diagnosed with a slight OA by a doctor. The Kellgren–Lawrence grading system scores between 0–2 level [12], as well as patients with mobility.

Design and Evaluation of an Innovative Assisting Device

171

Procedures. Observe the daily activities of a group of OA patients and record the type and frequency of events. First reach the patient’s living place, with the semi-structured interview to understand the regular life and movements of OA patients [13]. Specify some gesture transitions to observe body function. Ask what the assisting devices that you will use in your daily routine to understand the improvement of your use situation and subjective feelings are. 3.2

Experimental Evaluation

In this study, some objective scales will be used to quantify the status of OA patients. Through the WOMAC scale [14], gather the data of OA. Analysis of their needs for improvement. The aim of WOMAC scale divided into three aspects: pain, stiffness, and joint function to evaluate the structure and function of the hip and knee joints. The functional description is mainly for the lower limbs. You can use the entire system or pick a part of it when you use it. Scores can be recorded using VAS (visual analog scale) or 0–4 scales.

4 Design and Experiment 4.1

Design Criteria

To collect literature and observation on the effects of OA factors on posture transitions and blood circulation. The analyze the existing assisting devices’ function, organizing the mechanism of stimulation. Summed up the design criteria, and then design proposals. Then select the appropriate stimulation method and site of use. 4.2

Pillow Design Proposal

The design proposal will carry out the focus group method through the design criteria. The process of Focus group method is: (1) Experimental purposes. (2) The results of analysing OA patient daily life posture transition and the stimulation ways. (3) Discuss each other’s ideas. (4) Reduce the pain and stiffness in OA patients’ daily life. (5) Design the assisting devices for OA. (6) Discuss the design.

5 Expected Results Experimental evaluation of the expected results. Posture conversion of daily activities of OA patients and methods of stimulating deep blood circulation. Through analyzing the need for posture transitions. Reduce the inability can help improve the OA patients’

172

Y.-C. Lin and F.-G. Wu

quality of life. In this study, innovative assisting design helps to increase blood circulation of OA, reduce stiffness and pain.

References 1. Muthuri, S.G., et al.: History of knee injuries and knee osteoarthritis: a meta-analysis of observational studies. Osteoarthr. Cartil. 19(11), 1286–1293 (2011) 2. Hadjidakis, D.J., Androulakis, I.I.: Bone remodeling. Ann. N. Y. Acad. Sci. 1092(1), 385– 396 (2006) 3. Garip, Y., Güler, T., Tuncer, Ö.B.: Fatigue among elderly patients with knee osteoarthritis and its association with functional status, depression and quality of life. Ank. Med. J. 16(1) (2016) 4. World Health Organization: Chronic Rheumatic Conditions. Chronic diseases and health promotion (2012). http://www.who.int/chp/topics/rheumatic/en/ 5. Woolf, A.D., Pfleger, B.: Burden of major musculoskeletal conditions. Bull. World Health Organ. 81, 646–656 (2003) 6. Mazzuca, S.A., et al.: Pilot study of the effects of a heat-retaining knee sleeve on joint pain, stiffness, and function in patients with knee osteoarthritis. Arthritis Care Res. Off. J. Am. Coll. Rheumatol. 51, 716–721 (2004) 7. Pollard, B., Johnston, M.: The assessment of disability associated with osteoarthritis. Curr. Opin. Rheumatol. 18(5), 531–536 (2006) 8. Temmerman, O.P.P., et al.: In vivo measurements of blood flow and bone metabolism in osteoarthritis. Rheumatol. Int. 33(4), 959–963 (2013) 9. Zhang, W., et al.: Relationship between blood plasma and synovial fluid metabolite concentrations in patients with osteoarthritis. J. Rheumatol. (2015). https://doi.org/10.3899/ jrheum.141252 10. Sagar, D.R., et al.: Dissecting the contribution of knee joint NGF to spinal nociceptive sensitization in a model of OA pain in the rat. Osteoarthr. Cartil. 23(6), 906–913 (2015) 11. Weerapong, P., Hume, P.A., Kolt, G.S.: The mechanisms of massage and effects on performance, muscle recovery and injury prevention. Sports Med. 35(3), 235–256 (2005) 12. Petersson, I.F., et al.: Radiographic osteoarthritis of the knee classiﬁed by the Ahlbäck and Kellgren & Lawrence systems for the tibiofemoral joint in people aged 35–54 years with chronic knee pain. Ann. Rheum. Dis. 56(8), 493–496 (1997) 13. Van Teijlingen, E.: Semi-structured interviews (2014) 14. Hashimoto, H., et al.: Validation of a Japanese patient-derived outcome scale for assessing total knee arthroplasty: comparison with Western Ontario and McMaster Universities osteoarthritis index (WOMAC). J. Orthop. Sci. 8(3), 288–293 (2003)

Evaluation of Matching Degree Between Touch Gestures and User Mental Model Based on Event-Related Potential Ningyue Peng(&), Jing Ma, and Chengqi Xue School of Mechanical Engineering, Southeast University, Nanjing 211189, China [email protected]

Abstract. Designing gestures in multi-touch devices that are congruent with User Mental Model (UMM) is vital to remove the cognitive gulf, and reduce learning cost for users. In this study, a neurophysiological evaluation methodEvent-Related Potential was used to evaluate matching degree between touch gestures and UMM. Prime-probe paradigm was employed and two experimental conditions were compared-congruence and incongruence. Results show that negative component was elicited within the time window of 300 ms to 500 ms after the onset of probe stimuli in frontal, central and parietal lobe. With time went by, more negative components were found in the incongruence condition. Hereby, the former negative component is interpreted as a representation of the decoding process, and the later one interpreted as semantic matching with UMM. Results in this study may provide guidance for those assessing the usability of multi-touch gestures. Keywords: Matching degree Touch gesture Event-related potential Evaluation

User mental model

1 Introduction Thrust of multimedia technology has prompted a signiﬁcant changeover in the modal of human-computer interaction. A more natural and intuitive interaction strategy are advocated in recent years to echo the concept of ego-centered design. Touchinteraction, which enables users to directly manipulate the information system through gestures, has become a mainstream interactive mode on smartphones [1] and even large-sized interactive screens [2]. This new interaction method can narrow down the cognitive gulf resulted from differences between cognitive domain and representation domain, and have a facilitatory effect on users’ sense-making and reflection through affordances [3]. Deﬁning touch-interaction-based gestures that conform with the user’s mental model (including goals and expectations) can shrink study cost, especially for the green hand, and evoke positive emotions. This gives rise to the issue that how to evaluate the matching degree between user mental model and the touch gestures, and this issue is of great importance to maintain the learnability and usability of touch interaction. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 173–184, 2019. https://doi.org/10.1007/978-3-030-02053-8_28

174

N. Peng et al.

Mental model was ﬁrst put forward by Kenneth Craik, a Scotland psychologist, in the book of The Nature of Explanation [4]. According to Doyle and Ford, User Mental Model (UMM) is a long-lasting mental representation of the external world [5], which provides clues for users on how the system works and structure of the system [6, 7]. Thus, UMM is a lens through which users interact with the external world [8], including the representation world created by engineers and designers. Various subjective methods can be adopted to evaluate the matching degree between UMM and the representation domain in touch interaction, including Cognitive Walkthroughs, Heuristic Evaluation and Think Aloud, etc. These methods are ubiquitously used due to their convenience. However, disadvantages of such methods can not be overlooked, namely subjectivity and ambiguity. Such defects may impair the overall validity of evaluation. To overcome it, a more precise and sensitive technique - Event-related potential (ERP) was employed in this study. ERP is to uncover aspects of human sensory, cognitive and motor processes that underlie human thought and behavior [9] through capturing subtle potential fluctuations on the scalp [10]. ERP can be employed in researches on topics of violation of user intentions [11], social norm violation [12] and action sequence [13]. Among ERP components, N400 is a neural index of detection of unexpected anomalous stimuli [12], and it is sensitive to semantic mismatching information in actions [13]. It will be present when the upcoming stimulus is incongruent with users’ intentions or expectations. In this study, ERP investigation on matching cognition between tough-interactionbased gestures and user mental model was carried out. We interpret such matching cognition as a process of users’ decoding the gestures predeﬁned by designers. To our knowledge, there is no existing study on the cognitive mechanism of matching degree between UMM and predeﬁned gestures. And neurophysiological evaluation is also absent in this ﬁeld. Thus, some neural index is expected, in our study, to represent this kind of cognitive mechanism, and properties of such index can be used to evaluate the congruence between design and UMM. Potential ﬁndings of our study may overcome shortcomings of subjective and behavioral evaluation, and extend the evaluation methodology of touch-screen-based interaction usability.

2 Collecting and Constructing UMM Before investigating the cognition mechanism between UMM and touch gestures, collecting and constructing user mental model was required. Set of UMM for touch gestures were established via collecting the most commonly used gestures by participants. 2.1

Participants and Equipment

30 graduate students majoring in industrial design from Southeast University were enrolled as participants to construct UMM of touch gestures. All the participants are experienced in using touch-interaction-based devices. Ages of participants range from 23 to 28.

Evaluation of Matching Degree Between Touch Gestures and User Mental Model

175

Four D-Lab cameras were arranged around the participant to get a full-perspective capture of user’s gestures. Thinking path of each participants was recorded by D-lab cameras, which enables further analysis of UMM. An iPad was used to simulate the interaction effect after the user made their decision and gave the ﬁnal gesture which best matches the command word. The interaction effect (generated under After Effects, Copyright © 2018 Adobe) was to ensure a maximum simulation of the real touch interaction scenario. To eliminate effects from the prior knowledge of using iPad, a transparent plastic panel was put onto it, and participants were not aware that it was an iPad. An example of the simulated touch interaction scenario is shown in Fig. 1, and Experiment scenario is shown in Fig. 2.

Fig. 1. An example of the simulated touch interaction scenario in the experiment.

Fig. 2. Experiment scenario of collecting and constructing the set of UMM.

2.2

Materials for Constructing the Set of UMM

Ten pairs of most commonly used interaction commands were selected. They are listed as follows. Pair 1: Open vs. Close; Pair 2: Next page vs. Previous page; Pair 3: Enlarge vs. Shrink; Pair 4: Maximize vs. Minimize; Pair 5: Copy vs. Delete; Pair 6: Input vs. Cancel; Pair 7: Select vs. Uncheck; Pair 8: Move up vs. Move down; Pair 9: Move left vs. Move right; Pair 10: Accept vs. Decline.

176

2.3

N. Peng et al.

Procedures

Participants were ﬁrst briefed concerning the purpose of the experiment. The experimental test began with a brieﬁng in which the participants were given an overview of the experimental design and procedures. When the experiment began, they can explore the best-matching gesture without time limit. Once they made their decision, no cancellation was allowed. After ﬁnishing the task, participants were required to complete a 5-point Likert scale to rate difﬁculty, operative efﬁciency and satisfaction. 2.4

Constructing the Set of UMM for Touch Gestures

Gestures from all the 30 participants were recorded and went into further analysis. The most commonly used gesture for each command is deﬁned as the conforming users’ mental model. Table 1 shows the results of UMM. Graphical representations for each command are given in Fig. 3.

Table 1. Command words and the corresponding most commonly used gestures, which are represented as the users’ mental model. No. 1 2 3 4 5 6 7 8 9

Command Open Close Next Page Previous Page Enlarge Shrink Maximize Minimize Copy

10 11

Delete Select

12 13 14 15 16

Uncheck Input Cancel Move Up Move Down Move Left Move Right Accept Decline

17 18 19 20

Most commonly used gesture for command Click the icon with one ﬁnger Inward contraction with ﬁve ﬁngers Slide to the left with one ﬁnger Slide to the right with one ﬁnger Two ﬁngers open from inside to outside Two ﬁngers open from outside to inside Double click the page with one ﬁnger Drag current page to the bottom of window Long press with one ﬁnger and move towards the direction in which objects are copied Long press and move upwards with one ﬁnger Long press with one ﬁnger and drag from top to bottom to include the selected area Click the blank area with one ﬁnger Long press the input box Press and move leftwards with one ﬁnger Move upwards with one ﬁnger Move downwards with one ﬁnger Move leftwards with one ﬁnger Move rightwards with one ﬁnger Click the region within the dialog box Click the region outside the dialog box

Evaluation of Matching Degree Between Touch Gestures and User Mental Model

177

Fig. 3. Graphical representation of the touch gestures

3 ERP Experiment Based on the UMM collected and constructed in the ﬁrst phrase, Event-Related Potential (ERP) technique was adopted in this session to unveil the matching cognition mechanism and evaluate the matching degree between pre-deﬁned and touch gesture and UMM. 3.1

Participants and ERP Data Recording

23 participants were recruited from those who had already taken part in the ﬁrst experiment. It is to ensure that the UMM constructed in the last session has a high validity in this ERP experimental session. All participants have self-reported normal or corrected-to-normal vision. Ten female and 13 male participants ranging in age from 23 to 28 (mean of 25.25) completed the study. Participants were compensated and a bonus was offered to the highest-scoring participant in each between-subjects condition to provide motivation. ERP data was continuously recorded (band pass 0.5–100 Hz, sampling rate 1000 Hz) with ANT Ampliﬁer, adopting an electrode cap with 32 Ag/AgCl electrodes mounted according to the extended international 10–20 system and referenced to the

178

N. Peng et al.

left mastoid. The vertical ERP was recorded from the infra-orbital and supra-orbital electrodes on the right eye, and the horizontal ERP was recorded from electrodes on the outer canthi of both eyes. Electrode impedance was maintained below 5 kX in the experiment. Experimental materials were presented onto a 17-in. (43 cm in diagonal) screen with 1024 px multiple 768 px resolution. 3.2

Materials and Procedure

Five pairs of interaction command were selected according to the subjective evaluation on the difﬁculty, operative efﬁciency and satisfaction. All the selected commands have an average rating over 4.5 point (5 point in total), which are listed below. Pair 1: Next Page vs. Previous Page; Pair 2: Enlarge vs. Shrink; Pair 3: Select vs. Uncheck; Pair 4: Move Up vs. Move Down; Pair 5: Move Left vs. Move Right. Prime-Probe paradigm was used, in which command word is prime stimulus and touch gestures, in the form of short video, is probe stimulus. It has been demonstrated that videos elicit experiences similar to the perception of real world situations [14]. Thus, touch gestures have been played through videos, created via Adobe AffectEffects. The priming command word is to help users build up expectations about the subsequent gestures. Two conditions are contained in the experiment-matching and mismatching. Matching gestures for command words are according to the Table 1. Due to the pairing property of selected commands, mismatching gestures are the ones corresponding to the opposite command. All pairs of stimuli were randomly presented onto the screen and repeated for 8 times, with 4 times of matching condition and 4 times of mismatching condition, to guarantee a smooth brainwave in the off-line analysis. A withinsubject experiment design was employed. After getting familiar with the experiment procedure, participants entered the ERP experiment, which lasted twenty minutes per person. One core experiment unit is show in Fig. 4.

Fig. 4. Core experiment unit in the ERP experiment.

Evaluation of Matching Degree Between Touch Gestures and User Mental Model

179

4 Results and Analysis 4.1

Behavioral Data

Both evaluation time and accuracy were recorded and got further analysis. Mean evaluation time in matching condition is 1489.97 ms, with a standard deviation of 249.69 ms; mean evaluation time is 1767.46 ms, with a standard deviation of 362.99 ms in mismatching condition. Paired t-test was used during the analysis, where mean time for matching and mismatching condition were compared. Results show that there exists signiﬁcant difference in two conditions (t = 6.236, p = 0.000 < 0.01). Longer mean time in mismatching condition indicates that more hesitation may occur during their evaluation process. In another word, participants tend to give a more detailed processing of the stimuli, in order to reconcile the divergent command word and interaction gesture. In aspect of the accuracy rate representing the matching degree evaluation with a binary option (0 or 1), mean accuracy rate in matching condition is 96.85%, with a standard deviation of 3.62%; mean accuracy rate is 90.38%, with a standard deviation of 6.25% in mismatching condition. Signiﬁcant difference is found based on paired ttest (t = 5.568, p = 0.000 < 0.01). 4.2

ERP Data

ERP Data Analysis. EEG data was preprocessed offline by software of ASA 4.9 (ANT Neuro B.V., Netherlands). Electroencephalogram recordings were extracted from 200 ms before the onset of the touch screen gestures to 800 ms after the onset, with the ﬁrst 200 ms as baseline. Trials containing ampliﬁer clipping, bursts of electromyography activity, or peak-to-peak deflection exceeding ±80 lV and incorrect matching judgment were rejected before averaging. After rejecting all artifacts of EEG, at least 45 overlaps were available for each condition. EEGs were digitally rereferenced to the average of the left and the right mastoids and were ﬁltered with a 30 Hz low-pass ﬁlter (24 slope/db). At last, the EEG data were averaged separately for matching and mismatching conditions. ERP Data. According to former researches, N400 component is elicited in conditions of semantic violation and mismatching of the regular behavior sequence [13]. Thus, N400 component is also expected in our study. Based on the time window descriptions of N400 in Kutas and Federmeier’s [15], and Amoruso et al’s work [14], brainwave with a latency of 300–500 ms after the onset of probe stimuli were taken into account. Through visual inspection of the ERP waveform, both matching and mismatching condition elicited observable negative component during this time window in electrodes of frontal lobe (F3, F4, Fz); central lobe (C3, C4, Cz) and parietal lobe (P3, P4, Pz). Grand-averaged ERP in Fz, Cz and Pz in two conditions are displayed in Fig. 5. Larger amplitudes of negativity were found when the probe gestures are congruent with UMM. However, the difference does not reach to the signiﬁcant level according to 2 * 3 repeated ANOVA (F = 0.404, p = 0.532 > 0.05, η2 = 0.020). In addition, signiﬁcant main effect of electrode position was found (F = 4.020, p = 0.000 < 0.05, η2 = 0.167).

180

N. Peng et al.

Fig. 5. O Grand-averaged ERP waves recorded in Fz, Cz and Pz.

Fig. 6. Topographic map within the time window of 200 ms to 600 ms

No signiﬁcant interaction effect between matching condition and electrode position (F = 1.147, p = 0.335 > 0.05, η2 = 0.054). According to topographic map in Fig. 6, from 400 ms to 600 ms after the onset of the probe stimuli, more negativity is found, and they seem to spread from the left to right brain, from frontal to parietal lobe. To testify the effects of brain hemisphere and individual lobe on negativity within the time window of 300 ms to 500 ms, repeated ANOVA was performed. Results indicate that position of lobe carries signiﬁcant impact on the amplitude during 300 ms to 500 ms (F = 4.220, p = 0.022 < 0.05, η2 = 0.174). When gestures were congruent with UMM, amplitudes in frontal and central lobes are signiﬁcantly larger than those in parietal lobe according to pairwise comparison results (p1 = 0.022 < 0.05, p2 = 0.012 < 0.05). To compare the difference between the left, central and right hemisphere, we averaged amplitudes in left hemisphere (F3/C3/P3); central (Fz/Cz/Pz) and right hemisphere (F4/C4/P4) respectively. Amplitudes in left and central part were signiﬁcantly larger than those in the right part

Evaluation of Matching Degree Between Touch Gestures and User Mental Model

181

when gestures match UMM, which we deﬁne it as a left-hemisphere advantage. And only marginal difference was found between left, central and right brain part in the incongruence condition. Results are shown in Tables 2 and 3. Table 2. Average amplitudes in two conditions. Gestures congruent with UMM Left Central Right hemisphere part hemisphere –0.277 lV –0.261 lV 0.142 lV

Gestures incongruent with UMM Left Central Right hemisphere part hemisphere –0.122 lV –0.086 lV 0.126 lV

Table 3. Signiﬁcance between congruence and hemisphere according to pairwise comparison. Congruent with UMM Incongruent with UMM Left Central Left Central Central Right Right Central Right Right 0.911 0.036* 0.003* 0.791 0.094* 0.020 * indicates that the p value reaches to the signiﬁcant level (a = 0.05).

In addition, brain waves were more negative in incongruence condition, which began from around 500 ms after the onset of probe stimuli. No further statistic analysis was performed since there were hybrid components after 500 ms, and it is difﬁcult to extract one exact ERP component. However, the tendency that more negativity was elicited in the incongruence condition will be discussed in the next section.

5 Discussion Event-related potentials technique was employed to evaluate the usability of gesture design in a multi-touch interaction scenario. To our knowledge, this is the ﬁrst study in the research area of human-computer interaction. According to Sifaqui’s research, formation of UMM undergoes three levels: perceptive, computational and repository [16]. Users’ mental models when using gestureinteraction-based devices are generated based on their daily experience, knowledge structure and schemata. The ten pairs of commands that we chose in the experiment are the most commonly used ones in smart phone, touch-screen computers. A high familiarity with those selected materials is expected among all participants, which is to eliminate the interference of complexity and novelty. Meanwhile, being familiar with those interaction commands guarantees relatively consistent and constant mappings between command and the corresponding anticipated gesture before and after ERP experiment across participants. Presentation of command words and gestures in tandem conforms the rules and procedures based on which we interact with a multi-touch device. Imagine a real world

182

N. Peng et al.

scenario where we are required to perform a sequence of interaction procedures to ﬁnish one task. A command will emerge in our mind in the ﬁrst place according to the explicit representations onto the interface, e.g. close a full-screen window. Then, we might perform gestures, e.g. inward contract with ﬁve ﬁngers, and feedback is provided for users as an instruction. When presented with a gesture, regardless of its congruence with the command, there exists two main procedures-semantic decoding and semantic matching (procedures are illustrated in Fig. 7). In former studies, N400 is considered as a distinctive marker of semantic mismatch [17]. It is, thus, hypothesized before the experiment that more N400 would be evoked in the mismatching condition. Our original ERP results shows that no evident N400 were found in the two conditions within the regular time window (300 ms to 500 ms after onset), and it is attributable to two possible reasons. For one thing, the negative components captured in the time window of 300 ms to 500 ms might be associated with the decoding procedure that is involved in both of the two conditions. Thus no signiﬁcant effect was found between the negative amplitudes in two experimental conditions. Since materials in our experiment are presented in video, a format that is more complex than static screenshots [13] or photos [12]. It is therefore that latency of the expected N400 would delay. After ﬁnishing the decoding task, semantic matching will begin. As is reflected in Fig. 5, more negative brainwaves emerged from 500 ms posterior to probe onset to the terminal (1000 ms) in the incongruent condition. We conjecture that these negative parts contain N400 or N400-similar components that can be used to represent the semantic violation between gestures and user mental model. For another, the number of stimuli is not sufﬁcient enough to achieve desirable ecological validity. In our experiment, ﬁve pairs of commands were selected, and each of the command-gesture pair were repeated eight times in the ERP experiment. High repetition rate of stimuli may evoke memory, which results in the inapparent semantic matching evaluation-based ERP components. What is more, gestures can be classiﬁed into several groups, namely abstract gestures, compound gestures. Cognition and evaluation mechanisms vary across classiﬁcation. Thus, in the next study, it is recommended to investigate into different kinds of gestures. And a more systematic and inclusive UMM set of multitouch gestures is also expected.

Fig. 7. Procedures and timelines of cognition and evaluation

Evaluation of Matching Degree Between Touch Gestures and User Mental Model

183

6 Conclusion As a highly sensitive evaluation technique, event-related potential is used to uncover and represent the cognition and evaluation mechanism of multi-touch gestures. In our research, more negative amplitudes in frontal, parietal and central lobes were found in the incongruence condition after the onset of probe stimuli. However, both of the congruence and incongruence condition elicited negative components spanning from 300 ms to 500 ms posterior to the probe. We interpret the 300–500 ms negative effects as an index of gestures decoding, and the subsequent negative components are more likely to represent semantic matching cognition. And left-side advantage was found in both conditions, which conforms to the conclusion in former literatures that the left hemisphere controls the right-side actions. However, due to insufﬁcient samples of multi-touch gestures, ecological validity of our research will impair. But this research can be intended as a foundational step upon which future studies addressing evaluation the usability of multi-touch gestures could build. Acknowledgments. Author 1 gratefully acknowledge the grant from the Fundamental Research Funds for the Central Universities & Postgraduate Research & Practice Innovation Program of Jiangsu Province (No. KYCX17_0056). As well, all the authors want to acknowledge the grant from Natural Science Foundation of China (No. 71271053).

References 1. Kim, Y.H., Lee, J.H.: Game interface enhancement under smartphone platform focused on touchscreen interaction, pp. 45–61. Pergamon Press, Inc. (2015) 2. Kukimoto, N., Onoue, Y., Aoki, K., Fujita, K., Koyamada, K.: HyperInfo: interactive large display for informal visual communication. In: International Conference on Network-Based Information Systems, pp. 399–404 (2014) 3. Valdes, C., et al.: Exploring the design space of gestural interaction with active tokens through user-deﬁned gestures. In: SIGCHI Conference on Human Factors in Computing Systems, pp. 4107–4116 (2014) 4. Craik, K.J.W.: The nature of explanation. Stud. Q. J. 14(55), 44 (1967) 5. Doyle, J.K., Ford, D.N.: Mental models concepts revisited: some clariﬁcations and a reply to Lane. Syst. Dyn. Rev. 15(4), 411–415 (1999) 6. Senge, P.M.: Mental models. Plan. Rev. 20, 4–44 (1992) 7. Johnsonlaird, P.N.: Mental models: towards a cognitive science of language. Inference Conscious. 47(1/2), 438–445 (1983) 8. Forrester, J.W.: Industrial dynamics. J. Oper. Res. Soc. 48(10), 1037–1041 (1997) 9. Luck, S.J., Kappenman, E.S.: ERP components: the ups and downs of brainwave recordings. In: The Oxford Handbook of Event-Related Potential Components, pp. 3–30. Oxford University Press (2012) 10. Berti, S., Schröger, E.: A comparison of auditory and visual distraction effects: behavioral and event-related indices. Brain Res. Cogn. Brain Res. 10(3), 265 (2001) 11. Shibata, H., Gyoba, J., Suzuki, Y.: Event-related potentials during the evaluation of the appropriateness of cooperative actions. Neurosci. Lett. 452(2), 189–193 (2009)

184

N. Peng et al.

12. Mu, Y., Kitayama, S., Han, S., Gelfand, M.J.: How culture gets embrained: cultural differences in event-related potentials of social norm violations. Proc. Natl. Acad. Sci. U. S. A. 112(50), 15348–15353 (2015) 13. Reid, V.M., Striano, T.: N400 involvement in the processing of action sequences. Neurosci. Lett. 433(2), 93–97 (2008) 14. Amoruso, L., et al.: N400 ERPs for actions: building meaning in context. Front. Hum. Neurosci. 7(57), 57 (2013) 15. Kutas, M., Federmeier, K.D.: Thirty years and counting: ﬁnding meaning in the N400 component of the event related brain potential (ERP). Annu. Rev. Psychol. 62(1), 621–654 (2011) 16. Sifaqui, C.: Structuring user interfaces with a meta-model of mental models. Comput. Graph. 23(3), 323–330 (2013) 17. Szucs, D., Soltész, F., Czigler, I., Csépe, V.: Electroencephalography effects to semantic and non-semantic mismatch in properties of visually presented single-characters: the N2b and the N400. Neurosci. Lett. 412(1), 18–23 (2007)

Co-creation Workshop Oriented to the Autonomous Elderlies in Chile: Unveiling the Experience of the Participants Juan Carlos Briede-Westermeyer1(&), Cristhian Pérez-Villalobos2, Javiera Ortega-Bastidas2, and Isabel Leal-Figueroa1 1

Department of Art and Design Technologies, Universidad del Bío-Bío, Avda. Collao 1202, Concepción, Chile [email protected] 2 Department of Medical Education, Universidad de Concepción, Concepción, Chile

Abstract. A Co-creation workshop was held with the participation of elderly adults, pre-elderly adults, professionals from the industrial design area, entrepreneurs and health professionals, in which a descriptive-comparative observational methodology was assessed. The results show a diversity of deﬁning words mentioned by the participants. A semantic weight is presented for each section of the instrument, related to the words which were used most often by the subjects participating in the study. The weight of the semantic networks of each section observed, allow us to have essential feedback about the experience in a co-creative process. This becomes extremely relevant, when subjects who the design of a product is aimed at, participate. This allows being even more contextualized and gathers the real needs of a given social group. Keywords: Co-creation Product design Human-Centered design Generative research Elderly Semantic networks

1 Introduction Human-centered design requires using techniques which communicate, empathize, stimulate and promote the interaction of the people involved, obtaining an understanding of their needs, wishes and experiences which goes beyond what the people directly realize [1]. Aging stands out among the topics that this type of design can currently face. It constitutes one of the great challenges that society must face over the coming years. Demographic aging is a process which has been established as an economic impact phenomenon on public policies and on the development of communities and families, projecting itself as one of the most important social transformations of our time [2]. Motivated by the increase in life expectancy and the reduction of birth rates [3], aging, apart from becoming a global challenge, is one of the main sanitary concerns [4, 5], focusing the discussion on aspects like mental health, infectious diseases, chronic

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 185–190, 2019. https://doi.org/10.1007/978-3-030-02053-8_29

186

J. C. Briede-Westermeyer et al.

illnesses and, deﬁnitively, integrated health care, but it also focuses on aspects associated to resources, specialists and health insurance [2]. For this reason, aging and the elderly present complex problems which must be faced in an interdisciplinary manner, looking to design products and services that meet cultural demands and differences [6] along with biopsychosocial changes within this segment of the population [7]. The early phases of the design process, known as Front End of Innovation (FEI) or Fuzzy-Front End (FFE) [8] constitute the opportunity where the product’s strategy is formulated, opportunities are identiﬁed and chosen, ideas are generated, evaluated and chosen, the concept is developed and then tested [9]. Co-creation is an alternative to use at this point, as it allows having a collective act of creation which is experienced jointly by two or more people and it is understood as a special case of collaboration where the intention is to create something new that was not known about beforehand [10], with the goal of approaching the future users who it will be designed for [11]. Generative research, which according to Sanders [12],… “is done at the very start of the design process’s Fuzzy-Front End, where often it is not known whether the design process’s deliverable will be a product, a service, an interface or something else entirely. Here, the goal of the pre-design phase is to ﬁnd the problems to be resolved and identify the opportunities to explore and determine what needs to be done and what does not, what is designed and what is manufactured…” [12], involving the participants, far from the conventions and removing the limits between roles, so that the individuals can participate at a balanced, horizontal level [11]. A Co-Creation Workshop was held in this context, where the elderly, pre-elderly, industrial design professionals, entrepreneurs, physicians, psychologists, physiotherapists and other health professionals took part. The workshop lasted a week, with three sessions, where the co-creation methods, the deﬁnition of the problem and creation of prototypes were taught via Challenge Based Learning [6], before ending with the deﬁnition of areas of future opportunity. The goal of this study is to present the perception of the participants’ personal contribution in this Co-Creation Workshop, whose focal issue was the elderly.

2 Methodology The workshop was evaluated using a qualitative methodology, with a descriptivecomparative observational design. 19 people participated in a Co-creation Workshop that was focused on problems of the independent elderly. The participants were chosen via a non-probabilistic quota sampling so that they represented diverse proﬁles: elderly, pre-elderly, industrial designers, entrepreneurs and health professions. The work was done in groups comprising at least one representative member of each proﬁle. At the end of the workshop, the participants answered a questionnaire where they could share their experiences. The instrument for this, was built using the natural semantic networks technique and comprised four sections where the subjects had to indicate the deﬁning word which they felt was most closely related to the following topics: (1) my contribution to the group work; (2) my communication with other members of the group; (3) teamwork in

Co-creation Workshop Oriented to the Autonomous Elderlies in Chile

187

my group and (4) what I learned. The Natural Semantic Networks processing proposed by Figuera et al., [13] was used for its analysis. After building a table with all the deﬁning words that were mentioned by the workshop’s participants, the two main semantic network values were calculated: (a) J Value, which represents all the deﬁning words mentioned by the participants. This allows us to obtain an indicator of the network’s semantic richness. That is to say, more words, more richness; (b) M Value, is related with the hierarchy with which each of the subject’s deﬁning words has highest weight. This allows obtaining the semantic weight of each word. The group of M values is known as SAM set (Semantic Association Memory). The SAM sets are the basis for later intra and inter group comparisons and correlations [13]. However, for this study they are analyzed descriptively.

3 Results and Discussion The results show a diversity of deﬁning words mentioned by the participants. A semantic weight is presented for each section of the instrument, related to the words which were used most often by the subjects participating in the study. For the my contribution to the teamwork section, a total J value of 88 deﬁning words was seen, from which ten hierarchies were obtained. A ﬁnal diversity of seven words was seen, with “devise” having the highest semantic weight (SAM = 38) (Tables 1, 2, 3 and 4). Table 1. Results - valuation of words in my contribution to the teamwork Stimuli Hierarchy My Contribution 1 2 3 4 10 9 8 7 Devise 0 1 1 3 Collaboration 3 0 0 0 Creativity 1 1 1 0 Experience 1 1 1 0 Listen 1 1 0 1 Knowledge 1 2 0 0 Motivation 1 1 0 0

5 6 0 1 0 0 0 0 0

6 5 0 0 1 1 1 0 1

7 4 0 0 0 0 0 0 0

8 3 0 0 0 0 0 0 0

9 2 0 0 1 0 0 0 0

10 1 0 0 0 0 0 0 0

Total freq M value 5 4 5 4 4 3 3

38 36 34 32 31 28 24

The rest of the deﬁning words were: creativity, collaboration, experience and listen. Words like knowledge and motivation appeared further down the list. For the section, my communication with other team members, a total J value of 86 deﬁning words was seen, from which six hierarchies were obtained. Six ﬁnal deﬁning words were seen, with “flow” and “listen” being the words with highest semantic weight (SAM = 21). The rest of the deﬁning words were: respect, valuable, motivation and interesting. For the teamwork section, a total J value of 81 deﬁning words was seen, from which six hierarchies were obtained. Six ﬁnal deﬁning words were seen, with the word

188

J. C. Briede-Westermeyer et al. Table 2. Results of word valuation in my communication Stimuli Hierarchy My Communication 1 2 3 4 5 6 5 4 3 2 Flow 0 2 2 1 0 Listen 1 1 1 2 0 Respect 3 0 0 1 0 Valuable 0 1 0 1 2 Motivation 0 1 1 1 0 Interesting 0 0 0 0 0

6 1 0 0 0 0 0 2

Total Freq M Value 5 5 4 4 3 2

21 21 21 12 12 2

Table 3. Results - valuation of words in teamwork Stimuli Teamwork

Hierarchy 1 2 3 4 6 5 4 3 Union 1 2 0 0 Collaboration 2 0 0 1 Respect 0 1 1 0 Interesting 1 1 0 0 Commitment 1 0 1 0 Flow 0 0 2 1

5 2 0 0 1 0 0 0

6 1 0 0 0 0 0 0

Total Freq M Value 3 3 3 2 2 3

19 16 14 14 13 12

Table 4. Results - valuation of words in what I learned Stimuli Hierarchy What I learned 1 2 3 4 5 6 5 4 3 2 Co-create 2 1 0 1 0 Design 1 2 0 1 0 Devise 3 2 0 0 0 Team spirit 1 1 1 0 0 Creativity 1 0 1 1 0 Collaboration 0 0 1 0 0

6 1 0 0 0 0 0 1

Total Freq M Value 4 4 3 3 3 2

20 19 18 15 13 6

“union” having a SAM of 19. The rest of the deﬁning words were: collaboration, respect, interesting, commitment and flow. Finally, for the section, what I learned, a total J value of 74 deﬁning words was seen, from which six hierarchies were obtained. A total of six ﬁnal deﬁning words were seen, where “co-create” had the highest semantic weight (SAM = 20). The rest of the deﬁning words were: design, devise, team spirit, creativity and collaboration.

Co-creation Workshop Oriented to the Autonomous Elderlies in Chile

189

The diversity of the deﬁning words obtained, allow identifying those relevant processes and experiences that occurred as a result of the workshop. Speciﬁcally, they allow identifying the way in which the users got involved in the design process studies.

4 Conclusions The nature of this study allows explaining how diverse views about a work and design phenomena, on coming together, can nourish the same creation experience. The weight of the semantic networks of each section observed, allow us to have essential feedback about the experience in a co-creative process. This becomes extremely relevant, when subjects who the design of a product is aimed at, participate. This allows being even more contextualized and gathers the real needs of a given social group. Upon valuing the design, it is left clear that a great number of the participants, without their proﬁle being relevant, contributed to the collaborative design process. On evaluating teamwork and communication, the participants demonstrated having the willingness to listen and to involve the ideas of the group’s participants in the design of a solution, valuing the flow, that is to say, harmoniously leading the design process towards a ﬁnal, and convincing result. Regarding the ﬁnal lesson learned in the workshop, the participants mention cocreation, ﬁrst of all, and design secondly. This shows that independent of their proﬁle, they managed to understand and reach the target of the workshop. The workshop’s ﬁnal evaluation upholds that the participants understood their contribution to the design and, therefore, understood the basis of collaborative design. Acknowledgments. This work was supported by the CONICYT FONDECYT 1171037 project.

References 1. Giacomin, J.: What is human centred design? Des. J. Int. J. Asp. Des. 17(4), 606–623 (2014) 2. Comisión Económica para América Latina y el Caribe: Derechos de las personas mayores: retos para la interdependencia y autonomía (LC/CRE.4/3/Rev.1), Santiago (2017) 3. Mora, T., Herrera, F. (ed.) Convención Interamericana sobre la Protección de los Derechos Humanos de las Personas Mayores: Análisis de brechas legislativas y propuestas para su implementación en Chile. Ediciones Servicio Nacional del Adulto Mayor, Santiago (2018) 4. Bayarre, H.D.: Múltiples perspectivas para el análisis del envejecimiento demográﬁco. Una necesidad en el ámbito sanitario contemporáneo. Rev. Cuba. Salud Pública 43(2), 313–316 (2017) 5. Mejía, M., Rivera, P., Urbina, M., Alger, J., Maradiaga, E., Flores, S., Rápalo, S., et al.: Discapacidad en el adulto mayor: Características y factores relevantes. Rev. Fac. Cienc. Méd., 11(2), 27–33 (2014). Accessed http://www.bvs.hn/RFCM/pdf/2014/pdf/RFCMVol112-2014.pdf 6. Briede-Westermeyer, J.C., Pérez-Villalobos, C., Bastías-Vega, N., Bustamante-Durán, C., Olivera-Morales, P., Parra-Ponce, P., Delgado-Rivera, M., Cabello-Mora, M., CamposCerda, I.: Interdisciplinary experience for the design of health care products. Rev. Med. Chile 145, 1289–1299 (2017)

190

J. C. Briede-Westermeyer et al.

7. Briede-Westermeyer, J.C., Leal-Figueroa, I., Pérez-Villalobos, C.: Designers, are they stars or facilitators of interdisciplinary work? The co-creation and consensus in the design of products for elderly. Revista 180(40), 20–29 (2017) 8. Heck, J., Rittiner, F., Steinert, M., Meboldt, M.: Iteration-based performance measurement in the fuzzy front end of PDPs. Procedia CIRP 50, 14–19 (2016) 9. Stevanović, M., Marjanović, D., Štorga, M.: Managing the process of preparation for product development - ideas assessment and evaluation. In: DS 84: Proceedings of the DESIGN 2016 14th International Design Conference, pp. 1155–1164 (2016) 10. Sanders, L., Simons, G.: A social vision for value co-creation in design. Open Source Business Resource, December 2009 11. Sanders, E.B.N., Stappers, P.J.: Co-creation and the new landscapes of design. Co-design 4 (1), 5–18 (2008) 12. Sanders, E.: Perspectives on Participation in Design [Portal de servicios de consultoría y educación en formas colectivas de creatividad] En MakeTools (2013). Accessed http://www. maketools.com/articles-papers/Sanders2013Perspectives.pdf 13. Figueroa, J., González, E., Solís, V.: Una aproximación al problema del signiﬁcado: Las redes semánticas. Rev. Latinoam. Psicol. 13(3), 447–458 (1981)

Do Design Outcomes Get Influenced by Type of User Data? An Experimental Study with Primary and Secondary User Research Data Abhishek Dahiya(&) and Jyoti Kumar Indian Institute of Technology Delhi, New Delhi, India [email protected], [email protected]

Abstract. In User Centered Design (UCD), designers employ various user research methods to collect user data which help them in making design related decisions. While collecting primary user data, designers get a ﬁrst-hand experience of users’ problems and their task environments. This paper reports an experimental study with 20 novice designers under influence of primary and secondary user data. Design solutions generated by novice designers were analyzed to observe design thinking. It was found that under the influence of primary user data, novice designers suffered more design ﬁxations, had less divergent thinking but the concepts generated were more detailed with focus on users’ tasks. In light of the ﬁndings, this paper argues that novice designers get ﬁxated with existing solutions used by the users while doing user research and ﬁnd it difﬁcult to think of novel solutions while ideating. Keywords: Ill-deﬁned problem solving User Centered Design User research

Design cognition

1 Introduction User research is one of the initial steps in User Centered Design(UCD) process. User research is necessary because it help designers to look at the design problem through user’s perspective. According to Norman [1] designers may have a different mental model or mental representation of problem than their users. Hence, design solutions proposed by designer using his mental model of the problem, might not work for the user [1]. Further, designers are expected to ensure that problem is solved with users’ accepting the proposed design solution. In order to do that designers, need to understand users’ interests, motivations, beliefs, needs etc. Therefore, user research helps designers in looking at the problem from user’s perspective and developing solutions accordingly. There are various methods suggested by design practitioners and researchers for user research. Some of these methods are ethnographic research, contextual enquiry, interviewing, shadowing, focus groups etc. All these methods produce rich data which is helpful to designers in taking various design decisions in the design process. Often design ﬁrms and industries employ dedicated teams for user research. Data thus © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 191–197, 2019. https://doi.org/10.1007/978-3-030-02053-8_30

192

A. Dahiya and J. Kumar

collected from user research is presented to the designers to help them build an understanding of their users. Authors deﬁne user data collected by user research team and presented to designers as secondary user data. Unlike this case, there are design ﬁrms and industries in which designers themselves go out and do the user research. Authors deﬁne user data collected by designers themselves using user research methods as primary user data. The authors posit that there might be difference in development of user understanding in both the cases. There could be various reasons for this phenomena, for example the designer might misinterpret the information presented, or the designers might wrongly associate the personality studied etc. The aim is to investigate the changes in design solutions in both the cases i.e. when a designers work with primary and secondary user data. This paper presents an experimental study to understand the effect of primary and secondary user data on design solutions. Two groups of industrial design students were studied; one being subjected to user research before solving a design problem. The other group was asked to solve design problem on the basis of user data presented to them. The analysis of design cognition was done through qualitative assessment of the solution sketches made by them along with deep qualitative interviewing.

2 Importance of Understanding Users in Design Existing literature has argued the importance of understanding users in the design process. The term User Centered Design (UCD) was coined to connote a design process in which user remains at focus [1]. UCD has been deﬁned as “a philosophy based on the needs and interests of the user, with an emphasis on making products usable and understandable” [2, 3]. The innate need for keeping the user in spotlight during the design process is further justiﬁed by Nielsen [4] arguing that “What might be ‘friendly’ or ‘easy’ for one user may not be friendly or easy for another [3]. The key idea of user centered design is to approach the design problem from the user’s perspective [5]. Thus, role of a designer is to solve the problem and to ensure that the user accepts the design solution with minimum effort to learn how to use it [6].

3 Methods to Study Design Thinking Design literature reports several studies that have attempted to observe design thinking patterns of a designer [7, 8]. Design cognition has often been studied by verbal protocols, where designers’ behaviors were analyzed [9]. Cognitive processes of designers have often been coded and analyzed through frequency of occurrence of a behavior of interest [9]. Some other research has also used semantic analysis of the verbal protocols or observed behaviors [10]. Analysis of design thinking has also been done using sequence of sketches or by the outcome sketches developed by designers while ideating for solutions [10, 11]. In this paper, design solutions generated by designers has been used to understand the influence of user data on design thinking. Expert ratings has been used on the design outcome sketches on ‘novelty’, ‘ﬁxations’ and ‘practicality’ in design thought. ‘Novelty’ has been argued to be a key characteristic of a creative

Do Design Outcomes Get Influenced by Type of User Data?

193

problem solving process [12]. Also, ‘ﬁxations’, that is, repetition of certain elements of design across several, has been used to assess the quality of designing thinking [11, 13]. Fixations are argued to represent blocks in a designer’s thinking process that inhibits novel ways to approach a solution and limits the scope of creative problem solving. ‘Practicality’ is also taken as an important parameter of design thinking. Practicality denotes to a pragmatic thinking approach to idea generation for a design problem [14]

4 Experiment Design In this study, total of 20 design students participated. To half of them, that is to 10 participants, prior users research data presented through a standard persona, scenario and problem description was given. To the other half, a problem description was given and no persona and scenario description was provided. Instead they were instructed to visit the user in their environment and develop their own persona and scenario as per a given standard structure. The ﬁrst group has been called ‘secondary user data group’ or group ‘S’. The second group which collected data and prepared their own persona and scenario has been called ‘primary user data group’ or group ‘P’. 4.1

Design Task

Participants were given one problem each. Each of the 10 problem statements were created after the authors visited and observed the 10 different users. Each of the 10 different problems were related to one of the 10 users and no two problems came from same user or same problem statement from two users. This arrangement ensured that each of the 10 primary user data based designers had different users to visit while simultaneously same user was represented through persona and scenario to the 10 secondary user data based designers. The different users selected were common community workers like gardener, cobbler, painter, rickshaw puller, temple priest, bicycle mechanic etc. The problems were reported to the participates through a very crisp phrases which was fairly open ended. For example, some of the problem statements were: a tool kit for cobbler, worship plat for priest, grass cutter for gardener etc. These problem statements were given to participants on a printed paper. 4.2

Participants

A total of 20 students participated in the study. The participants in this experiment were Masters in Industrial Design students at Indian Institute of Technology, Delhi. Participants constituted 13 males and 7 females with age ranges between: 23–27 yrs. (average age: 24.6 yrs., Std. dev.: 1.18 yrs.).

194

A. Dahiya and J. Kumar

5 Data Analysis Qualitative data in the form of sketches and verbal protocols was collected. Design thinking is facilitated by sketching and in turn thus the sketches become an insight into the designers thinking process. The sketches were thus used to evaluate the quality of design thinking. The sketches submitted by the participants were evaluated by three experts for ‘Novelty’, ‘Fixation’ and ‘Practicality’ as discussed in Sect. 3 these parameters have been reported in literature as some of the markers of quality of design thinking [10]. Two of the three experts were industrial designers one of which was an interaction designer having ﬁve years of experience in design industry, other had one year’ experience in manufacturing industry. Third expert was associated with design academics for more than seven years. Experts were asked to give a rating between 1 to 10 on three parameters, ‘Novelty’, ‘Fixation’ and ‘Practicality’. Prior to rating, experts were explained the three constructs with examples as below: Novelty: Ideas categorized as non-novel corresponded to concepts partially or fully similar to the existing solutions. Other non-novel ideas depicted solutions currently being used in existing real life situations. Any other types of ideas were categorized as novel. Any sketch rated as 1 is taken as least novel and rating 10 was taken as most novel solution (Figs. 1, 2 and 3).

Fig. 1. Design solutions for grass cutting tool for gardeners, examples of novel (left) and nonnovel(right) ideas.

Fixations: Ideas rated highest in ‘ﬁxation’ corresponded to concepts which had repetitions of key attributes (Physical form, functioning principle, direction of use) in successive design solutions. Ideas rated low in ‘ﬁxation’ corresponded to concepts that had less or no repetition of key attributes. Any sketch rated as 1 was taken as least ﬁxated and rating 10 was taken as most ﬁxated solution. Practicality: Ideas rated as high on ‘practicality’ corresponded to concepts which denoted state of being ‘useful’ or ‘applicable’ in indented users’ current real life situations. Ideas rated low on ‘practicality’ were concepts which had ambiguity on the applicability of design solution. Any sketch rated as 1 was taken as least practical and rating 10 was taken as most practical solution. Qualitative unstructured in-depth interviews with all the 20 participants were conducted post sketching where the aim was to understand the ﬁrst person account of designers’ approach to design solution. Interviews were recorded and transcribed.

Do Design Outcomes Get Influenced by Type of User Data?

195

Fig. 2. Design solutions of toolkit design, examples of ﬁxation with existing solution- a toolbox (left) and non-ﬁxated solutions – a tool shirt (right).

Fig. 3. Design showing practical details in conceptual solutions for utensil washing tool (left) and grass cutting tool (right).

The interview data was analyzed for emergent themes in design thinking and was corroborated with the expert ratings.

6 Results and Conclusion Scores obtained through experts rating on design solutions were analyzed through two tailed t-test and the differences were found to be statistically signiﬁcant between the two groups of participants. Results indicate that participants working with primary user data (Group P) scored less in novelty but high in ﬁxation and practicality. On the other hand, participants working with secondary user data (Group S) scored more in novelty and less in ﬁxation and practicality (Fig. 4). Post design, interviews were conducted with designers and the interview data was qualitatively analyzed. One of emerged themes from interview data was difference in the way two groups had processed the persona information. 9 out of 10 participants working with secondary user data had focused on solving problems that were ‘reported’ in user information format. Whereas 6 out of 10 participants working on primary user data said in the interview that while they had created the person information themselves but during the design thinking they naturally focused on the memory of the user’s observations rather than what was textually described in the

196

A. Dahiya and J. Kumar

Fig. 4. Box plot showing expert ratings for both groups(left), t values for novelty, ﬁxation and practicality (right).

persona description. And the participants in ‘group P’ used the word ‘felt’ often to describe the problems with persona. For e.g.: They reported that they ‘felt’ that a fruit cart vendor works in hot weather conditions therefore, providing space for keeping water bottle in fruit cart design will be an important feature. Another emergent theme from the interview data was difference in the way participants felt creative freedom while designing. It was noticed that the ‘group P’ participants (7 of 10) were more conditioned by the constraints they tacitly felt in users’ environment though they were not thoroughly described in the persona text. One of the participant reported that it was “hard for him to let go off his initial ideas” that he thought of while doing the user research. The ‘group S’ participants did not feel any constraint and were able to apply their creativity with less bounds This paper concludes that type of user data, that is primary user data collected by designers and secondary data collected by others, has an influence on the design thinking process. The difference is reflected in the design outcomes as well as in their design thinking process.

References 1. Norman, D.: The Design of Everyday Things: Revised and Expanded. Hachette, New York (2013) 2. Norman, D.A.: The Design of Everyday Things/Emotional Design/Design of Future Things. Basic Books, New York (1988) 3. Corry, M.D., Frick, T.W., Hansen, L.: User-centered design and usability testing of a web site: an illustrative case study. Educ. Technol. Res. Dev. 45(4), 65–76 (1997) 4. Nielsen, J.: Usability Engineering. Morgan Kaufmann (1994) 5. Norman, D.A., Draper, S.W.: User Centered System Design: New Perspectives on HumanComputer Interaction. Erlbaum Associates, Hillsdale (1986) 6. Preece, J.: Supporting user testing in human-computer interaction design. In: LNCS, pp. 256–267 (1991)

Do Design Outcomes Get Influenced by Type of User Data?

197

7. Chai, K.-H., Xiao, X.: Understanding design research: a bibliometric analysis of design studies (1996–2010). Des. Stud. 33(1), 24–43 (2012) 8. Dorst, K., Cross, N.: Creativity in the design process: co-evolution of problem–solution. Des. Stud. 22(5), 425–437 (2001) 9. Kim, E., Kim, K.: Cognitive styles in design problem solving: insights from network-based cognitive maps. Des. Stud. 40, 1–38 (2015) 10. Tversky, B., Suwa, M.: Thinking with sketches. In: Tools for Innovation, pp. 75–84. Oxford University Press (2009) 11. Cardoso, C., Badke-Schaub, P.: Fixation or inspiration: creative problem solving in design. J. Creat. Behav. 45(2), 77–82 (2011) 12. Sarkar, P., Chakrabarti, A.: Assessing design creativity, Design Studies (2011) 13. Smith, T.F., Waterman, M.S.: Identiﬁcation of common molecular subsequences. J. Mol. Biol. 147, 195–197 (1981) 14. Kaufman, J.C., Sternberg, R.J. (eds.): The Cambridge Handbook of Creativity. Cambridge University Press, New York (2010)

Medical Device Design Challenges Based on Users Hierarchy and Their Correlation with Illness Fabiola Cortes-Chavez(&), Maria Giovanna-Trotta, Paulina Manzano-Hernandez, Alberto Rossa-Sierra, and Gabriela Duran-Aguilar Universidad Panamericana, Facultad de Ingeniería, prolongación Calzada Circunvalación Poniente 49, 45010 Zapopan, Jalisco, Mexico [email protected]

Abstract. The absence of industrial designers in the design of medical devices, which are currently designed mostly by engineers and/or professionals with less study in the human centered design methodologies, as well as the rigorous regulations that govern the design of medical products, have diminished the importance of implementing a hierarchy of users during the design process. Most importantly, this hierarchy varies depending on the stage of the disease in which the patient is located. The medical devices should be designed taking into account the hierarchy of users in importance, priority of needs and their correlation with the stage of the patient’s illness. Keywords: Medical device design Engineering design User-centered design Design process Design challenges Usability User hierarchy Industrial design

1 Introduction All project developments start by considering a problem, or inconsistent functions in the system that fail to answer the user´s needs. Medical devices are based on an improvement of the correspondence between needs and functions related to both the clinical environment and the different users and the prioritization of their needs. In hospitals, due to the vulnerability and the relevance of the user conditions, certain design procedures have to be planned according to the institution´s regulation plan and design optimizations must be done through veriﬁcation/validation. To reduce risk and improve patient safety, the human factors have to be the base of the entire design development process [7]. The convenience of use, the adaptability, and the ease of use are parameters required to achieve speciﬁed goals with effectiveness, efﬁciency, and satisfaction, within a speciﬁc context. The FDA (Food and Drug Administration) underlines the importance of following three essential steps to guarantee the control and reduce the hazards in the process of

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 198–203, 2019. https://doi.org/10.1007/978-3-030-02053-8_31

Medical Device Design Challenges Based on Users Hierarchy

199

medical device design. Where, these steps are: 1- Identify use-related hazards, 2- Develop and apply strategies to mitigate or control use related hazards, 3- Demonstrate a safe and effective device use through Human Factors validation testing [8]. The medical experience which is traditionally poor in engagement, has a huge opportunity of being more integrated and highly compromised by considering the user’s [9]. The United Kingdom´s National Patient Safety Agency studied the incidence of medical design in promoting accidents. They agreed on the responsibility of medical devices in facilitating accidents, recognizing that the devices have been designed without developing a deep acknowledgment on the users’ expectations [6]. To improve the patients’ healthcare services, industry depends upon the supply of well-designed medical devices. A medical device characterized as “well designed” is clinically effective, safe, and meets the people’s needs. [5].

2 Challenges of Designing Medical Devices Typically, a design process, requires the designer to be involved in the consideration of complex interactions between product, subsystems, environment and user. However, in the case of the design of medical devices, this problem is further complicated by the presence of a large number of interested parties and particularly complex use environments [3]. Engineering design is a demanding process where the engineering design teams are typically not made up of experts in the medical domain. This situation can complicate the achievement of an optimal end result of the process because of the lack of knowledge of potential users or environments of use. Solutions proposed in this process within medical environments are constantly driven by clinical and biological contexts. These two contexts set design requirements and constraints that need to be contemplated in order to propose a design that is able to anticipate and adapt to contextrelated challenges that could arise [2]. Hence the successful measurement and compliance of the user’s requirements during the design process will result in less product recalls and modiﬁcations, and patient safety and device effectiveness will be improved. It is relevant to highlight that potential users of medical devices make up a highly heterogeneous group [5]. Interdisciplinary contributions between engineering design and industrial design is fundamental to launch successful products to the market since product design involves a multidisciplinary perspective. Nevertheless, it is known that engineering design and industrial design have signiﬁcantly different design practices, and their design approaches are, in some way, opposed to each other. Industrial designer´s role involves enhancing the user´s experience when interacting with a product and making sure that the outside form an interface plus to coherent aesthetic and ergonomic principles. When collaborating with industrial designers, engineering designers participate in the implementation of the design concept developed by industrial designers. Engineering designers aid in making the product work and take the necessary actions to make it reliable and manufactured. This leads to different approaches among them [4].

200

F. Cortes-Chavez et al.

3 Analysis of the Medical Device Design Process and Its Risks Currently the processes that patients undergo in medical contexts (diagnosis, monitoring and treatment) have become more dependent on the use of medical devices. This dependence implies the need of competent and skilled staff that can properly use these devices [1]. Different studies have reported multifactorial causes involved in the frequency and severity of adverse consequences due to misuse of medical devices. Aspects such as the lack of training or the failure to update medical personnel, old or deﬁcient equipment, and erroneous procedures contribute to the happening of negative events such as accidents [1]. The design of medical devices is one of the most complex design process due to the amount of variables that need to be considered such as technology, innovation time, norms, quantity of chemicals, sterilizable materials, and most importantly the users. The number of users that need to be considered is bigger than in other consumer products. In this study we will only consider seven of them, which are considered as the most important because of their role as user. The users considered in this study are: patient, family, general doctor, nurse, technical staff, maintenance staff and specialist. The design of medical devices requires the speciﬁcation of a big number of users and it is of utter importance to identify the priority between them to achieve a correct balance between their needs. Prioritizing the importance of each one of them with respect to the product should be a constant practice because it varies in each stage of the disease that the patient is developing. The challenge is to design a device that fulﬁlls all users and adapts correctly with the different users in each stage of the disease, these stages are shown in Fig. 1.

Fig. 1. Disease stages.

Currently, the design of medical devices tries to be functional in each stage. A medical device (product), should not take the risk of satisfying all users in all stages of the disease since it is almost impossible to achieve and this could create a large

Medical Device Design Challenges Based on Users Hierarchy

201

number of risks (mistakes). Figure 2 exempliﬁes how the development of a disease in two patients can differ greatly not only by avoiding one or several disease stages but by differing on the time spent on each stage as well, hence the consideration of an “all stages device” is risky.

Fig. 2. Disease development in two patients, stage and time variations on the whole process can be easily perceived.

The design of medical devices is complex due to the increase of variables in comparison to the consumer product. In order for the medical device to be efﬁcient in its functions, it is indispensable to deﬁne the users and the priority of use between them since the patient is not always the main user of the medical device. The users and their priorities vary as the patient’s health increases or decreases. The design of medical devices should be flexible to adapt to the level of disease of the patients as their health status progresses. Each patient is different and their process is not foreseeable, which is why the design of the product should be thought to be flexible in each change of stage. For this reason, we have developed a hierarchical scheme that prioritizes the user depending on the stage of the disease based on the previously mentioned ﬁndings. Figure 3 shows the priority scale proposed for this analysis and the hierarchical scheme that demonstrates how priorities vary in each stage and with each user. Besides the aforementioned scheme, a scheme that covers a maximum of two disease stages (Fig. 4) is also proposed to increase the efﬁciency of the products since it is practically impossible for a current product to be 100% efﬁcient, considering that users and their priorities change at each stage of the patient’s illness.

202

F. Cortes-Chavez et al.

Fig. 3. Application of the priority scale depending on the disease stage and the user experiencing it.

Fig. 4. Two disease stages coverage scheme. The (circle in red) shows the approach used nowadays, the (circles in green) show the new proposed scheme with focus on only two stages.

4 Conclusion There are four variables that need to be considered in the design of medical devices, (1) the amount and variety of users - priority of needs of each one of them, (2) the stages of the disease - necessities change, (3), the priority of users - varies on each stage, and (4) the time spent on each stage as well. Correlating these four variables forces medical device designers to perform an extensive study to effectively design the adaptability of the product in the medical

Medical Device Design Challenges Based on Users Hierarchy

203

context. This is why we propose a medical product design that considers solely a maximum of 2 disease stages in its use to reduce the risk, increase the medical device’s efﬁciency, and avoid the product satisfaction from being diminished.

5 Discussion The design of medical devices should consider the possibility of (1) being able to adapt in all stages of the disease, (2) be flexible in the use to consider the different types of users, (3) remember that in each stage the priority level of each user varies and this forces the product to adapt to each one of them and (4) the time spent on each stage of the disease varies. These four variables are currently not correlated during the design of medical devices and should be studied more thoroughly. This is a great challenge for the future of medical device design.

References 1. Clarkson, D.M.: Medical device guidebook: a browser information resource for medical device users. Med. Eng. Phys. 41, 97–102 (2017) 2. Hagedorn, T.J., et al.: A concept ideation framework for medical device design. J. Biomed. Inform. 55, 218–230 (2015) 3. Hagedorn, T.J., et al.: An information model to support user-centered design of medical devices. J. Biomed. Inform. 62, 181–194 (2016) 4. Kim, K., Lee, K.: Collaborative product design processes of industrial design and engineering design in consumer product companies. Des. Stud. 46, 226–260 (2016) 5. Martin, J.L., et al.: A user-centred approach to requirements elicitation in medical device development: a case study from an industry perspective. Appl. Ergon. 43(1), 184–190 (2012) 6. Martin, J.L., et al.: Capturing user requirements in medical device development: the role of ergonomics. Physiol. Meas. 27, 8 (2006) 7. Privitera, M.B., et al.: Human factors in the design of medical devices – approaches to meeting international standards in the European Union and USA. Appl. Ergon. 59, 251–263 (2017) 8. Schmettow, M., et al.: An extended protocol for usability validation of medical devices: research design and reference model. J. Biomed. Inform. 69, 99–114 (2017) 9. Wilkinson, C.R., De Angeli, A.: Applying user centred and participatory design approaches to commercial product development. Des. Stud. 35(6), 614–631 (2014)

Putting Chatbots to the Test: Does the User Experience Score Higher with Chatbots Than Websites? Amélie Beriault-Poirier1, Sandrine Prom Tep2(&), and Sylvain Sénécal3 2

1 imarklab, Montréal, QC, Canada ESG Business School, University of Quebec in Montreal, Montréal, QC, Canada [email protected] 3 HEC Montreal, Montréal, QC, Canada

Abstract. Chatbots are robots that simulate conversation with human users through instant messaging services. Though the technology is innovative and trendy, what kind of user experience do they provide? To answer this, we conducted an exploratory study with chatbots from different sectors of activity on Facebook Messenger. We invited ten participants to complete six precise search tasks using these chatbots and the corresponding websites from each brand. Based on observation and interviews according to a qualitative approach, our study shows that the user experience scores higher with websites than chatbots, while abandonment rate is higher with chatbots, even though they generate more positive emotions. As a conclusion, until artiﬁcial intelligence improves the technology, chatbots have some catching up to do as they are do not score higher than websites to fulﬁll users’ expectations. Keywords: Chatbots Websites

User experience Online consumer behavior

1 Introduction Chatbots are robots that simulate conversation with human users through instant messaging services like Facebook Messenger. In the last year, several brands launched chatbots and promoted them heavily through ad campaigns. For example, you can now chat with a virtual chef (in French) on Quebec’s pork producers Facebook page. Guess what? The virtual chef can suggest many recipes to integrate pork in your menu. Such marketing initiatives usually catch the interest of consumers, businesses and media. It’s trendy, innovative and cool technology-wise, but what kind of user experience do they provide? Are they easy to use, useful and more fun to use? Can chatbots make our life easier? To answer this, we conducted an exploratory study with chatbots from different sectors of activity on Facebook Messenger. We put chatbots from Whole Foods (food and beverage), Tommy Hilﬁger (clothing) and Skyscanner (travel) to the test. We invited ten participants to complete six precise search tasks using these chatbots. We asked them to do the same with websites from each brand. For example, © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 204–212, 2019. https://doi.org/10.1007/978-3-030-02053-8_32

Putting Chatbots to the Test

205

participants had to ﬁnd a salad recipe using Whole Foods’ chatbot and a salmon recipe using Whole Foods’ website. As mentioned earlier, this study is exploratory and was initiated backwards, starting with a practical question but without any literature review. It is important to note that a lot is written about chatbots in the managerial press in order to promote their tech innovativeness, but scarcely no scientiﬁc marketing research work exists yet on chatbots to assess neither the online service quality they provide nor the user’s satisfaction using them. The goal of this study hence it to conduct applied research and get ﬁrst answers on our practical questions [1]. In doing so, it will allow for a better understanding of the user experience with chatbots, and direction in future research work. In the next sections, we describe the qualitative approach used to conduct this study, and the three major ﬁrst-observed conclusions resulting from the data analysis.

2 Methodological Approach The research was conducted according to scientiﬁc principles, and here are described the essential elements from the methodological approach used for this study. We recruited ten university students through email and social media. We asked interested students to share their contact information for us to do a follow-up. We invited a total of four men and six women aged between 21 and 29 years old to participate in our study in Montreal. Participants were not related to the moderator and never met beforehand. Usability tests were conducted with individual participants. They lasted an hour each on average and took place from November 27th to 29th 2017. Participants did the following tasks without assistance like they would at home: Find… • • • • • •

A salad recipe to serve as starter using Whole Foods’ chatbot; A salmon recipe to serve as main course using Whole Foods’ website; A new coat using Tommy Hilﬁger’s chatbot; New black pants using Tommy Hilﬁger’s website; A flight to New York City for a weekend in March 2018 using Skyscanner’s chatbot; A flight to Paris for a week in May 2018 using Skyscanner’s website.

To avoid sequence bias in the study, we randomized task order from one participant to another to alternate brands (Whole Foods, Tommy Hilﬁger and Skyscanner) and tools (chatbots and websites). For example, some participants began searching recipes with Whole Foods’ chatbot while others began searching pants on Tommy Hilﬁger’s website. The data collection lasted 3 days. We collected the following data to compare websites to chatbots for a total of 60 observations (10 participants x 6 tasks): • • • • •

Time required to complete each task or before abandonment; Abandonment rate for each task (abandonments/participants); Responses to a user experience evaluation questionnaire after each task; Participants’ facial expression during each task; Qualitative feedback from participants about chatbots after tasks.

206

A. Beriault-Poirier et al.

To measure the user experience, we analyzed participants’ facial expressions using the Noldus FaceReader tool at the Tech3lab in HEC Montreal, and adapted the USE questionnaire [2] to evaluate usefulness, ease of use, ease of learning and satisfaction. Participants were asked to evaluate the extent to which they agreed or disagreed with eight statements. We used a 7-point Likert scale from 1 “Strongly Disagree” to 7 “Strongly Agree” to capture participants’ responses. We evaluated each dimension (e.g., usefulness) with two statements.

3 User Experience Is Better with Websites Than Chatbots After each task, participants evaluated usefulness, ease of use, ease or learning and their satisfaction on a scale of 1 to 7. Results indicate that participants ﬁnd websites more useful, easier to use and easier to learn than chatbots. They are also more satisﬁed with websites than chatbots. There is a statistically signiﬁcant difference between chatbots and websites for each dimension mentioned above. Conﬁdence level is 95% for ease of learning, and 99% for usefulness, ease of use and satisfaction (see Fig. 1).

Fig. 1. User experience evaluation by dimension

According to overall evaluation, participants like Whole Foods’ chatbot the most (Fig. 2). They are equally satisﬁed with user experience provided by its chatbot and website. However, they preferred Tommy Hilﬁger (Fig. 2) and Skyscanner’s websites (Fig. 3) than chatbots.

Putting Chatbots to the Test

Fig. 2. Whole Foods and Tommy Hilﬁger chatbots

207

208

A. Beriault-Poirier et al.

Fig. 3. Skyscanner chatbot and website

Participants like to enter ingredients they have at home to get recipe suggestions in a few seconds. Chatbots and recipe search seem like a winning combination. This is probably why two participants spontaneously said they would like Ricardo Cuisine to launch a chatbot… but we should get back to Whole Foods! You have a butternut squash on hand and want to serve a salad as a starter for the Holidays? Done. Whole Foods’ chatbot (Fig. 2) satisﬁed our participants quite quickly but with Tommy Hilﬁger, something was missing. Participants felt they could not see the whole product selection to ﬁnd the best product for their needs. According to them, search options are too limited with the chatbot. They ﬁnd search ﬁlters easier to use on Tommy Hilﬁger’s website. It should be noted that the brand’s chatbot seems programmed to make very broad recommendations. For example, here is what I get when I ask for a “black party dress”. The chatbot understands I want a dress but nothing more. If I get angry by adding an exclamation mark, it still shows me the same dresses all over again.

Putting Chatbots to the Test

209

4 Abandonment Rate Is Higher with Chatbots As researchers, our goal is always to create a setting that accurately reflects a real use scenario. With this in mind, we asked participants to navigate just like they would at home. Thus, some abandoned one or several tasks because they couldn’t ﬁnd what they were looking for. Table 1 shows the abandonment rate for each task. Table 1. Abandonment rate

Seven out of ten participants abandoned one or two tasks with a chatbot while only three abandoned a task on a website. This is due to the fact that some tasks are longer to complete with a chatbot. Participants took an average of 3:04 min to complete a task with a chatbot or decide to give up and 1:39 min to complete a task on a website. On average, they need 85% more time to complete a task when using a chatbot (See Table 2). Table 2. Time for task completion

Whole Foods stands out once more. Participants take almost the same amount of time to complete a task using its chatbot and website. They take twice as much time with Tommy Hilﬁger and Skyscanner’s chatbot than website because there are more steps to go through. Let’s have a look at Skyscanner’s case. To search for plane tickets on Skyscanner. ca, you need to give basic information (from, to, when and who), click “Search flights” and voila. You are done. It is not as quick with Skyscanner’s chatbot. You need to communicate each element one after the other to be understood. This makes the process much longer.

210

A. Beriault-Poirier et al.

Besides additional steps, the fact that chatbots have limited understanding of human language annoyed participants. It even caused some of them to abandon tasks. They feel chatbots do not understand them and do not communicate clearly what information they want. For example, Skyscanner’s chatbot asks users when they want to travel but does not specify which date format it accepts. Such imprecision forced participants to repeatedly enter the same information before ﬁguring out what was expected.

5 Chatbots Generate More Positive Emotions We analyzed participants’ facial expression with FaceReader, a facial expression analysis software. Speciﬁcally, we looked at the average positive or negative emotion intensity during a task [3]. It is called emotional valence and varies from –1 to 1. A –1 valence indicates a 100% negative emotion, 1 a 100% positive emotion and 0, a neutral reaction. If you gave up on chatbots at this point, you will be surprised to ﬁnd out they generated more positive emotions than websites. Participants’ average emotional valence is –0,26 with chatbots compared to –0,37 with websites. As for user experience, there is a statistically signiﬁcant difference between chatbots and websites at a 95% conﬁdence level. However, this time, chatbots did better than websites. Average valence is negative in both websites and chatbots (See Fig. 4). This is normal considering online search is not particularly hedonic. In most utilitarian contexts, emotional valence is negative or very close to neutral. Our results show that participants expressed more positive emotions while using chatbots. Here, Whole Foods is no exception and results are similar for all three chatbots and websites.

Fig. 4. Emotional valence by task

Websites did better than chatbots in terms of user experience and success rate. Talking to a “person”, whether robot or human, is generally more pleasant than using a website interface. It is one of chatbots’ strengths. Based on these results, we can

Putting Chatbots to the Test

211

presume the more humanlike a chatbot is, the more positive emotions it will generate. For example, Quebec’s pork producers’ virtual chef is quite different from the chatbots we tested. It has a name (Alex) and a face (Fig. 5).

Fig. 5. Alex: Quebec’s pork virtual chef

6 Conclusion: Chatbots Have Some Catching up to Do Getting a quick and precise answer is what participants like the most about using chatbots. However, they ﬁnd them limited and are irritated by not being understood right away. Despite these shortcomings, they still believe they could use chatbots in the future. They assume chatbots will become more intelligent and reactive as technology improves. On a scale of 1 to 7 where 1 means “Not at all likely” and 7 “Very likely”, they rate their intention of using chatbots in the future at 3,7/7 on average. While artiﬁcial intelligence is evolving, chatbots have potential to offer a pleasant and customized experience to Internet users. They can also generate positive emotions. Of course, there is still work to do before consumers widely adopt them. Chatbots need to become at least as intelligent as websites. Search engines are still the better option. Marketers should keep in mind that consumers refer to websites as the benchmark when it comes to online search. For example, by telling Alex (Quebec’s pork producers’ virtual chef) that I am a vegetarian, I expected it to react like a website. I expected it to say it can only provide pork recipes or that it does not have recipes for me. Unfortunately, it does not understand me. The website does. When I search for “vegetarian recipe” on Quebec’s pork site, the search engine tells me there is no results. We should keep in mind that the average Internet user is used to websites and that chatbots are still a new concept. That being said, familiarity could partly explain why participants prefer the user experience on websites. Chatbots are particular because they provide information in a Question-Answer format. Websites give information in a different manner. They provide a lot of information at the same time with a variety of tools to help users ﬁnd what they need (e.g., menus, search engines, etc.). As a managerial takeaway form our study, it seems like chatbots have even more room for

212

A. Beriault-Poirier et al.

improvement for precise and utilitarian search and could eventually outdo websites in that situation. This is what the high success rate for recipe search using Whole Foods’ chatbot and low success rates for more precise tasks tend to tell us. As a follow-up to this research, it would be interesting to compare emotions generated by chatbots like Alex to emotions generated by less “human” chatbots. As a conclusion, chatbots have some catching up to do as they are not better than websites to fulﬁll users’ expectations. However, we believe this might certainly change in the future, with artiﬁcial intelligence improvements lying ahead.

References 1. Dabholkar, P.A.: Consumer evaluations of new technology-based self-service options: an investigation of alternative models of service quality. Int. J. Res. Mark. 13(1), 29–51 (1996) 2. Lund, A.M.: Measuring usability with the USE questionnaire. Usability Interface 8(2), 3–6 (2001) 3. Vasileios, T., Moridis, C.N., Economides, A.A.: Measuring instant emotions based on facial expressions during computer-based assessment. Pers. Ubiquitous Comput. 17(1), 43–52 (2013)

IHSED 1: User Interface Design Applications and Human-Systems Integration

Making Multi-team Systems More Adaptable by Enhancing Transactive Memory System Structures – The Case of CDM in APOC Dirk Schulze Kissing1(&), Carmen Bruder1, Nils Carstengerdes2, and Anne Papenfuss2 1

Institute of Aerospace Medicine, Department Aviation and Space Psychology, Deutsches Zentrum für Luft-und Raumfahrt (DLR), Sportallee 54a, 22335 Hamburg, Germany {Dirk.Schulze-Kissing,Carmen.Bruder}@dlr.de 2 Institue of Flight Guidance, Department Human Factors, Deutsches Zentrum für Luft-und Raumfahrt (DLR), Lilienthalplatz 7, 38108 Brunswick, Germany {Nils.Carstengerdes,Anne.Papenfuss}@dlr.de

Abstract. The DLR project ‘Inter Team Collaboration’ (ITC) aims to provide systems engineers with tools and human factors concepts that allow a systemic access to the social side of socio-technical systems. A main design question for implementing Collaborative Decision Making (CDM) in APOC is how to induce collaborative decision making in a dynamic environment of ATM to make it more adaptive and resilient. Our main assumption is that the establishment of a Transactive-Memory System (TMS) is the basic predisposition for a successful implementation of intensive CDM. A TMS reflects linkages across MTS boundaries. Assumedly, its emergence is a function of social structures (like motives), but also of communication structures. The MTS is conceptualized as a nonlinear dynamical system (NDS), where CDM is conceived as an attractor to system-behavior. Recurrence analyses on behavioral data assessed within Human-in-the-Loop-experiments will be applied to identify MTS transition phases in reaction to perturbations. Keywords: Human factors Systems engineering APOC Collaborative Decision Making Multi Team System Transactive Memory System Nonlinear Dynamic System

1 Introduction Airports are complex systems with multiple interrelated processes and multiple process owners who pursue their focal goals and have their own perspective on the system [1]. With Airport Collaborative Decision Making (A-CDM) a concept that promotes information-sharing was introduced. It was assumed that by providing stakeholders with an integrated view of airport operations, coordination of airport processes is enhanced. But soon it became obvious that information sharing alone does not sufﬁce to optimize stakeholder coordination, as their processes are still planned without knowledge about intentions, goals and actions of other (cooperating or competing) © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 215–220, 2019. https://doi.org/10.1007/978-3-030-02053-8_33

216

D. Schulze Kissing et al.

parties at the same airport [1]. Partly conflicting goals of airport stakeholders and their non-sharing of all information about their plans makes it difﬁcult to assess the impact of a decision for actions to the other’s plans, and vice versa [1]. So the A-CDM concept was developed into a solution called Total Airport Management (TAM). TAM develops ideas for intensive collaborative decision making above the level of information sharing in Airport Operation Centers (APOCs) (see [2] for detailed description of APOC processes and use cases). In Human Factors literature CDM is regarded to be the essence of team performance [3]. But implementing highly collaborative CDM like in the APOC concept comes along with challenges and limitations. One major reason identiﬁed by [4] is low trust between the (in part competing) CDM entities. However, trust issues were observed to reduce with collaboration experience, insofar the collaboration process unfolded in a positive manner resulting in collaborative beneﬁts, like providing an airline with a greater ability to control the allocation of unavoidable delay among their flights (see [5]). But also other collaboration barriers are reported, related to differences in viewpoints because of different process views and because of different interests [3]. Another main design concern for implementing highly collaborative CDM is how to ensure its adaptivity and resilience in the dynamic environment of ATM, with its sometimes turbulent transitional phases. Adaptivity is conceived as the capacity of a work system to achieve its goals, despite the emergence of circumstances that perturb it from a predetermined course by pushing it towards the boundaries of its competence envelope [6]. How can the work system APOC, which is designed for enhancing adaptivity of the airport system, succeed in transition situations created by perturbations, inducing time pressure for action? Does it show stability in its already occupied parameter space? [6] What happens if the APOC work system is pushed beyond its competence envelope? Is it capable of emitting resilience, so it can achieve a new state of stability? [6].

2 A Conceptual Framework for Inter-team Collaboration Analysis Taking over a systemic view, regardless of the scale of analysis, may it be a single user tool or a whole socio-technical system, is regarded to be a main characteristic of high quality Human Factors [7]. Human Factors should be embedded in system design, especially in the design of socio-technical-systems with a systems engineering approach (cf. [8]). [9] deﬁnes Cognitive Systems Engineering as ‘a speciality discipline of systems development that addresses the design of socio-technical systems’ (p. 1). And [8] claims: ‘Understanding and applications of theory of complex adaptive and dynamical systems should signiﬁcantly improve the effectiveness of human-centered design efforts in large systems-of-systems’ (p. 983). The aim of the DLR project ‘Inter Team Collaboration’ (ITC) is to provide systems engineering with tools and human-factors concepts that support systems thinking about the social side of these socio-technical systems. Three cornerstones of the systemic human-factors analysis approach of ITC to the design of a socio-technical system like APOC are proposed: (a) contextualising APOC within a Multi-Team System

Making Multi-team Systems More Adaptable by Enhancing TMS

217

(MTS) (b) introducing Transactive Memory System (TMS) as a moderating variable ‘between the heads’, and (c) conceptualizing APOC as a Nonlinear Dynamic System (NDS) (cf. [10]). 2.1

Contextualization: APOC Within a Multi Team System

To our view a concept from organizational research, the Multi-Team System (MTS) promises to be a fruitful ﬁrst cornerstone of our systemic human factors approach to the design of the socio-technical system that is laid down in the APOC operational concept. Providing deeper analysis into its social context of use is deemed necessary to fully assess the impact of this technical solution for CDM. [11] provide an analysis on the (in-) efﬁciencies of collaborative problem solving tasks when members of teams are from distinct, yet interdependent organizations. Referring to [12] they characterize ATM as a distributed problem solving system in which a large number of decision makers are distributed across organizations and thus often have conflicting and competing goals. [12] emphasize that it is the data and the knowledge about the system distributed among the decision makers (which is inherently incomplete because of the characteristics of complex systems; cf. [13]) ‘that ties the very complex, highly distributed system together’ ([12] p. 100). Their conceptualization of the social context of use of ATM systems parallels the MTS construct proposed by [14]. [15] state that MTS are distinguished by being smaller than their embedding organizations, with outcome-interdependent teams sharing at least one common distal goal. However, when teams embedded within different organizations are working on a common goal, they have to work across the system boundaries of these organizations. MTS that are cross-boundary are likely to have less shared value and motivation to collaborate than ‘internal’ MTS, with teams coming from one organization1 [15]. These potential issues have to be addressed in the systemic humanfactors analysis of APOC. Research shall yield to ﬁnd answers to questions like: How reciprocal or intense is the interdependence within APOC? How close do stakeholders work in the assumed ‘intense’ collaboration to accomplish shared goals? What intensity is possible and advisable? The APOC can be conceived as a process interdependent MTS, where by deﬁnition component teams share several functions necessary for effective collective action [cf. [15]. This should encompass boundary spanning activities, environmental sense making, task ordering and tactical planning, communicating key information, the timing and coordination of sequential and synchronous actions, and the monitoring and backup of MTS actions (cf. [14]). For experimental studies in ITC a speciﬁc coding system for communication acts will be provided to assess when each of these activities and processes emerge within APOC during CDM. As the analysis will be on the systemic level, boundary spanning activities, as a construct from the Open-Systems Paradigm, will come into research focus. Boundary spanning is a concept that encompasses activities at 1

The topic of trust between teams in (internal) MTS is raised by [16]. They contextualize teams by pointing out that within an organization teams are dependent on information, products, or services from other teams, ‘which must be trusted to take their intentions into consideration and constructively resolve potential conflicts of interest’ [16] p. 276.

218

D. Schulze Kissing et al.

the interface between organizational units, from simple information exchange to complex and real-time behavior integration and coordination [17]. Boundary spanning activities speciﬁcally aim at real-time, mutual adjustment between focal teams within an MTS [17]. Within this conceptual framework the APOC can be understood as a tool for the necessary boundary spanning activities within the MTS at airports. 2.2

Introducing Transactive Memory System (TMS) as a Moderating Variable

TMS is the second cornerstone of our systemic human factors approach to the design of the socio-technical system that is layed down in the APOC operational concept. It guides to the mitigation approach followed in ITC. TMS is an emergent property of collaborating cognitive systems. A TMS can be mediated by social structures, like the motives within focal teams to exchange information and to collaborate (for more information cf. [18]). TMS emergence should also be a function of communication structures, modality of communication, fluidity of communication linkages among focal teams, as well as a function of the MTS formal state, from early to mature According to [19]’s conceptualization a TMS is a dynamic form of cognitive processes unfolding on an inter-individual level which promotes the organized differentiation of knowledge within the social system (the TMS structure) by transactive encoding, storage, and retrieval processes. To their view, a TMS is not limited to a socially shared cognition of ‘who knows what’. They propose that teams with good TMS show higher performance than teams that simply show a shared understanding of who knows what. They also assume that performance is a function of the degree to which the group’s knowledge is differentiated. It is this distributed knowledge which [11] assume to be the glue that ties an MTS together. In ITC this central proposition (the higher the quality of TMS structure and processes, the better the performance) will be tested for the APOC social system. A central question will be: Do TMS metrics show an effect size large enough to signiﬁcantly influence APOC performance? We propose that, like information sharing on the technical side, the establishment of a TMS is the basic predisposition for successful ‘intensiﬁcation’ of CDM on the social side. 2.3

Conceptualizing APOC as a Nonlinear Dynamic System

The third cornerstone of our systemic HF analysis approach is a methodological one. Assuming that MTS are best conceptualized as dynamical-systems, recurring patterns of MTS behavior will come under close scrutiny. For [20] the underlying theoretical concept of dynamical systems simply introduces a temporal element (not only the occurrence of behavior, but also its recurrence) for understanding systems behavior. Recurrence analysis has already been assessed to examine group dynamics (e.g. [21]). According to [21] recurrence analysis (see [23]) is a technique with roots in the systems dynamics literature that was developed to reveal the properties of complex, nonlinear systems. A number of articles have shown that recurrence analysis can be useful for

Making Multi-team Systems More Adaptable by Enhancing TMS

219

understanding the emergence of coordination in dyads (e.g. [23]), and even in small groups (see [24]). Our method to assess system dynamics is the recurrence quantiﬁcation analysis [22]. Based on behavioral data recorded in HitL-Simulations hypotheses about emerging recurrence patterns within APOC that indicate adaptability (i.e. search for stability in an already occupied parameter space) or resilience (achievement of new states of stability) will be tested.

3 Outlook As ITC is a human factors project which by deﬁnition should be design driven, another challenge will be the deﬁnition of metrics, i.e. thresholds or decision points used to specify value judgements [6] to evaluate our current (as well as prospective) sociotechnical system design. The state of work will be presented in upcoming paper. In next steps experimental studies in laboratory settings within a control-center task environment, as well as large scale HitL-simulations of ‘CDM in APOC’ scenarios will be designed to assess TMS emergence within MTS, and how this might enhance CDM adaptability during system transition phases. According to [10] there are many types of change that systems can produce. They are represented by numerous modelling structures such as attractors, self-organization, emergence and synchronization. A future goal is to have a model of the APOC social system, serving as a representation to derive testable hypotheses from, to validate model assumptions and modify the model guided by empirical evidence. The CDM will be conceived as an attractor to system-behavior, and recurrence analyses of continuously assessed behavioral metrics will be performed to identify MTS transition phases in reaction to perturbations.

References 1. Papenfuss, A., Carstengerdes, N., Schier, S., Günther, Y.: What to say when: guidelines for decision making. An evaluation of a concept for cooperation in an APOC. In: Proceedings of the 12th USA/Europe Air Trafﬁc Management Research and Development Seminar (ATM2017) (2017) 2. Bogers, H., Linde, M., Matas Sebatia, J.R., Álvarez Escotto, I., Martin Espinosa, J.I., Ciprián Tejero, B., et al.: D16 - OFA05.01.01 Consolidated OSED Edition 3, Part 2, Brussels, Belgium (2015) 3. Coury, B.G., Terranova, M.: Collaborative Decision Making in Dynamic Systems, 35th edn, pp. 944–948. SAGE Publications, Sage (1991) 4. Vail, S., Churchill, A., Karlsson, J., McInerney, T., Domitrovich, J., Phillips, T.: Guidebook for Advancing Collaborative Decision Making (CDM) at Airports (2015) 5. Ball, M.O., Chen, C.Y., Hoffman, R., Vossen, T.: Collaborative decision making in air trafﬁc management: current and future research directions. In: New Concepts and Methods in Air Trafﬁc Management, pp. 17–30. Springer (2001) 6. Hoffman, R.R., Hancock, P.A.: Measuring resilience. Hum. Factors 59(4), 564–581 (2017)

220

D. Schulze Kissing et al.

7. Dul, J., Bruder, R., Buckle, P., Carayon, P., Falzon, P., Marras, W.S., Wilson, J.R., van der Doelen, B.: A strategy for human factors/ergonomics: developing the discipline and profession. Ergonomics 55(4), 377–395 (2012) 8. Karwowski, W.: A review of human factors challenges of complex adaptive systems: discovering and understanding chaos in human performance. Hum. Factors 54(6), 983–995 (2012) 9. Lintern, G.: Cognitive Systems Engineering, 15 August 2017. https://pdfs.semanticscholar. org/ca96/1aa540819e0783514ab58bbdf035c877df74.pdf 10. Guastello, S.J.: Nonlinear dynamical systems for theory and research in ergonomics. Ergonomics 60(2), 167–193 (2017) 11. Obradovich, J.H., Smith, P.J.: Problem solving in a distributed collaborative environment: The necessity of shared knowledge within the air trafﬁc management system, pp. 100–104 (2002) 12. Billings, C., Smith, P.J., Woods, D.D., McCoy, E., Denning, R., Sarter, N., Dekker, S.: Advanced air transportation technologies: Problem deﬁnition and exploration of a solution space. Human-centered technologies and procedures for future air trafﬁc management, Report, Contract No. NAG2–995 (1997) 13. Funke, J.: Complex problem solving. In: Encyclopedia of the Sciences of Learning, pp. 682– 685. Springer (2012) 14. Mathieu, J.E., Marks, M.A., Zaccaro, S.J.: Multi-team systems. In: International Handbook of Work and Organizational Psychology, vol. 2, no. 2 (2001) 15. Zaccaro, S.J., Marks, M.A., DeChurch, L.A.: Multiteam systems: an introduction. In: Multiteam Systems, pp. 18–47. Routledge (2012) 16. Somech, A., Khalaili, A.: Team boundary activity: its mediating role in the relationship between structural conditions and team innovation. Group Organ. Manag. 39(3), 274–299 (2014) 17. Davison, R.B., Hollenbeck, J.R.: Boundary spanning in the domain of multiteam systems. In: Multiteam Systems: An Organization Form for Dynamic and Complex Environments, pp. 323–362 (2012) 18. Kanfer, R., Kerry, M.: Motivation in multiteam systems. In: Multiteam Systems: An Organization Form for Dynamic and Complex Environments, pp. 81–108 (2012) 19. Lewis, K., Herndon, B.: Transactive memory systems: current issues and future research directions. Organ. Sci. 22(5), 1254–1265 (2011) 20. Gorman, J.C., Dunbar, T.A., Grimm, D., Gipson, C.L.: Understanding and modeling teams as dynamical systems. Front. Psychol. 8, 1053 (2017) 21. Knight, A.P., Kennedy, D.M., McComb, S.A.: Using recurrence analysis to examine group dynamics. Group Dyn. Theor. Res. Pract. 20(3), 223 (2016) 22. Webber, C.L., Marwan, N.: Recurrence Quantiﬁcation Analysis: Theory and Best Practices. Springer International Publishing (2015) 23. Fusaroli, R., Tylén, K.: Investigating conversational dynamics: interactive alignment, Interpersonal synergy, and collective task performance. Cogn. Sci. 40(1), 145–171 (2016) 24. Schulze Kissing, D., Bruder, C.: Relating gaze patterns to assess coordination within small groups, Research Report 2018. DLR Institute of Aerospace Medicine (2018, to be published)

Deﬁning User Needs for a New Sepsis Risk Decision Support System in Neonatal ICU Settings Through Ethnography: User Interviews and Participatory Design Richard Harte1,2(&), Leo R. Quinlan2,3, Evismar Andrade1,2, Enda Fallon4, Martina Kelly4, Paul O’Connor5, Denis O’Hora6, Patrick Pladys7,8, Alain Beucheé7,8, and Gearoid ÓLaighin1,2 1

Electrical and Electronic Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland [email protected] 2 Human Movement Laboratory, NUI Galway, University Road, Galway, Ireland 3 Physiology, School of Medicine, NUI Galway, University Road, Galway, Ireland 4 Mechanical Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland 5 General Practice, School of Medicine, NUI Galway, University Road, Galway, Ireland 6 School of Psychology, NUI Galway, University Road, Galway, Ireland 7 Centre Hospitalise Universities de Rennes (CHU Rennes), rue Henri Le Guilloux, Rennes, France 8 Faculté de Médicine de l’Université de Rennes, Rennes, France

Abstract. The diagnosis of late onset sepsis in neonates is complex and therefore usually late, resulting in increased risks. The Digi-NewB project proposes a novel solution to this problem, by designing a non-invasive Decision Support System (DSS) which will use vital signs, images and sounds to measure the risk of sepsis in the preterm infant and therefore support clinicians in diagnosis. The introduction of any new system to a neonatal intensive care unit (NICU) environment presents a challenge for designers who must account for a technology laden environment and a demanding work-load for clinicians. To deﬁne the user needs and therefore build the ﬁrst use cases for such a system, a multi-method approach was adopted and is described in this paper. This approach consisted of a period of ethnography, eleven semi-structured interviews and the application of a prototyping exercise based on the principles of participatory design. Keywords: User-Centered Design Neonatal

Participatory design Codesign

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 221–227, 2019. https://doi.org/10.1007/978-3-030-02053-8_34

222

R. Harte et al.

1 Introduction Neonatal sepsis contributes substantially to neonatal morbidity and mortality, and is an ongoing major global public health challenge [1]. The condition can cause signiﬁcant respiratory and neurological problems and is responsible for 13% of all neonatal deaths, and 42% of deaths in the ﬁrst week of life [2]. Preterm infants are a group who are particularly vulnerable to sepsis. There are 300,000 preterm infants born each year in the EU who are at a 25% risk of developing late onset sepsis. The diagnosis of late onset sepsis is complex and is difﬁcult to detect early in its evolution, resulting in increased risks for the infant. The Digi-NewB project [3] proposes to design a non-invasive Decision Support System (DSS) which will use traditional vital signs, novel imaging techniques and audio analysis to measure the risk of sepsis in the preterm infant and therefore support clinicians in making an early diagnosis. The introduction of any new system to the neonatal intensive care unit (NICU) environment presents a challenge for designers who must account for a technology laden environment and a demanding work-load for clinicians [4]. The Digi-NewB project is following a three-phase user-centered design (UCD) process. This process is illustrated in Fig. 1. In UCD, design solutions are presented to the end-users at various stages throughout the process and the design is iterated based on their feedback [5]. In participatory design, the act of ‘making’ is brought to the fore of the process through the use of probes, toolkits or prototyping [6]. In this study, we describe the User Research Phase of our process where we seek to understand the problem of the detection of sepsis from the perspective of potential endusers and engage the users in the act of ‘making’. Therefore the end-user becomes not just a design subject and a design participant; but also becomes a design partner or ‘codesigner’ [7]. The feedback from these activities will be used to develop the ﬁrst Digi-Newb system Use Case (Fig. 1) which will consist of task-flow descriptions, storyboards and user proﬁles [8]. The development of Use Cases is a well-established method to deﬁne, organize and prioritize user requirements in medical device development [9].

Fig. 1. This paper describes the User Research Phase (Yellow Bubble) of the Digi-NewB UCD Process

Deﬁning User Needs for a New Sepsis Risk Decision Support System

223

2 Methodology A three stage methodology was utilized which merged traditional qualitative user research methods such as ethnography and user interviews, with codesign prototyping. The entire study took place onsite in Centre Hospitalier Universitaire (CHU) de Rennes and CHU de Angers in Brittany, France over the course of 11 days with full ethical consent from the relevant hospital boards. The three stages were as follows: Ethnography. A team of two researchers carried out immersive, focused ethnography across six shifts (60 h in total) in the (Neonatal Intensive Care Unit) NICU of each hospital, observing and shadowing 25 different clinicians (nurses and doctors) to gain a ‘day in the life’ perspective of their routine work. The researchers sought to understand the work-flow of the clinical teams, their approach to patient care and how they interacted with different technology in the unit. Written notes and video/audio recordings were taken and transcribed. The researchers observed in a ‘fly on the wall’ manner, and when necessary would intervene to ask speciﬁc questions of the staff, particularly after a certain procedure (during a period of downtime) had been carried out or an incident had occurred. For example “could you explain why it is necessary to carry out that procedure and what are the conditions that lead to such a procedure?” Interviews. From the observations in Stage 1, a list of seven questions was generated (Table 1). Semi-structured interviews were conducted with eleven clinicians (6 nurses, 3 neonatologists, 1 pediatrician and 1 intern doctor). The questions were asked in the same order to all interviewees, participants were encouraged to return to certain topics if they recalled something later in the interview.

Table 1. Questions generated from ethnography which were asked in interviews 1. 2. 3. 4. 5.

What is your experience of sepsis in the unit? Do you consider it a big problem? What are the current procedures for diagnosing or monitoring sepsis? Is there one strong indicator of sepsis occurring or something you look for straight away? How could the procedures be improved by technology? How do you think the information will be presented to the user? In terms of interfaces type, number of interfaces? (Reference Map of Reanimation Unit) 6. Do you think Digi-NewB would require an alarm system? 7. What do you think of Digi-NewB having the ability to present live video feeds from each chamber to a computer in the central station?

Codesign Exercise. Based on the principles of participatory design, the same (stage 2 interviewees) eleven clinicians participated in a generative exercise where they created interface prototypes of what potential information they would like to see on a DigiNewB user interface. It was intended that these prototypes would give form to ideas, and allow designers and developers to explore with participants what interface elements might be feasible. The participants made sketches on mock paper screens using colored pens and pencils and were encouraged to discuss their thought process while sketching

224

R. Harte et al.

such that the researchers could understand the information needs of the users [6], and why they felt they needed or did not need certain interface features, elements or functions.

3 Results Ethnography. The NICU was observed to be a challenging environment for clinicians, who while providing care must work with the demands of various interfaces, such as patient monitors. In the NICU Level 3 (most critical care level), there could typically as many as 4 interfaces per patient illustrating the already heavy burden of information that workers must negotiate. The nurses’ routine depends on the needs of the patient(s) assigned to them. Each nurse may be assigned 2–4 infants and they check each infant every 1–4 h, depending on what needs to be done in terms of feeding, cleaning and moving, and the current health status of the patients. Alarms are extremely prevalent in the neonatal unit, with false alarms a regular occurrence. Sepsis is a major preoccupation for staff, who must exercise caution when using invasive devices such as catheters and feeding tubes. Interviews. When diagnosing sepsis, clinical teams process a large amount of information from different sources in order to make a clear decision on treatment. Staff noted that the way in which signs related to sepsis diagnosis are currently examined is discontinuous. It relies on the experience of the clinician to see metrics at one particular point in time and make a concrete judgement based on the available data. Not every observer will have the same level of experience which leaves the door open for different outcomes and delays. In some cases, the doctors are required to make a decision based on subtle behaviors that the baby exhibits that are often only observed with many years of experience. Even an experienced senior doctor may only make 3 or 4 rounds a day, so is not in constant contact with patients to always make effective decisions. The difﬁculty with sepsis was summed up by one nurse: “There is no ideal way to diagnose sepsis. The biomarkers (from blood tests) which we have at the moment are not very sensitive or speciﬁc. In order to detect a problem like sepsis, a change in behavior of the baby needs to be detected. Behavioral signs are often picked up ﬁrst by the parents and nurses who will inform the doctor. These signs may include the baby being less active, a change in sleeping pattern or different sound activity from the baby. When the baby is septic the evolution can be very quick and dangerous, requiring quick prescriptions of anti-biotics.” Clinicians expressed a need for a system which could present clear and concise information on the patient’s sepsis risk. The potential ability to use technology to both monitor the past trend of sepsis risk and the related vital signs as well as looking at the current metrics in real time as a way to acquire an early diagnosis was greeted enthusiastically by both nurses and doctors. However, the clinicians, in particular the NICU nurses, expressed concerns about introducing another interactive system into an environment where workload and interaction with technology was already pushed to its limits. There was also conflicting opinions about the location of interfaces, 8 of the 11 interviewees expected that it would only be necessary to have one central interface per

Deﬁning User Needs for a New Sepsis Risk Decision Support System

225

unit at the nurse’s station and therefore avoid adding further clutter to rooms and bedsides. The remainder preferred a personalized interface for each patient at the bedside. There was also a ﬁrm divide over the perceived usefulness of having a camera feed coming from the room to a central station monitor as part of the system, due to concerns over data protection and privacy. This mixed feedback underlined the challenge for the design team who must address the concerns and requirements of the different stakeholders. Codesign Exercise. Each of the 11 participants provided a series of sketches of what they think a sepsis risk monitoring interface would look like. All were in agreement that the interface would need to be on a screen which would show the sepsis risk in a clear and concise manner, easily readable by those with varying levels of experience and which allows the user to explore the data in further detail if needed. How data is to be presented and visualized was one of the main themes of the sessions. Figure 2 shows some examples of the sketches that were provided during the sessions.

Fig. 2. Prototypes generated by end-users in the codesign stage

Without any prompting as to what they should sketch, participants began to draw their preferred data visualization style. For example, some preferred bar charts over line graphs, others preferred numerical tables, while others still preferred pie charts. Some preferred the use of color codes to illustrate changing risks while other preferred the use of numerical risk indexes or probabilities. The preferences may reflect their experience with other similar technologies or their level of training. During the ethnography it was made clear that nurses and doctors rarely rely on one metric or piece of information. This was supported by one doctor during the codesign activity; “I think all caregivers realize that there are many confounders that can vary all of these so-called markers, so I think they won’t be conﬁdent in just one number without trying to understand what generated it and become more conﬁdent in the information.” This sentiment is reflected in the sketches in Fig where each prototype example contains multiple blocks of data using different visualizations and forms of notiﬁcation.

226

R. Harte et al.

4 Discussion and Conclusion Based on the data from this three stage process, we now have a more complete picture of how the Digi-NewB System might function in the current user workflow. Taskbased scenarios were developed and were described in the form of storyboards (Fig. 3) to help the design team visualize the different tasks and scenarios that a user might be expected to carry out with the system.

Fig. 3. Storyboards were created showing the current workflow of nurses. Illustrating the workflow in this manner was important in order to understand how the Digi-NewB system would ﬁt into the current NICU workflow. A: Nurse observes changing risk index on bedside monitor(s) B: The nurse informs the shift doctor C: The doctor investigates further

The Use Case was exposed back to the same group of end-users from this study, such that the designers could verify if the stories, tasks and scenarios had been accurately captured and if the end-users agreed that this is how the Digi-NewB system would be used. This revised Use Case will now serve as the basis for the next stage of prototyping and making. This study has demonstrated the combined use of usercentered design methods such as ethnography and user interviews, and participatory design methods such as prototyping. The result is an in depth understanding of the problem of sepsis detection within the NICU context of use, and the preferences and needs of the typical end-users. The ethnography and interviews allowed designers to gain an empathic understanding of the difﬁcult context in which nurses and doctors work within the NICU. The large quantity of complex and urgent work involved in caring for preterm infants was observed in conjunction with the potential for tragedy on a daily basis. The study has also demonstrated the different preferences and needs which even seemingly uniform users profess. With this in mind, the use of the codesign prototyping exercise has allowed users to illustrate and visualize their needs and has provided the ultimate stakeholder, the caregiver, with an immediate stake in the design of this system, thereby treating the primary end-user as design partners rather than design subjects.

Deﬁning User Needs for a New Sepsis Risk Decision Support System

227

References Qazi, S.A., Stoll, B.J.: Neonatal sepsis: a major global public health challenge. Pediatr. Infect. Dis. J. 28(1 Suppl), S1–S2 (2009) Zea-Vera, A., Ochoa, T.J.: Challenges in the diagnosis and management of neonatal sepsis Challenges in the diagnosis and management of neonatal sepsis. J. Trop. Pediatr. 61(1), 1–13 (2015) Värri, A., Kallonen, A., Helander, E., Ledesma, A., Pladys, P.: The Digi-NewB project for preterm infant sepsis risk and maturity analysis. 10(2–3), 330–333 (2018) Melton, K.R., Ni, Y., Tubbs-Cooley, H.L., Walsh, K.E.: Using health information technology to improve safety in neonatal care: a systematic review of the literature. Clin. Perinatol. 44(3), 583–616 (2017) Harte, R., et al.: A human-centered design methodology to enhance the usability, human factors, and user experience of connected health systems: a three-phase methodology. JMIR Hum. Factors 4(1), e8 (2017a) Debrah, R.D., de la Harpe, R., M’Rithaa, M.K.: Design probes and toolkits for healthcare: identifying information needs in African communities through service design. Des. J. 20(1 Suppl), S2120–S2134 (2017) Sanders, E.B.-N., Stappers, P.J.: Probes, toolkits and prototypes: three approaches to making in codesigning. CoDesign 10(1), 5–14 (2014) Harte, R., et al.: Human-centered design study: enhancing the usability of a mobile phone app in an integrated falls risk detection system for use by older adult users. JMIR Mhealth Uhealth 5(5), e71 (2017b) Yuan, M.J., Finley, G.M., Long, J., Mills, C., Johnson, R.K.: Evaluation of user interface and workflow design of a bedside nursing clinical decision support system. Interact. J. Med. Res. 2(1), e4 (2013)

FLOW: A Software Application Designed to Help Older Adults Build Distance Interaction Wonsil Jang(&), Stephen Gilbert, and Sunghyun Kang Iowa State University, Ames, IA 50011, USA {swiriwon,gilbert,shrkang}@iastate.edu

Abstract. Lack of communication methods due to older adults’ physical and mental decline produce a signiﬁcant communication gap with friends and family, which creates social and mental distance with them. Also, relatively few studies have been dedicated to developing social media technologies that are older-adult-friendly. The main goals of this initial study were (1) to develop a prototype of story-based communication application based on a touch-based tablet application; and (2) to assess the prototype design using a mixed-method approach. Although anecdotal stereotypes suggest that technology is harder for older adults to learn, this research found that older adults are willing to learn and use communication applications. Results also suggested that intuitive visual cues might play an important role in enabling older adults to grasp and understand the structure of an application. Therefore, this paper tries to expand the scope of designers’ knowledge for the development of tablet-based applications for older adults and ﬁnd the possibilities of data usage for the related gerontology areas of study. Keywords: Older adults Tablet-based application Social media for older adults Communication method Communication application for older adults

1 Introduction Sharing emotions, opinions and thoughts is an important aspect of communication between humans. As technology has rapidly developed, human interaction has dramatically moved to online spaces through social communication applications and media. According to the Pew Research Center, although 78% of younger adults aged 32–49 used online communication through social networking and only 46% of older adults over 65 used it in 2013. Sinclair and Grieve [1] examined social connectedness of older adults in both faceto-face and online connections. They found that older adults have similar feelings of social engagement through either online connections or face-to-face conversation. This led us to assume that online social networks would be an alternative method for older adults who are less able to connect face-to-face with others or who are trying to reconnect with distant old-time friends. Thus, social network services (SNS) could be a good starting place for their online communication for who are less able to connect © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 228–234, 2019. https://doi.org/10.1007/978-3-030-02053-8_35

FLOW: A Software Application

229

face-to-face with others. Social communication through technology could play a huge role in the lives of older adults, and interactive technology for social communication is expected to be an assistive method for diminishing isolation problems and improving mental health for older adults [2]. Technology has the possibility to enable older adults to lead healthier lives with their intuitive social connection. However, according to Jung et al. [3], older adults are concerned with using Facebook or other SNS due to issues related to privacy and superﬁcial interactions, which makes them less comfortable sharing their personal stories in online public spaces. Therefore, this study focused on developing an application with an older-adult-user friendly interface in a secured environment that can help them communicate with others.

2 Method 2.1

Participants

There were eleven participants in this study, ﬁve of which were female and six were male. They ranged in age from 60 to 85 years old. All participants had college degrees and nine had advanced graduate degrees. They were all married and lived with their spouses at the time of this study. 2.2

Design Structure

Flow is a tablet-based web communication and social media app. It was developed for older adults to provide them with an individual-centered service in which they can control their individual stories and allow people in a speciﬁc group to see their stories. It provides an easy individual communication platform, in which older adults can create diverse types of communication: creating recorded stories, adding comments, and participating in chat rooms. Recordings of their voices will convey their emotions and mood; also, users can easily categorize and access their stories anytime. This is different from other social media such as Facebook, which is based on prompt newsfeed stories and does not categorize or retrieve stories as easily as Flow. With this in mind, Flow’s design structure has two main menus: My Story and Community Chats. 2.3

Application Design

The tablet interface design followed several web design guidelines [4–8] regarding font size, color combination, layout, icons, and sound. The prototype was developed on an Apple iPad with a 2048-by-1536-pixel resolution at 264 dpi. For easy recognition of differences in the menus, complementary color sets were set up in the application: Green for the Login, orange for My Story, and blue for Community Chats. A three-row grid was used to provide effective use of space and hierarchy structure in the vertical interface. Several prototyping programs were tried such as Justinmind, InVision, CSS with Javascript, and Axure was used for ﬁnal prototyping. The prototype used in this study did not have a working voice recording function. Also, this prototype did not allow simultaneous users to be online, so task scenarios were somewhat scripted.

230

W. Jang et al.

2.4

Study Methods

A mixed methods approach was used to better understand the target audience in this initial study [9]. Data were collected through ﬁeld observation, interviews, and a survey. A research team visited participants’ homes and conducted the user study in a comfortable environment. They conducted the study with a given iPad and only their hands and ﬁnger movements were recorded. It took each participant less than an hour to complete the user study.

3 Results and Discussion 3.1

Design Results

Analysis of participants’ answers and reactions during the user study showed that, in general, most participants kept in touch with their friends or family by phone or email more than 8 times per week and met people in person more than 8 times per week through direct meetings in person. Bias: (1) Even though most participants had computer experience, they were not familiar with social media and they sometimes had a negative opinion about it. They believed social media is too open-ended and complicated to use. The few participants who did use social media took an observer position, just gathering news from family members or friends. These ﬁndings showed that older adults need a social communication method that ﬁts their proﬁle and demands. (2) An additional ﬁnding in the study was that most participants had no previous iPad experience and originally were somewhat negative in their thinking about touch-based tablets. They thought their heavy ﬁngers and dry skin might possibly inhibit their use of an iPad, but our observation was that the problems were due more to how they use their ﬁngers in pressing the screen and how long they maintained pressure on the screen. More experience in using a tablet would probably quickly solve such problems. (3) Additionally, despite many research studies having shown that voice-based methods are easier for older adults to use [10], eight participants (73%) in the second interview said they would rather use typing due to its accuracy and because it allowed enough time to review their responses. Since most participants had advanced degrees, they might be somewhat familiar with writing from their own past academic activity. Most participants agreed that a recording method would be easy but felt that using a recording function might imply they had a physical impairment. E.g., one participant said, “I am ﬁne with writing now. I may need this recording function later.” This could be one of the reasons they might think a recording function would not be a comfortable way to upload a story. However, after experience with Flow, several participants changed their mind to prefer recording function. It shows that user-friendly experience will be enough to change their behaviors. Qualitative and quantitative data: Participants made mistakes while operating the application and searching for the menus. Interviews and observations suggested the application has problematic portions that should be modiﬁed. Despite these issues, the observations showed that participants exhibited positive attitudes toward Flow.

FLOW: A Software Application

231

(1) Quantitative data indicated that most survey questions related to communication rendered above-average mean values(mid-point). In Question 9, most participants agreed that Flow would help enhance their memory by providing interaction with others (Mean: 4.182, SD: 0.603). These results conﬁrm ﬁndings by Ford et al. [11] that older adults tend to be positive and overestimate their actual performance, even though they might have trouble operating touch-based devices. (2) Qualitative data indicated that the interface application needs improvements for use by older adults. For example, most participants were unsure about the right steps to take while navigating the app, and only three participants completed the user study without asking questions. Navigation (shortcut menu): Flow had two main menus, My Story and Community Chats. Participants were confused with two shortcuts, Friend’s News and Create Story, which were located in the main menu. Several participants suggested that Friend’s News and My Friends, which was located in My Story, should be merged and placed in the main menu as My Friends. Therefore, My Friends became a new main menu, which includes Friend’s News. Navigation (cancel button): The most problematic aspect with the Cancel button was its name. Most participants expected to encounter a back-arrow icon to return to the previous page. This problem was solved by change the button’s name to Back with an arrow icon. Navigation (comment): Flow has a comment area that allows users to add their comment in their friend’s story. It had several complex functioning buttons to process and upload stories, which made most participants hesitate to press buttons. Participants thought the microphone icon was the recording button. Another reason why they could not recognize the microphone, or the recording buttons was that there was no printed word to describe the function of either icon. Also, most participants said the recording button should be green instead of red, because red means “danger” for them, not “recording.” Finally, most participants reacted to icons with a speciﬁc shape that use less cognition function such as the microphone icon. Intuitive design development is very important, and several design guidelines for older adults discuss imprinting icons with words that make sure the icons’ function. It is a more efﬁcient way to help older adults have conﬁdence in their choices. This problems with icon recognition were solved by redesigning the uploading method and icons. 3.2

Recommendation for New Design

Several factors derived from this initial user study may help designers develop a better application. Although participant sample size was small, the fact that most participants had similar answers, opinions, and suggestions have made it clear that there are speciﬁc considerations that designers should be aware of when designing applications for older adults. Design elements for older adults should be intuitive, quickly recognizable, and require minimal attention. Structure: Designs should have parallel menu structures with only one deep hierarchy for ease of use. Providing enough space in the design elements permits older adults to use a menu intuitively without requiring unnecessary cognitive ability. A shortcut with

232

W. Jang et al.

different menu hierarchies was problematic and participants were confused and struggled to understand the menu functions. As a result, the new design has a simpliﬁed structure (Fig. 1).

Fig. 1. Revised new design and structure of flow

Words: To develop a service for most generations and be easily accessible to older adults, common word usage for easy approaches are recommended. For example, participants preferred the word “Back” instead of “Cancel” (Fig. 1). Icons: The icon shape was the ﬁrst recognizable cue for older adults, and a word added to the icon provides conﬁdence in their choice. Color is a secondary cue in recognizing the function of an element, but color itself cannot often describe the meaning of the icon, so a recognizable shape would be the ﬁrst priority condition for the icon (Fig. 1). Behavior: Older adults tend to overestimate their performance during user testing [11]. A given question delivered in several different ways may help determine actual behavior. An older adults’ answer of “Fine” may not really mean “Fine,” so researchers must observe their behavior carefully. For this reason, multiple-user study methods would be appropriate for the study of older adults by providing different perspectives that could help understand older adults’ experiences and preferences more effectively.

4 Limitations Participants’ level of education was very high and not representative of the majority of older users. Nine out of the 11 participants held Ph.D. degrees, meaning that results from this study do not provide sufﬁcient generalizability beyond this population.

FLOW: A Software Application

233

Although the application prototype was interactive, its recording feature was not available, and participants were not able to use like programmed one. Thus, participants could not have the same experience they might have if it were an actual fullyfunctioning app. This limitation particularly decreased the chance of observing how participant bias might affect the voice recording function. Although this was a functional problem more than an interface problem, it made users feel uncomfortable using the application and may have affected the study’s results. Plans for a follow-up user study are under way, and include: increasing sample size, and having participants coming from several different socioeconomic and educational backgrounds. In the next phase, the prototype will feature fully-functioning recording, cloud-based storage of stories, and simultaneous users, as well as improved instructions and icons with older adults-friendly wording.

5 Conclusions and Future Work Results from this initial study recommend the development of a speciﬁc design methodology for future development of apps for older adults. Results also show that older adults have different perspectives compared to younger adults in terms of design and usability for social media apps. Thus, this study suggests that the new user interface should be developed as a programmed application capable of all functions. An app like Flow can also be used to monitor patterns of usage of the application by older adults. Certain patterns of usage could be monitored for improvement of social health. By offering a method of collecting social interaction levels and monitoring health, Flow also offers an information system that supports data driven science and addresses the societal challenge of isolation among older adults.

References 1. Sinclair, T.J., Grieve, R.: Facebook as a source of social connectedness in older adults. Comput. Hum. Behav. 66, 363–369 (2017) 2. Horowitz, A., Brennan, M., Reinhardt, J.P., MacMillan, T.: The impact of assistive device use on disability and depression among older adults with age-related vision impairments. J. Gerontol. Ser. B Psychol. Sci. Soc. Sci. 61(5), S274–S280 (2006) 3. Jung, E.H., Walden, J., Johnson, A.C., Sundar, S.S.: Social networking in the aging context: why older adults use or avoid Facebook. Telemat. Informat. 34, 1071–1080 (2017) 4. Agelight, L.: Interface design guidelines for users of all ages, pp. 1–17 (2001) 5. Coyne, K.P., Nielsen, J.: Web Usability for Senior Citizens: Design Guidelines Based on Usability Studies with People Age 65 and Older. Nielsen Norman Group (2008) 6. Hodes, R.J., Lindberg, D.A.B.: Making your website senior friendly. National Institute on Aging and the National Library of Medicine (2002) 7. Holt, B.J., Komlos-Weimer, M.: Older Adults and the World Wide Web: a Guide for Web Site Creators (2001) 8. Zhao, H.: Universal usability web design guidelines for the elderly (age 65 and older). Universal Usability in Practice (2001)

234

W. Jang et al.

9. Barg, F.K., Huss-Ashmore, R., Wittink, M.N., Murray, G.F., Bogner, H.R., Gallo, J.J.: A mixed-methods approach to understanding loneliness and depression in older adults. J. Gerontol. Ser. B Psychol. Sci. Soc. Sci. 61(6), S329–S339 (2006) 10. Dickinson, A., Hill, R.L.: Keeping in touch: Talking to older people about computers and communication. Educ. Gerontol. 33(8), 613–630 (2007) 11. Ford, A.B., Folmar, S.J., Salmon, R.B., Medalie, J.H., Roy, A.W., Galazka, S.S.: Health and function in the old and very old. J. Am. Geriatr. Soc. 36(3), 187–197 (1988)

Learning from Human Behavior to Improve Preventative Health Information Systems Remberto Martinez1(&), Marcos Tong1, Luis Diago2, and Jaana Lindstrom3 1

Extensive Life Oy, Tampere, Finland {Remberto.Martinez,Marcos.Tong}@health-e-living.com 2 Interlocus Inc., Yokohama, Japan [email protected] 3 National Institute for Health and Welfare, Helsinki, Finland [email protected]

Abstract. Health services are often oriented to be ubiquitous for persons comfortable with mobile device and web technologies usage. In this environment, the main question is: can computer systems learn from human behavior to improve preventative health information systems? Current healthcare information systems focus on ensuring adherence to treatment through Just-In-Time Adaptive Interventions (JITAI). JITAI is used for preventative health and lifestyle behavior change interventions. The main problem for JITAI is lack of proper models of user’s behavior allowing effective interventions. Our novel approach adds contextual variables to the behavior learning system embedded in the message contents of lifestyle modiﬁcations interventions. The main result obtained from the proposal was a 33,8% perceived effective SMS intervention using behavior modeling with reduced specialized feedback. Chronic diseases like diabetes type 2 (DT2) are preventative when lifestyle behavior can be modiﬁed towards healthier habits. The research results open a new range of applications in context-aware computing and improve the effectiveness of JITAI for preventative chronic diseases. Keywords: Lifestyle intervention Preventative health systems

Human behavior modelling

1 Introduction The prevalence of diabetes type 2 (DT2) and hypertension has increased rapidly in recent decades and this trend will continue as the global population ages. Since these chronic diseases have asymptomatic preclinical phase the prevalence of undiagnosed cases is also of clinical and public health concern [1]. Feel4Diabetes (“Families across Europe following a healthy Lifestyle 4 Diabetes prevention”, F4D for short) is an EUfunded project* (2014–2019) aiming to develop, implement and evaluate an evidencebased and potentially cost-effective and scalable intervention program to prevent type 2 diabetes across Europe, primarily focusing on families from vulnerable groups. The current preliminary outcomes of the ﬁrst year revealed that F4D intervention appears to © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 235–241, 2019. https://doi.org/10.1007/978-3-030-02053-8_36

236

R. Martinez et al.

have favorable effects on a wide range of anthropometric and biochemical indexes for both children and their parents [1]. However, a follow up examination after a completion of the second year of the intervention is necessary to reveal important learnings about the effectiveness of the intervention. In this project, the main question was: can computers learn from human behavior to improve preventative health information systems? In this work we focus on the data obtained from motivational SMS used on F4D project to answer whether intervention was effective and/or perceived effective by users in the high-risk group. After users are selected to participate in the intervention, they were educated on lifestyle changes, biometrics measured and included also in SMS follow up during the second year. Three learning methods are compared to predict the user perceived beneﬁt from the collected data. The rest of the paper is organized as follows. First, Sect. 2 describes the system, how data was obtained and what behavior model speciﬁcation was used in F4D-SMS intervention. Section 3 poses the main problem addressed in this paper, whether the SMS intervention is perceived effective by the participant and can the data be used to predict the correctness of answers. Finally, results, conclusions and future work are presented in last section.

2 Overall System for SMS Intervention Mobile phone messaging is an effective and acceptable method to deliver advice and support towards lifestyle modiﬁcation to prevent type 2 diabetes [8]. The F4D SMS intervention aimed to induce or maintain healthy changes in physical activity and diet among participants of the F4D High Risk Families. The design required a low-cost low-maintenance system to reinforce the behavioral skills learn in the previous phase of project according to HAPA model [6]. Figure 1 shows the design.

Fig. 1. Overall system design for F4D SMS intervention

The F4D participants were not required having a smart phone with Internet connection but only a featured phone able to send and receive SMS. While the intervention was designed to be interactive (participants were able to receive educational messages and answer to questions), the SMS questionnaire was conceived, translated and culturally adapted so participants could receive 2 messages per week and answer to maintenance questions every 2 months. The coach (country researcher) had access to a management tool accessible via phone or web with authorized credentials to help participant’s registration in each group. Once the participant is registered she can answer the invitation SMS with a numeric value (replying with 1 to continue or 9 to opt

Learning from Human Behavior to Improve Preventative Health Information Systems

237

out). In case of no answer the system registers the incomplete state and tries to resend the message three times otherwise the user is considered dropped out. For those participants not sure what to reply to the invitation question, the coach was allowed to override the topic selection (coach becomes a data agent) during registration phase. In this case the participant is considered to have answered the initial question. After that the system continues to send 2 SMS per week automatically with messages containing educational information associated with the topic selected, and including timing, intention, content and representation components [5]. 2.1

Data Collection

The data collection from SMS interactions is summarized in Table 1. There were more than 140 participants per each country within the intervention groups having high risks of developing DT2 according to the diabetes risk score [2]. Each country could have one or several coaches, and for each participant was recorded whether they were able to register (active), as well as whether they accept receiving topical messages or just abort the messaging after several rounds (opted out). Table 1. Participants per country. Country Belgium Bulgari Greece Finland Hungary Spain

Coach # Active 1 36 1 11 4 247 2 82 1 0 1 115

Opted out Total 75 143 41 226 5 288 48 142 20 209 36 185

From the period of continuous motivational message reception, it was possible to change the selected topic after a period of 2 months. The participant was able to reply to an SMS question at any time with just a text number between 1 and 5, or 9 to abort. Figure 2 shows an overall of the collected data features. The values for Opted-out and Drop-out are stored as YES-NO according to participant reply. All answers were stored for later processing (see “Last Answer” summary above) as well as how many times the topic was changed (“Num Change Topic”) to have an indication of participant engagement. In the case there was no answer received during a week, the system repeated the question up to 3 times. After that the participant is considered dropped if the coach did not agented a topic initially. The data was cleaned to merge equivalent answers and replace wrong answers as not answered. The histograms for the total number of SMS (“Total Num SMS”) and the maximum number of SMS (“Max SMS in a Round”) are also shown in Fig. 2 as a measure of system usage.

238

R. Martinez et al.

Fig. 2. Overview of data collected

2.2

Human Behavior Modelling in F4D SMS Intervention

The HAPA model was used in the SMS questionnaire to evaluate the participant current state according to one of HAPA phases (motivation/volition/maintenance) [6]. The initial SMS text was designed as an invitation to participate in the intervention and required consent from the participant (motivation phase). Here the participant forms the intention according to its own believes, perceived self-efﬁcacy and outcomes expectations [2]. The following question were designed to determine the maintenance and volition states: “How satisﬁed are you with your current life style and wellbeing? 1 means very unsatisﬁed, 5 means fully satisﬁed”. For the maintenance phase the content was oriented to promote healthy lifestyle while the volitional phase includes 2 more questions to allow participants select a speciﬁc topic of interest between physical activity and nutrition preferences. During this phase the content of messages was designed to support self-efﬁcacy, emotional commitment and failure learning (action planning and recovery in HAPA). Before topic selection it was added question to verify whether participant followed the intended design: “How ready are you to make lifestyle changes (1 means not ready, 5 means very ready)”. This allowed to redirect the participant to the motivational state with content oriented to self-efﬁcacy, outcome expectations and risk perception. After a cycle of 2 months, the system asked the question: “How beneﬁcial have these messages been for your lifestyle? Reply 1: not beneﬁcial, 5: very useful, 9: stop sending SMS”. The answer to this question is of high interest here, any number between 1 and 5 allowed the participant to continue for 2 more months while 9 considered the participant as opted out (see Table 1). The answer 1, 2 classiﬁes as not beneﬁcial, 3 as cannot tell and 4, 5 as beneﬁcial.

Learning from Human Behavior to Improve Preventative Health Information Systems

239

3 Prediction of Perceived Beneﬁts Perceived effectiveness is a judgment under uncertainty data science problem. While emotions might explain why human behavior changes according to environment challenges, the decision making under uncertainty is conditioned by previous learning experiences, demographics and personality traits. In [4] the theory of planned behavior is used as basis for modelling intention formation since it incorporates many cognitive variables: intentions, outcome expectancies and perceived behavioral control. In this paper a statistical approach is used to validate whether the answers to the perceived beneﬁts question can be used also to predict was the SMS intervention a positive experience, how it is remembered and was the perceived beneﬁt just a biases impression or statistically sounded. In this case the answer can be used directly as an estimate of the effectiveness of the intervention. However, as shown in Fig. 2, the “Last Answer” number of responses 9 is very high (238). The number of missing values for “Last Answer” is 820 (69%) and for the other three variables (“Num Change Topic”, “Max SMS in a Round” and “Total Num SMS”) is 463 (39%). Very little attention has received the problem of missing values in the community of machine learning when it comes to applying a classiﬁer on prediction with missing values [3]. There are several methods to solve this problem, in this paper we explore 3 of the classiﬁers available in WEKA [7] to get an idea of the percentage of accuracy that classiﬁers could achieve for the prediction of perceived beneﬁts in F4D-SMS intervention data set. WEKA tool is available as open source with implemented algorithms like Multilayer Perceptron, Bayes Naïve and Zero R.

4 Preliminary Results The SMS interaction was designed to run continuously for one year with check point questionnaire cycle of 2 months. While the SMS interactivity was simple and assumed that mobile skills were minimal, it was noticed a large amount of missing answers in several parts of questionnaire. In 5 of 6 countries the missing values occur completely at random except for the case of Hungary where the feature for skipping topic selection was not used by coaches and participant did not understand how to reply to SMS or were not comfortable with SMS at all. There were challenges in SMS content translations and cultural adaptation, especially for countries using Cyrillic alphabet in their mobiles. One of biggest challenge was to ﬁt the essence of content in 150 characters. These features increased the costs for automatic trafﬁc generation as new regulations on privacy and prepaid service number transactions online are limited within European countries speciﬁc implementations of GSM standard. Three WEKA algorithms were run in test mode 10-fold cross-validation with full training set and compared in % of correctly classiﬁed instances as shown in Table 2. The features used were those in Fig. 2: number of users opted out, number of users drop out, num change topic, total num SMS and max SMS in a round. For “Last Answer” prediction all 238 instances including the answers containing value 9 (opt out) were considered missing values.

240

R. Martinez et al. Table 2. Weka classiﬁers comparison for 1193 instances. Algorithm Bayes naïve Multilayer perceptron Zero R

Correctness 30.8271% 25.5639% 24.812%

5 Conclusions and Future Work Mobile phone SMS intervention is an effective preventive method to promote healthy lifestyle changes in economically vulnerable European populations with considerable risk to develop DT2. Most active countries were Greece, Finland and Spain followed by Belgium and Bulgari. Hungary participants were not able to use the SMS intervention as no answer was received from the engagement question. From all collected answers to the intervention beneﬁts perceived on scale 1 to 5, a 41.3% answered little beneﬁt (1,2), a 24.8% answered not sure (3) and a 33.8% answered some beneﬁt (4,5). The best algorithm for classiﬁcation was Bayes Naïve with 30.8271% correctly classiﬁed instances. There was not notable influence on the number of topic changes and the total number of SMS but the max number of SMS in one round had stronger correlation for answers 4–5. This might be explained with the number of different messages for each topic, as the system reuses the messages once the last one is sent. Topics related to physical activity had an average diversity of 19 messages while topics on nutrition had an average of 16 messages. The amount of missing values for the data collected was very high (89%) which explain the poor correctness of the classiﬁers. With the available data set it can be concluded that only 33.8% of participants perceived the SMS intervention beneﬁcially. As SMS has being used lately for marketing and advertising in society the informative perception could be challenged as next research. In future works it would be possible to analyze the data from questionnaires previous to the SMS intervention and understand what proﬁles are more impacted by the motivational SMS. It might be possible to simulate missing values with formal speciﬁcations of human behavior to improve classiﬁers correctness.

References 1. Manios, Y., et al.: Preliminary outcomes of the 1st year of an intervention aiming to promote healthy lifestyle in a large-scale European cohort: Feel4Diabetes study. In: Proceedings of 53rd Annual Meeting of the European Diabetes Epidemiology Group, p. 74 (2018) 2. Lindström, J., Tuomilehto, J.: The diabetes risk score: a practical tool to predict type 2 diabetes risk. Diabetes Care 26, 725–731 (2003). https://doi.org/10.2337/diacare.26.3.725 3. Saar-Tsechansky, M., Provost, F.: Handling missing values when applying classiﬁcation models. J. Mach. Learn. Res. 8, 1217–1250 (2007) 4. Brailsford, S.C.: Healthcare: human behavior in simulation models. In: Kunc, M., Malpass, J., White, L. (eds.) Behavioral Operational Research. Palgrave Macmillan, London (2016) 5. op den Akker, H., et al.: Tailored motivational message generation: a model and practical framework for real-time physical activity coaching. J. Biomed. Inform. 55, 104–115 (2015)

Learning from Human Behavior to Improve Preventative Health Information Systems

241

6. MacPhail, M., Mullan, B., Sharpe, L., MacCann, C., Todd, J.: Using the health action process approach to predict and improve health outcomes in individuals with type 2 diabetes mellitus. Diabetes Metab. Syndr. Obes. Targ. Ther. 7, 469–479 (2014) 7. Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data mining software: an update. SIGKDD Explor. Newsl. 11, 10–18 (2009). https://doi.org/10. 1145/1656274.1656278 8. Ramachandran, A., et al.: Effectiveness of mobile phone messaging in prevention of type 2 diabetes by lifestyle modiﬁcation in men in India: a prospective, parallel-group, randomised controlled trial. Lancet Diabetes Endocrinol. 1(3), 191–198 (2013). https://doi.org/10.1016/ S2213-8587(13)70067-6. Epub 2013 Sep 11

Prototyping a User Interface for a New Sepsis Risk Decision Support System Using Participatory Design Richard Harte1,2(&), Leo R. Quinlan2,3, Evismar Andrade1,2, Enda Fallon4, Martina Kelly4, Paul O’Connor5, Denis O’Hora6, Patrick Pladys7,8, Alain Beucheé7,8, and Gearoid ÓLaighin1,2 1

Electrical and Electronic Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland [email protected] 2 Human Movement Laboratory, NUI Galway, University Road, Galway, Ireland 3 Physiology, School of Medicine, NUI Galway, University Road, Galway, Ireland 4 Mechanical Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland 5 General Practice, School of Medicine, NUI Galway, University Road, Galway, Ireland 6 School of Psychology, NUI Galway, University Road, Galway, Ireland 7 Centre Hospitalise Universities de Rennes (CHU Rennes), rue Henri Le Guilloux, Rennes, France 8 Faculté de Médicine de l’ Université de Rennes, Rennes, France

Abstract. Digi-NewB is a system currently being developed to monitor and predict the risk of sepsis in infants within a Neonatal Intensive Care Unit (NICU) setting. More than 300,000 preterm infants are hospitalized each year in a European care unit. Sepsis diagnosis is complex and therefore usually late, resulting in an increased risk. In this paper, we present on our experience with applying a participatory design based prototyping method to create user interface (UI) concepts for the Digi-Newb system and then testing the prototypes with end-users. Prototype making within the participatory design framework was found to be an effective method to rapidly develop potential design solutions, utilizing the experience of the end-user as a design partner. Keywords: Prototyping Sepsis risk

User-centered design Participatory design

1 Introduction Neonatal sepsis contributes substantially to neonatal morbidity and mortality, and is an ongoing major global public health challenge [1]. The condition can cause signiﬁcant respiratory and neurological problems and is responsible for 13% of all neonatal deaths, and 42% of deaths in the ﬁrst week of life [2]. The risk of a preterm infant acquiring late © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 242–248, 2019. https://doi.org/10.1007/978-3-030-02053-8_37

Prototyping a User Interface for a New Sepsis Risk Decision Support System

243

onset sepsis is approximately 25%, resulting in signiﬁcant rates of death and respiratory and neurological sequelae. Sepsis diagnosis is complex and therefore usually late, resulting in an increased risk for the patient. The Digi-NewB project [3] proposes a novel solution to this problem, by designing a non-invasive Decision Support System (DSS) which will use vital signs, images and sounds to measure the risk of sepsis and therefore support clinicians in the diagnosis. The introduction of any new system to a NICU environment presents a challenge for designers who must account for a technology-laden environment and a demanding work-load for caregivers [4, 5]. The Digi-NewB project is following a three-phase User-Centered Design (UCD) process (Fig. 1). A pre-design User Research phase was previously carried out to understand the context of use of a potential Digi-NewB system within the NICU. This User Research phase involved a period of immersive focused ethnography, a series of end-users interviews and a generative codesign prototyping exercise which produced the ﬁrst concepts for the system User Interface (UI). The feedback gathered from these activities led to the construction of the ﬁrst Use Case and the ﬁrst list of User Requirements.

Fig. 1. The yellow tile shows the current study in the context of the overall Digi-NewB UCD process. This study describes the generation of the ﬁrst UI prototype

In this paper we present the next stage of development, a period of iterative participatory paper prototyping to enhance and validate UI concepts, and therefore create the ﬁrst prototype for user testing. Within the participatory design framework, prototyping is seen as a means to evoke focused discussions on concepts, to test hypotheses and to confront the reality in which the system will exist [6].

2 Methodology The methodology for generating and testing prototypes consisted of two stages: Stage 1: Prototype Generation through Parallel Design: Using the Use Case and User Requirements document as the initial reference, a multi-disciplinary team of four designers each independently constructed a set of UI paper prototypes [7]. Creating multiple prototype alternatives in parallel has been shown to encourage designers to effectively discover unseen constraints and opportunities and promotes a culture of

244

R. Harte et al.

comparison and compromise between concepts [7]. The four prototype sets were then analyzed by a cognitive walkthrough methodology, using Nielsen’s usability heuristics [8] as a guide with which to assess the 4 different designs. A moderator assessed the performance of each design and created a uniﬁed paper prototype design, which featured elements of the four starting designs. This uniﬁed paper prototype was the basis of a detailed evaluation by end-users and subsequent iterative reﬁnement. Stage 2: Prototype Evaluation and Iteration: The paper prototype was then evaluated in three formative cycles using 11 clinicians (8 nurses and 3 doctors) from the NICU at a university hospital in France. After each cycle the prototype was updated by the usability engineering team in response to user feedback [9]. It was explained to the participants that they were considered codesign partners in this process. This involved users providing think-aloud feedback on the actions required to complete each task, the flow of the navigation, the screen layout and the terminology on the screens. This provided the designers with insight into the information needs of the users [10]. Additional feedback was gathered by an After-scenario Questionnaire (ASQ) and the completion of the System Usability Scale (SUS). The entire process is illustrated in Fig. 2.

Fig. 2. Methodology for making, evaluating and evolving the Digi-NewB Interface

3 Results 3.1

Parallel and Converged Design

The features from each of the four parallel prototypes which best captured the use case and user requirements were merged into the uniﬁed prototype. For example, one of the prototypes allowed the user to explore past risk trends in different timelines. The uniﬁed paper prototype was put forward for exposure to end user representatives. An example of the convergence of multiple paper prototype screens into one is illustrated in Fig. 3. The uniﬁed paper prototype consisted of the following features (F1-F7); F1: A start or login screen with which the user logs into the system, F2: A screen or notiﬁcation to check if the system is set up, F3: A screen(s) to enter patient details or setup a new patient on the system correctly, F4: A screen to check the current sepsis risk of the patient and to make a note of or log the sepsis score i.e. Traceability, F5: A function to change certain dynamic clinical parameters such as belly distension, F6: A function to view the sepsis risk in more detail on a new screen, F7: A screen which shows a summary for each patient in the unit.

Prototyping a User Interface for a New Sepsis Risk Decision Support System

245

Fig. 3. Features and functions from different screens were converged for the patient setup screen.

3.2

Iterative Prototyping

For the evaluation of the uniﬁed paper prototype, the participants were asked to complete two tasks; (1) Setup a new patient; (2) Review current status of the patient and update the patient clinical data. These two tasks allowed the participants to explore and comment on all screens of the paper prototype. Evaluation of Rev1 Prototype. The participants pointed out that certain information should be made more prominent, particularly the sepsis risk number which is the most important piece of information. Pieces of information were also not well grouped. The concept of traceability (F4) created some confusion as to why the feature was necessary or what exactly it was to be used for. Users said the concept needed to be developed such that users knew exactly what was being traced and how many times per day they would have to enter their PIN. Nurses also felt that the level of data entry required at the patient setup (F3) and during the update of clinical metrics (F5), would add unacceptable time to their workload. Evaluation of Rev2 Prototype. To make data entry more efﬁcient, a feature which would allow users to swipe an ID card rather than enter a PIN was added to the Rev2 prototype. The number of data entry steps required for admitting a new patient was reduced to minimize the time burden on users. Participants still identiﬁed some issues with the layout of information in the Rev2 Prototype. Evaluation of Rev3 Prototype. Clinicians found the Rev3 prototype layout clear and easy to navigate. They made some minors comments of layout and information being presented. But still expressed concerns about the level of interaction required which they felt would contribute a signiﬁcant burden on their already full workload. In particular they were concerned about the level of data entry. A ﬁnal paper prototype was created in response to this ﬁnal evaluation. Figure 4 illustrates how the same

246

R. Harte et al.

screen on the prototype evolved from the Rev1 to Rev3. It can be seen how certain concepts were developed in more detail from the ﬁrst iteration to the third iteration.

Fig. 4. The screen from Prototype Rev1 shows simple details including sepsis risk score, historical trend breakdown of individual metrics; The screen from Prototype Rev2 develops a tab system such that the user can see different levels of detail in each tab; In the Rev3 prototype the user is provided information on sepsis risk, clinical details and medication details. The tab system is now on the left hand side of the screen to allow for space along the top bar

The ﬁnal paper prototype was sent to developers to generate a functional software format. Figure 5(A and C) shows examples of the screens in the ﬁnal paper prototype which were then converted to software prototypes (B and D).

Fig. 5. (A) shows the nurses’ central station screen where users can see a summary of all the patients in the unit; (B) Shows the software version of A; (C) Shows the personal screen for each patient which can be viewed via an interface in the room or can be viewed by clicking on a tile from the central station screen; (D) Shows the software version of C, in response to comments by users in the ﬁnal evaluation the tabs were moved from the side of the screen back to the top of the screen

Prototyping a User Interface for a New Sepsis Risk Decision Support System

247

The SUS and ASQ scores which were captured after each evaluation are shown in Table 1. It can be seen that the metrics are improving after each iteration. This increase is in response to the prototype increasing ﬁdelity as well to the designers responding to user comments and implementing desired features and removing unwanted or confusing screen elements.

Table 1. SUS and ASQ metrics captured after each evaluation for each task Evaluation Task Rev1 Task 1 Task 2 Rev2 Task 1 Task 2 Rev3 Task 1 Task 2

Ease of use Efﬁciency 5.5 6 6 6 7 6.7 6.5 6.5 7 6.5 6.5 6.5

SUS SUS learnability 72 73 82

73

90

90

4 Discussions and Conclusion This prototyping process conducted within a participatory design framework was found to be an effective and efﬁcient method to rapidly develop potential design solutions. The intimate nature of the sessions utilized the experience of the end-user as a design partner. Paper prototyping was found to be a flexible tool for iteration purposes, especially when concepts are not clearly deﬁned and it is not efﬁcient to spend time developing software prototypes which may need to be changed rapidly. The low ﬁdelity nature of the prototypes allowed users to be forthright and honest in their opinions on the design solutions presented. As pointed out in previous literature, paper prototyping does not require extensive expertise in code, a lack of which could hinder early development and prevent designers from exporting certain concepts. However, it must also be stated that paper prototypes should be used in this early fluid design phase, but the migration to software prototypes as soon as possible to allow for more comfortable and effective usability and human factors analysis. The next phase in the design process will see the ﬁnal software prototype produced at the end of the process be exposed to user testing, before a new use case is developed to reflect the feedback from those activities.

References 1. Qazi, S.A., Stoll, B.J.: Neonatal sepsis: a major global public health challenge. Pediatr. Infect. Dis. J. 28(1 Suppl), S1–S2 (2009) 2. Zea-Vera, A., Ochoa, T.J.: Challenges in the diagnosis and management of neonatal sepsis. J. Trop. Pediatr. 61(1), 1–13 (2015)

248

R. Harte et al.

3. Värri, A., Kallonen, A., Helander, E., Ledesma, A., Pladys, P.: The Digi-NewB project for preterm infant sepsis risk and maturity analysis. Finn. J. Ehealth Ewelfare 10(2–3), 330–333 (2018) 4. Joseph, R.A.: New technology in the NICU: challenges to parents and clinicians. Neonatal Netw. 35(6), 367–374 (2016) 5. Braithwaite, M.: Nurse burnout and stress in the NICU. Adv. Neonatal Care 8(6), 343 (2008) 6. Rodgers, P., Yee, J.: The Routledge Companion to Design Research. Routledge, Abingdon (2014) 7. Dow, S.P., Glassco, A., Kass, J., Schwarz, M., Schwartz, D.L., Klemmer, S.R.: Parallel prototyping leads to better design results, more divergence, and increased self-efﬁcacy. ACM Trans. Comput.- Hum. Interact. 17(4), 18:1–18:24 (2010) 8. Nielsen, J., Molich, R.: Heuristic evaluation of user interfaces. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, New York, NY, USA, pp. 249–256 (1990) 9. “Parallel & Iterative Design + Competitive Testing = High Usability,” Nielsen Norman Group. https://www.nngroup.com/articles/parallel-and-iterative-design/. Accessed 26 Feb 2018 10. Sanders, E.B.-N., Stappers, P.J.: Probes, toolkits and prototypes: three approaches to making in codesigning. CoDesign 10(1), 5–14 (2014)

A Proposal for an Affective Design and User-Friendly Voice Agent Heesung Park(&), Jeongpyo Lee, Sowoon Bae, Daehee Park, and Yenah Lee Samsung Electronics, 56, Seongchon-gil, Seocho-gu, Seoul, Republic of Korea {Hux.park,jp0212.lee,sowoon.bae, daehee0.park,yenah.lee}@samsung.com

Abstract. In these days, most technical research conducted in relation to the voice agent which is mounted on mobile devices. However, research on the affection of users who use voice agents has not been studied yet. In this study, we assumed that users’ affective responses to three Voice Agents (Siri, Bixby, and Google Assistant) might be different. To do this, we asked the participants to perform four tasks (voice registration, checking information, using the speciﬁc functions, and joking). Hence, affective responses were measured by SAM (Self-assessment Manikin). Finally, we found out the difference of users’ affective responses to each voice agent. Then we propose design factors for userfriendly voice agent on the mobile. Keywords: Mobile voice agent

Affective design HCI

1 Introduction Touch interaction has become a common method of interaction on the smartphone. However, in recent years voice interaction technologies have developed, and a variety of products applying this technology have been released to the market. The most popular mobile voice agents are Apple Siri, Samsung Bixby, and Google Assistant. These products use a voice agent coupled-with a touch screen based user interface. Many IT companies are continually striving to deliver a better experience to their users. A mobile voice agent recently has been increasingly regarded as an outstanding example of a way to convey a better experience to a user. Also, Voice interaction in HCI offers advantages for users [1]. Voice agents based on verbal interaction allow users to use mobile devices beyond the limitations of existing mobile interaction methods. There is a list of papers related to voice on Google’s speech processing website. However, many studies focus on the technology, such as the accuracy of voice recognition [2–5]. However, in the case of voice interaction, auditory modality is signiﬁcant since the user receives and expresses affection through auditory modality [6]. Therefore, affection is regarded as an important factor [7]. In this study, we have assumed that there is an effective difference between each voice agent across different platforms (Apple Siri, Samsung Bixby, and Google Assistant). We depicted what factors have appeared in these differences. We asked participants to perform various tasks with the voice agents at their disposal, and then © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 249–255, 2019. https://doi.org/10.1007/978-3-030-02053-8_38

250

H. Park et al.

their affective responses were measured. In order to make direct measurements, a selfassessment manikin model considered which could easily measure valence and arousal across various spectrums [8]. Then, we analyzed those factors which showed signiﬁcant difference between the lineup of mobile voice agents. Finally, in terms of affective design, we will explore and propose a key element in voice agent design based on the research results. 1.1

Experiment Design and Methods

In order to avoid bias from previous voice agent experiences, we conducted experiments with participants who have never used a Voice Agent before. We chose Apple Siri, Samsung Bixby, and Google Assistant. In order to conduct the experiment, we designed four tasks such as voice registration and calling, checking simple information, using speciﬁc functions, and joking. After the performance of a task, we evaluated the affective responses of each agent through SAM [9, 10]. After performing each task, three papers with a picture of SAM’s Valence, Arousal, and Dominance were given to the participants. In the next stage, participants were asked to choose a picture similar to what they had felt while performing the task. Task 1 is to register his/her voice in Agent and call the agent. Task 2 requires asking the Voice Agent about the current weather conditions in Vietnam, and conﬁrming the weather information. In Task 3, the participants were asked to send a text message about ‘lunch tomorrow’. In Task 4, the participants were told to ask the voice agent to sing a song, and they then conﬁrmed the results. Finally, we conducted an in-depth interview regarding their preferences, as well as opinions on the strengths and weaknesses of the voice agents.

2 Results We use Self-Assessment Manikin method with 9-scale point scale. The genders of the participants were as follows: 3 males and 9 females, and the mean age was 27.1. Valence: it described only if there was a signiﬁcant difference in the valence scores between task 1 and task 4 (no signiﬁcant difference task: 2, 3) (Table 1). In addition, the average valence score was 6.67 for Siri in Task 1. And in Task 4, Google Assistant (GA) earned the highest score, which was 6.75 (Table 2). Table 1. Two-way ANOVA (Valence_ Task 1) Source of variation Voice agent Error Total

SS df MS F P-value F-crit 17.56 2 8.78 5.83 0.01 3.44 33.11 22 1.51 93.22 35

A Proposal for an Affective Design and User-Friendly Voice Agent

251

Table 2. Two-way ANOVA (Valence_ Task 4) Source of variation Voice agent Error Total

SS df MS F P-value F-crit 29.39 2 14.69 6.84 0.004 3.44 47.28 22 2.15 174.97 35

Arousal: According to Table 3, there was a signiﬁcant difference in the Arousal Score in Task 4, which asked Agent to sing (no signiﬁcant difference task: 1, 2 and 3). The average Arousal score was the highest for Google Assistant at 6.25. Table 3. Two-way ANOVA (Arousal_Task4) Source of variation SS df MS F P-value F-crit Voice agent 44.06 2 22.03 7.2 0.003 3.44 Error 67.28 22 3.06 Total 206.22 35

Dominance: According to Table 4, there was a signiﬁcant difference in the Dominance Score of Task 4, which asked the Agent to sing (no signiﬁcant difference task: 1, 2 and 3). Also, the highest average Dominance score was earned by Google Assistant, it was 5.33. Table 4. Two-way ANOVA (Dominance _Task4) Source of Variation SS df MS F P-value F-crit Voice agent 19.39 2 9.69 5.83 0.009 3.44 Error 36.61 22 1.66 Total 189.64 35

Self-assessment: According to Table 5, Siri is the most preferred agent among users, and it showed the highest recognition accuracy. In addition, Google Assistant is regarded as the best voice agent to comprehend users’ intention. Table 5. Ranking of Self-assessment Ranking 1 2 3

Preference Siri GA Bixby

Recognition Siri GA Bixby

Understand Feedback GA Same Bixby Siri

252

H. Park et al.

Perceptual Map: A perceptual map containing 125 emotions (Morris [11]). The perception map can be mapped to emotion text by conversion of the SAM score. We found the emotion texts of each agent in Task 1 and 4 (Table 6). Table 6. Emotion text of tasks 1 and 4 (which were signiﬁcantly different in ANOVA) Task 1 Task 4 Siri Bixby GA Siri Bixby GA Co-operative Concentrating Nonchalant Co-operative Concentrating Nonchalant

3 Discussion We tried to ﬁnd out if the signiﬁcant difference between the voice agents on the several situations. According to the overall results, there was no signiﬁcant difference in the SAM valence scores of tasks 2 and 3. This result indicates that there is no signiﬁcant difference in affection when participants use a voice agent for a special purpose, such as checking the weather or sending a message. Valence score of Apple Siri is highest among the three. The participants’ opinions on Siri are described to friend. Example of “Siri gets the feeling of listening to my words”, “When I call Siri, I feel the same afﬁnity as calling a friend.” Registering the voice of the user on the voice agent and calling the voice agent is regarded as the starting point to using the voice agent. When interacting with Apple Siri, a separate user interface is provided for only the voice agent. Also, when the user calls Siri, the main command is “Hey, Siri”. From this we can assume that being sympathetic is related to the mapping from the emotion text of the Perceptual Table to the ‘Co-operative’ (see Table 4). On the other hand, Bixby also provides separate interfaces for voice interaction. However, different emotions are afforded, as compared to Siri. When the user calls Bixby for the ﬁrst time, Bixby will listen to the voice of the user with a small listening animation on the screen. This indicates that Bixby is less likely to listen to the user’s voice. Also, when the user calls Bixby, the main command is “Hi, Bixby”. Participants felt awkward when calling Bixby during the experiment since the phrase felt less friendly compared to that used for Apple Siri. According to Table 4, calling Bixby is matched to “Concentrating”. It seems that the users are required to concentrate on Bixby more to have conversation with it. Task 4 is a function in which there is a high degree of freedom to exchange jokes with an Agent. In Task 4, Google Assistant was the highest in terms of the three affections (Valence, Arousal, and Dominant). Although Siri and Bixby sometimes did not play a song, Google Assistant always played songs, which led to a positive response. Google Assistant provided various ways to play a song. Even if the agent did not sing a song by itself, it played the previously recorded music, or gave a witty answer that his ‘song teacher is singing’. The agent also sometimes answered the experimenter’s question with a bit box instead of a song, or with a witty song. This indicated that the Perceptual Table’s Google Assistant is related to interests. However, Samsung Bixby sometimes gave a positive answer at the request of the experimenter, but sometimes a rejection. When issuing a positive response, the Voice Agent sung a

A Proposal for an Affective Design and User-Friendly Voice Agent

253

song or it recommended other methods. However, the user was noticeably timid or courageous. In the case of Apple Siri, the experimenter largely declined to request for the Agent to sing a song. As can be seen in the Perceptual Table, the emotion text for Siri is ‘Quietly indignant’. In Task 4, Google Assistant had the highest Dominance score. The participants rated the level of understanding of their speak. Google Assistant responded positively to all questions of the participants in a systematic fashion. However, the participants felt that they were playing with the machine. We found that there is a difference in emotion between each participant when registering voices with agents and performing functions with high degrees of freedom, such as when telling jokes. As we have seen in both tasks, the role of the voice agent is important in helping the function succeed. However, when the voice agent was regarded as a friendly target, the Valence Score was highly evaluated. If the designer tries to design a voice agent friendly, we recommend the following: First, we recommend that the designer should design a voice command which the user can use to set a custom friendly word. For example, the user sets the speciﬁc word as a voice command to call the voice agent. The speciﬁc word could be “Hey John” if the users feel “Hey John” is friendlier, and the terminology is used frequently. According to Table 5, Apple Siri had the lowest score in “Understanding of users words.” Understanding of user contexts by the voice agents is an important factor for the voice agents providing the correct functions, consistent with user intentions. On the other hand, Apple Siri was chosen as the most preferable voice agent. It influenced the ranking of user preference. Thus, we propose that the initial voice command to call the agent’s attention should be customizable. Second, we assumed that the visual interface on the display when a voice agent executing, represented how much the voice agent was focusing on user speech. Furthermore, the listening animation depicted the way in which the voice agent responds to the users friendly. Apple Siri and Samsung Bixby already provided a separate screen for only voice agent. Thus, users of Apple Siri and Samsung Bixby could recognize easily that they were interacting with a voice agent. Although, Google Assistant provides a separate user interface, some users could not recognize when they were interacting with a voice agent (found in In-depth Interview). In addition, it induced that Google Assistant is located nearby “Nonchalant” emotion. Therefore, we recommend that the voice agent should represent a message which is input verbally. Also, if the voice agent shows a listening animation, this may be helpful in understanding the status of voice recognition. Thus, the system provides feedback which indicates that if the voice agent understands the context of user command or not. (see Fig. 1). Third, the voice agent should show an effort towards trying to complete a task is requested by the user even though the user asks for a task which cannot be easily performed. In our experiment, task 4 was asking the voice agent to sing a song. Singing a song is not directly involved in the capabilities of the voice agent. According to what was mentioned above, we found that Google Assistant accepted user requests friendly without refusal. This is displayed as the “Interested” emotion. Thus, an analysis of the result indicates that accepting user requests without refusal incites emotions that are more positive.

254

H. Park et al.

Fig. 1. User-friendly voice agent interface

4 Conclusion Voice agents are regarded as a key feature towards simplify the existing mobile experience. We assumed that there were affective differences between the three Voice Agents (Siri, Bixby, and Google Assistant). Through SAM, we found a signiﬁcant difference in Task 1 and 4, Based on the results, we found various insights through the in-depth interview after associating the emotional text of each agent’s perceptual map. Additionally, we proposed three design improvements for the voice agents in affective design.

References 1. Cohen, P.R., Oviatt, S.L.: The role of voice input for human-machine communication. Proc. Nat. Acad. Sci. 92(22), 9921–9927 (1995) 2. Halpern, Y., Hall, K.B., Schogol, V., Riley, M., Roark, B., Skobeltsyn, G., Bäuml, M.: Contextual prediction models for speech recognition. In: INTERSPEECH, pp. 2338–2342 (2016) 3. Coward, S.W., Stevens, C.J.: Extracting meaning from sound: nomic mappings, everyday listening, and perceiving object size from frequency. Psychol. Rec. 54(3), 349–364 (2004) 4. Huang, T.Y., Huang, P., Chen, K.T., Wang, P.J.: Could Skype be more satisfying? a QoEcentric study of the FEC mechanism in an internet-scale VoIP system. IEEE Netw. 24(2), 42–48 (2010) 5. Chen, K.T., Tu, C.C., Xiao, W.C.: Oneclick: a framework for measuring network quality of experience. In: INFOCOM 2009, pp. 702–710. IEEE, April 2009 6. Kostov, V., Fukuda, S.: Emotion in user interface, voice interaction system. In: 2000 IEEE International Conference on Systems, Man, and Cybernetics, vol. 2, pp. 798–803 (2000) 7. Nass, C., Jonsson, I.M., Harris, H., Reaves, B., Endo, J., Brave, S., Takayama, L.: Improving automotive safety by pairing driver emotion and car voice emotion. In: CHI 2005 Extended Abstracts on Human Factors in Computing Systems, pp. 1973–1976. ACM, April 2005 8. Bradley, M.M., Lang, P.J.: Measuring emotion: the self-assessment manikin and the semantic differential. J. Behav. Therapy Exper. Psychiatry 25(1), 49–59 (1994)

A Proposal for an Affective Design and User-Friendly Voice Agent

255

9. Desmet, P., Overbeeke, K., Tax, S.: Designing products with added emotional value: development and application of an approach for research through design. Des. J. 4(1), 32–47 (2001) 10. Lang, P.J.: Behavioral treatment and bio-behavioral assessment: Computer applications (1980) 11. Morris, J.D.: Observations: SAM: the self-assessment manikin; an efﬁcient cross-cultural measurement of emotional response. J. Advertising Res. 35(6), 63–68 (1995)

Methodologies for the Design of ATM Interfaces: A Systematic Review Joel Aguirre1(&), Arturo Moquillaza1,2, and Freddy Paz1 1

Pontiﬁcia Universidad Católica del Perú, Lima 32, Lima, Peru {aguirre.joel,amoquillaza,fpaz}@pucp.pe 2 Universidad San Ignacio de Loyola, Lima 12, Lima, Peru [email protected]

Abstract. Usability is nowadays an aspect that determines the success of any software product, and the embedded systems in the Automated Teller Machines (ATMs) are not the exception. This fact represents a challenge for the software developers that are forced to guarantee the design of usable interfaces. Some attempts to achieve the required degree of usability in a software of this domain, involve the application of user-centered techniques in the elaboration process of the design proposals for the functionality of the ATMs. However, there is a lack of a formal methodological procedure. In this research, a systematic review was conducted to identify the approaches that are used to guide the design process of ATM interfaces. From a total of 492 studies, only 10 were selected as relevant. Although usability is an important aspect, only a small number of authors report the use of frameworks, that are still in the process of validation. Keywords: User interfaces Automated Teller Machine

Systematic review User-centered design

1 Introduction Through the years, advances in Information and Communications Technologies have allowed transforming the way in which any kind of organizations offer their services, and according to Camilli [1], multiplying their channels and point of contact with their customers. In this sense, according to the study by ASBANC [2] (a Peruvian consortium of banks), the ﬁnancial system has undergone a series of major changes linked to the way in which ﬁnancial entities relate to their users. For example, today ﬁnancial clients can carry out many of their operations through alternative channels to traditional bank branches. According to ASBANC [2] and Mahmood and Shaikh [3], ATMs (Automated Teller Machines) were the ﬁrst major effort of banks to decentralize the delivery of their services. The ﬁrst ATM was installed by Barclays Bank in 1967, and nowadays there is practically no bank that operates without one. In accordance with Kamﬁroozie and Ahmadzadeh [4], ATMs have evolved, from being machines with the sole purpose of delivering money, to be complex machines capable of doing more ﬁnancial operations. However, in line with Zhang et al. [5], this increase in the functions of an ATM can generate confusion in users, causing the loss of the sense of security in the system to © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 256–262, 2019. https://doi.org/10.1007/978-3-030-02053-8_39

Methodologies for the Design of ATM Interfaces

257

have more complex menus and eventually stop choosing that channel as a transactional medium. According to Moquillaza et al. [6], the low consideration for the needs of the end user in the design of ATM interfaces has generated serious problems of usability and accessibility, generating discomfort and inducing operational errors. This may be due to the fact that although there are many design guides for web interfaces and mobile devices, there is little evidence on how to apply these principles to ATM interfaces. The paper is organized as follows: Sect. 2, where we describe the process to design and perform the systematic review, Sect. 3, where we detail the obtained results, and Sect. 4, where we present conclusions and future work.

2 Systematic Review The systematic review was conducted based on the methodology proposed by Kitchenham and Charters [7] [8] with the aim to reduce the researcher bias and conduct an impartial review. The phases proposed for the review were review planning, conduct of the review, and result analysis. 2.1

Research Questions

The deﬁnition of the research question was based on the PICOC (Population, Intervention, Comparison, Outcome, Context) method. The criteria proposed by Petticrew and Roberts [9] were used to deﬁned general concepts in order to formulate the research questions: • Population: Automated Teller Machine user interfaces. • Intervention: User interfaces usability. • Outcome: Usability methodologies and methods for the design of highly usable user interfaces. • Context: Academic and business context. The formulated questions were: • RQ1: Which usability methodologies or frameworks have been proposed for the design of ATM interfaces? • RQ2: Which usability methods or techniques have been used in the design of ATM interfaces? • RQ3: Which usability methods or techniques have been used in the design of user interfaces in the banking software domain? 2.2

Search Strategy

Based on the PICOC criteria described in the previous section, we deﬁned our search strategy. The aim of having a search strategy was to deﬁne a way to select the most relevant studies that could answer the research questions.

258

J. Aguirre et al.

First, we deﬁned the search terms taking into consideration the PICOC criteria, transforming the original deﬁnitions into search terms. An evaluation of the different viable options for the search terms was performed, and we obtained the following search strings: • C1: (“methodology” OR “method” OR “framework” OR “guidelines” OR “design” OR “principles”) • C2: (“user-centered” OR “usable” OR “ATM” OR “Automated Teller Machine” OR “Automatic Teller Machine” OR “ATM interface” OR “user interface” OR “bank” OR “banking software”) • C3: (“usability” OR “interface” OR “user interface” OR “UX” OR “user experience”) Then, the basic search string used was: C1 AND C2 AND C3 Then, we deﬁned a two phases process to conduct the systematic review: the primary search and the secondary search. The ﬁrst phase, the primary search, was performed using four online databases, well known by their references for scientiﬁc articles and journal, conference proceedings and technical papers: ACM, ScienceDirect, IEEE Xplore and SpringerLink. The secondary search was performed as the second phase of the search process. We used the more relevant citations and references obtained during the primary search. 2.3

Study Selection

It was necessary to deﬁne a set of selection criteria to select the most relevant studies that would help us answer the research questions and exclude the ones which wouldn’t. The inclusion of the studies was determined by the following criterion: The study presents a usability methodology or process that have been applied in the design of ATM interfaces, banking software domain or related. The studies that fulﬁlled one the following criteria were excluded (criteria based on culture and linguistic, target population, and nature of the intervention [10]): • The study only present historical information of the search domain. • The study presents reviews of other studies or is a systematic review. • The study is a book or is not written in English. In addition to these criteria, we selected all the studies between 2011 and 2018 because of the proximity in time and to avoid reporting out-of-date studies. 2.4

Synthesis Strategy

The search for the results of the present systematic review was performed on April 27th, 2018. A total of 492 studies were retrieved from the four consulted databases. Only ten of those studies were relevant following the proposed selection strategy. Once performed the secondary search, we did not ﬁnd any other study that could contribute to the research questions. The amount of identiﬁed studies is shown in Table 1.

Methodologies for the Design of ATM Interfaces

259

Table 1. The total amount of papers classiﬁed by their origin database. Database Found papers Duplicated papers Relevant papers ACM 322 0 2 IEEE 39 0 3 Science direct 47 0 1 Springer 84 0 4 TOTAL 492 0 10

3 Results In order to answer the research questions, the reported studies were classiﬁed based on how they contribute to this objective. We present in the next subsections a detailed analysis of the results that contribute to answering each research question. 3.1

Frameworks for the Design of ATM Interfaces

The systematic review results reported only one study proposing a methodology for the design of ATM interfaces. In the study by Moquillaza and Paz [11], a user-centered design approach is used to propose a framework for the design of the ATM interfaces of a well-known ﬁnancial institution in Perú. The proposal was applied by the ATM developers team of the bank in mention and used an eight-step process to design the ATM interfaces. The methodology has a user-centered approach, and it introduces two techniques: video prototyping and storyboarding. Finally, they performed user tests with internal users from the bank; however, both authors agree that it is necessary a quantitative validation with formal methods. 3.2

Methods and Techniques for the Design of the ATM Interfaces

The results of the present systematic review also reported a considerable percentage of studies applying user-centered methods and techniques in the design of ATM interfaces. In the study of Camilli et al. [1], they used heuristic evaluation to empirically recognize the most common and frequent mistakes the user may incur into. The information gathered with the technique allowed the authors to redesign the user interface of the ATMs of a bank in Italy. Moreover, Zhang et al. [5] used heuristic evaluation and user tests to assess the usability and user experience in the design of user interfaces for ATMs. Moquillaza et al. [6] with the support of a group of undergraduate students from the University of Cauca, applied user-centered techniques, such as activity theory, user proﬁling, and user tests, to design user interfaces along one semester. The ﬁnal interfaces were approved by the stakeholders, one of the leader banks in Perú, BBVA Continental. In order to improve the usability in ATMs, Kamﬁroozie and Ahmadzadeh [4] proposed a model to design ATMs interfaces using personalization; using the historical transactions of users to predict actions and reduce the time they spend using an ATM.

260

3.3

J. Aguirre et al.

Methods and Techniques for the Design of Banking Software Interfaces

In the study conducted by Gumussoy [12], the author proposed a set of heuristics as a guideline for the design of user interfaces for banking software, making the acknowledge of the necessity of a guideline of methodology that allows us to design user interfaces to overcome the difﬁculties that the complex ﬁnancial system is involved into. Realpe-Muñoz et al. [13], proposed a process for the design of usable and secure authentication systems that involves the next activities: user requirements analysis, the empirical design made by the domain experts and heuristic evaluation to validate the interfaces designed, unfortunately, this process is incomplete because of lack of validation. Siebenhandl et al. [14] proposed a process to design a friendly interactive kiosk; in this case, a ticket vending machine, that introduces brainstorming for the design and user test and user stories for the evaluation of the interfaces designed. In the study of Wardhana et al. [15], the authors proposed a user-centered design process to improve the security perception of the user interfaces of a mobile app for communication between parents and their children; besides the requirements analysis, they proposed prototyping and an evaluation based on the ISO 9241 standard’s QUIM model to validate the usefulness of the process to improve the security perception in the ﬁnal users. Adama et al. [16] proposed a process to design mobile banking usable interface for novice users. The authors gathered information by conducting a literature review in order to ﬁnd the most common constraint that low literate and novice user had. Then, they design a prototype that would be tested and evaluated with user tests and SUS questionnaires.

4 Conclusions and Future Works The present systematic review had the objective of gathering the studies about methodologies and methods applied to the design of ATM user interfaces. It was performed following a predeﬁned protocol review. In this way, we identiﬁed 492 studies, from which just only 10 papers were selected as relevant for the research questions. From this analysis, we concluded that: • There is a lack of formal methodologies or frameworks that allow the design of usable interfaces for ATMs. Just one framework that uses a user-centered approach and introduces activities such us, video prototyping and storyboarding. • Most of the studies report the application of usability methods focusing in the requirement analysis, prototyping and usability evaluations in the ATM, banking software and security domain. However, they do not deﬁne a framework for the design of the interfaces. • The frameworks and design processes have not been validated. They are partially validated or do not have conducted any validation test. Most of the selected studies validate their proposals with empirical experiments and qualitative results.

Methodologies for the Design of ATM Interfaces

261

The results establish a lack of information about how to apply the usability knowledge in the ATM user interface design. This may cause several usability problems for the users when they use an ATM. As part of our future work, we will update and extend the present systematic review throughout other related domains such as the self-service domain, because it encloses the interactive kiosks, which the ATMs are an example, and secure and interactive systems, two important features of ATMs.

References 1. Camilli, M., Dibitonto, M., Vona, A., Medaglia, C.M., Di Nocera, F.: User-centered design approach for interactive kiosks: evaluation and redesign of an automatic teller machine. In: Proceedings of the 9th ACM SIGCHI Italian Chapter International Conference on Computer-Human Interaction: Facing Complexity, pp. 85–91. ACM, New York (2011) 2. ASBANC: Asociación de Bancos del Perú: Impacto Económico del Uso de los Cajeros Automáticos en el Perú (2016). https://goo.gl/NTWzDQ. Accessed 14 Aug 2018 3. Mahmood, T., Shaikh, G.M.: Adaptive automated teller machines. Expert Syst. Appl. 40(4), 1152–1169 (2013) 4. Kamﬁroozie, A., Ahmadzadeh, M.: Personalized ATMs: improve ATMs usability. In: Stephanidis, C. (eds.) HCI International 2011 – Posters’ Extended Abstracts. HCI 2011. Communications in Computer and Information Science, vol. 173, pp. 161–166. Springer, Berlin, Heidelberg (2011) 5. Zhang, M., Wang, F., Deng, H., Yin, J.: A survey on human computer interaction technology for ATM. Int. J. Intell. Eng. Syst. 6(1), 20–29 (2013) 6. Moquillaza, A., Molina, E., Noguera, E., Enríquez, L., Muñoz, A., Paz, F., Collazos, C.: Developing an ATM interface using user-centered design techniques. In: Marcus, A., Wang, W. (eds.) Design, User Experience, and Usability: Understanding Users and Contexts. DUXU 2017. LNCS, vol. 10290, pp. 690–701. Springer, Cham (2017) 7. Kitchenham, B., Charters, S.: Guidelines for Performing Systematic Literature Reviews in Software Engineering. Technical Report. Keele University and Durham University Joint Report (2007) 8. Salvador, C., Nakasone A., Pow-Sang, J.A.: A systematic review of usability techniques in agile methodologies. In: Proceedings of the 7th Euro American Conference on Telematics and Information Systems, article no. 17. ACM, New York (2014) 9. Petticrew, M., Roberts, H.: Systematic Reviews in the Social Sciences: A Practical Guide. Wiley, Hoboken (2005) 10. Meline, T.: Selecting studies for systematic review: inclusion and exclusion criteria. Contemp. Issues Commun. Sci. Disord. 33, 21–27 (2006) 11. Moquillaza, A., Paz, F.: Applying a user-centered design methodology to develop usable interfaces for an automated teller machine. In: Proceedings of the XVIII International Conference on Human Computer Interaction, article no. 26. ACM, New York (2017) 12. Gumussoy, C.A.: Usability guideline for banking software design. Comput. Hum. Behav. 62 (C), 277–285 (2016) 13. Realpe-Muñoz, P., Collazos, C.A., Granollers, T., Muñoz-Arteaga, J., Fernandez, E.B.: Design process for usable security and authentication using a user-centered approach. In: Proceedings of the XVIII International Conference on Human Computer Interaction, article no. 42. ACM, New York (2017)

262

J. Aguirre et al.

14. Siebenhandl, K., Schreder, G., Smuc, M., Mayr, E., Nagl, M.: A user-centered design approach to self-service ticket vending machines. IEEE Trans. Prof. Commun. 56(2), 138– 159 (2013) 15. Wardhana, S., Sabariah, M.K., Effendy, V., Kusumo, D.S.: User Interface Design model for parental control application on mobile smartphone using user centered design method. In: Fifth International Conference on Information and Communication Technology, pp. 1–6. IEEE, New York (2017) 16. Adama, V.N., Shehu, I.S., Adepoju S.A., Jimoh, R.G.: Towards designing mobile banking user interfaces for novice users. In: Marcus, A., Wang, W. (eds.) Design, User Experience, and Usability: Designing Pleasurable Experiences. DUXU 2017. LNCS, vol. 10289, pp. 181–197 (2017)

Design and Development of an Image-Based System to Facilitate Reading Comprehension of Chinese Classic Literature Tung-En Chien1(&) and Shelley Shu-Ching Young2 1 Institute of Information Systems and Applications, National Tsing Hua University, No. 101. Section 2 Kuang-fu Road, Hsinchu 30013, Taiwan (R.O.C.) [email protected] 2 Institute of Learning Sciences and Technologies, National Tsing Hua University, No. 101. Section 2 Kuang-fu Road, Hsinchu 30013, Taiwan (R.O.C.) [email protected]

Abstract. This study aims to develop an image-based reading system to help people increase reading comprehension of Chinese classic literature. Nowadays, people have new reading habits with the uses of handheld devices, such as IPhones, IPads. Thus, they are used to reading shorter messages and simple Chinese. It becomes an issue for younger generation, digital natives, to read classical Chinese literature. Therefore, in this study we, using the classical Chinese literature, Dream of the Red Chamber, as an example, try to develop a handheld-device-based system that combines graphic (word clouds) and the Internet to solve this problem. After the system is implemented, we will conduct this study to the college students, who have not read the Dream of the Red Chamber before. Participants will read some speciﬁc chapters. Before reading, they will use the word clouds function and create their own infographic. We will administer questionnaires to investigate the effectiveness of the image-based reading system. Keywords: Word cloud

Infographic Reading comprehension

1 Introduction Nowadays, the emerging multimedia and handheld devices such as smart phones and ereaders could provide people with many alternatively convenient ways of reading. Thus, people’s reading habits have also changed. People now enjoy reading short messages, short articles via the quick-scan approach. People become less patient in reading long works. New technology lets the people get interrupted easily when they are reading and feel difﬁcult to read the long literature like classic literature in long duration [1]. However, Chinese classic literature is very important, because they carry the heritage of Chinese culture and life value. They not only can help people develop good Chinese literacy, but also can get people inspired in life. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 263–267, 2019. https://doi.org/10.1007/978-3-030-02053-8_40

264

T.-E. Chien and S. S.-C. Young

So this study aims to develop an Image-based Reading system to support people’s Chinese classical literature reading. The system will combine ﬁgure (graphics) to meet modern people’s reading habits so as to enhance their content understanding. In this study we will use the well-known but very complicated Chinese classic novel, Dream of the Red Chamber, as an example for the system design. Word clouds and Infographics are the main graphics we utilize in this system.

2 Literature Review Word Cloud Word cloud (tag cloud), is a visualization of text or content. Tags are usually single words, and they will present in different font sizes and colors. This format is useful for quickly perceiving the most prominent parts by those characteristic. A word cloud is a visualization of word frequency in text ﬁle. Martin Halvey points some feature about wordclouds, and font size is helpful ﬁr users to ﬁnd important information quickly and easily. Position of tags is also important, and more users scan lists and clouds rather than read them [3]. Lohmann et al. conducted a comparative study on several tag cloud layout methods. They ﬁnd the tags’ font sizes had a strong effect on search speed. And more tags in the middle of the cloud attract more user attention than tags near the borders. Tags in the upper left quadrant are found more quickly [2]. Schrammel et al. conduct a study about Semantically Structured Tag Clouds. In this study they found the popularity of tag clouds to be explained by their ability to help users in getting a fast overview of a certain area. For different layouts, topically layouts can improve search performance for speciﬁc search tasks compared to random arrangements, and the semantic arrangement must be good enough. Otherwise, users will not be able to distinguish it from the random layouts [4]. Thus we know that tag’s arrangement is important, but most of the tag cloud visualizations do not allow users to change single tag’s positions, and just can change the cloud shape, font size or color. ManiWordle, a more flexible Wordle-based visualization [5], allows people to manipulate topography, color, position, and orientation of the individual words. It can let people design a customized word cloud in this study.

2.1

Infographics

With today’s overflow of information, people enjoy reading short messages, short articles via the quick-scan approach to get information. Infographics can help readers understand the key information quickly. Infographic is deﬁned as a visualization of data or ideas that tries to convey complex formation to an audience in a manner that can be quickly consumed and easily understood [6]. A good infographic will not only tell readers story, but also let reader want to know more about it. There are some major types of infographics: statistical based, timeline based, process based and location or geography based [7]. In this study we will choose the suitable type for the system.

Design and Development of an Image-Based System

265

3 System Design To support readers to read Chinese classic literature, we will use two types of graphics: Word clouds and Infographic in this system, and use the well-known but very complicated Chinese classic novel, Dream of the Red Chamber, as an example for the system design. We will introduce how to use them to help people deepen reading understanding to Chinese classic literature as follows (Fig. 1).

Fig. 1. System architecture diagram

Word Clouds Section. We will use some web word cloud generator to create each chapter’s cloud. Before reading, the readers can get an overview of the key characters, places and events in the book chapter before they read the corresponding chapters by it, and the word clouds help raise reader’s attention when they see these words. Each generator will create different outcomes by the given texts so we will continue to ﬁnd the better generator and use ManiWordle to create our word clouds. After they read, they also can rate the cloud to help the system referentially provide the better cloud to others. Infographics Section. After the they read the chapter, the system will provide readers with a tool to let them generate various types of relevant charts by themselves. Through the process of constructing a diagram, readers can organize what they have read. The output will be a relation chart, a time series chart or a map. The diagrams can demonstrate their understanding of the content. Moreover, they can use these charts to solve the problems of fragmented reading and quickly relocate and recall the contents of the book through the infographic they made.

266

T.-E. Chien and S. S.-C. Young

Moreover, user can share their infographic and see others ﬁnal product, to make sure his understanding is right or not by comparing with others infographic (Fig. 2).

Fig. 2. System usage flow chart

4 Conclusion In this study, we aim to help people to solve the problem of reading classic literature and increase their reading comprehension. Thus we design the image-based reading system with word cloud and infographics to cater people’s reading habits. Currently we are in the stage of system development. In future, we will improve our system by the result of formative evaluation and hope the ﬁnal system can enhance user’s reading comprehension with easy operation and high reading comprehension. Acknowledgments. Thanks for the research grant support from the Ministry of Science and Technology (MOST), R.O.C., under the grant No. 105-2420-H-007-011-MY3.

References 1. Reading: The Struggle. http://www.nybooks.com/daily/2014/06/10/reading-struggle/. Accessed 03 June 2018 2. Lohmann, S., Ziegler, J., Tetzlaff, L.: Comparison of tag cloud layouts: task-related performance and visual exploration. In: IFIP Conference on Human - Computer Interaction, pp. 392–404. Springer, Heidelberg (2009) 3. Halvey, M.J., Keane, M.T.: An assessment of tag presentation techniques. In: Proceedings of the 16th International Conference on World Wide Web, pp. 1313–1314 (2007)

Design and Development of an Image-Based System

267

4. Schrammel, J., Leitner, M., Tscheligi, M.: Semantically structured tag clouds: an empirical evaluation of clustered presentation approaches. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 2037–2040 (2009) 5. Koh, K., Lee, B., Kim, B., Seo, J.: Maniwordle: providing flexible control over wordle. IEEE Trans. Vis. Comput. Graph. 16(6), 1190–1197 (2010) 6. Smiciklas, M.: Power of Infographics Using Pictures to Communicate and Connect with Your Audience. Que Publishing, Indiana (2012) 7. Siricharoen, W.V.: Infographics: the new communication tools in digital age. In: The International Conference on E-technologies and Business on the Web (EBW 2013), pp. 169–171 (2013)

Human Factors Integrated System Validation in the Nuclear Power Plant Main Control Room Xiaoyang Ming(&), Guangwei Yu, and Shiguang Deng Heneng Building, #117 Xisanhuan Road, Haidian District, Beijing, China [email protected], [email protected], [email protected]

Abstract. According to NUREG-0711, Human Factors Veriﬁcation and Validation is one of the ﬁnal stages of Human Factors Engineering Program, determining whether an integrated system’s design conforms to HF design principles, and enables plant personnel to successfully perform their tasks and other operational goals, assuring plant safety. As the center of monitoring and controlling the Nuclear Power Plant, the Main Control Room is an integrated system comprising varieties of Human System Interfaces, procedures, operators and environment, which should all be included in HF V&V activities. Meanwhile, the comprehensive performance of operators interacting with human machine interfaces would also be evaluated by simulating plant operation conditions. Normally, the previous series of HF activities are known as Integrated System Validation, which would be implemented upon a high ﬁdelity Full Scope Simulator identical with the main control room panels. Keywords: Human factors Veriﬁcation and validation Integrated system validation

1 Introduction Main control room, the brain of a Nuclear Power Plant (NPP), is an integrated system comprising varieties of Human System Interfaces (HSI), procedures, operators and environment. As described in NUREG-0711, the Human Factors Engineering Program Review Model, Human Factors Engineering (HFE), as an integral activity throughout the design, operation, testing and maintenance process of NPP, plays a major role in supporting plant safety and providing defense in depth. Among the 12 different elements of HFE safety review, Veriﬁcation and Validation (V&V) is one of the crucial activities which determines whether an integrated system’s design (i.e., hardware, software, and personnel elements) conforms to HF design principles, and enables plant personnel to successfully perform their tasks and other operational goals, assuring plant safety. The HF V&V plan contains four different types of activities as following: (1) HSI Task Support Veriﬁcation (TSV), the activity that ensures HSIs and their characteristics meet all of the operator’s task requirements as deﬁned by the HFE task analysis process. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 268–274, 2019. https://doi.org/10.1007/978-3-030-02053-8_41

Human Factors Integrated System Validation

269

(2) HSI HFE Design Veriﬁcation (DV), the activity evaluates that the implemented HSIs and the environment in which it is used conform to HFE guidelines. (3) Integrated System Validation (ISV), a performance based evaluation of the integrated HSI design. (4) Human Engineering Discrepancy (HED) management, the process of identifying, documenting, evaluating and resolving all of the HEDs detected from the above three V&V activities. For this paper, the detailed introduction of ISV would be given.

2 HF ISV Introduction Generally, validation refers to the tests and evaluations that determine whether a design complies with functionality, performance and interface requirements, in practice, the validation process deﬁnes whether the appropriate design has been implemented. As stated in NUREG 0711, ISV is an evaluation, using performance-based tests, to determine whether an integrated system’s design (i.e., hardware, software, and personnel elements) meets performance requirements and supports the plant’s safe operation. HEDs are identiﬁed if performance criteria are not met. 2.1

ISV Objective

Basically, the HF integrated system validation intends to evaluate the acceptability of those aspects of the design that cannot be determined through such analytical means as HSI task-support veriﬁcation and HF design veriﬁcation. Before validation is performed, the resolution of all signiﬁcant HEDs identiﬁed in veriﬁcation process should have been given. Besides the test bed and scenario requirements, the test team, performance measurement and the test procedure should be planned particularly for the ISV. The data analysis methodology and the acceptance criteria should be considered speciﬁcally about the scenario and HSI features. With a performance based evaluation, ISV aims to conﬁrm: (1) That the HFE-adequacy of the integrated HSI conﬁguration in relation to crew primary tasks, and the adequacy in relation to crew secondary tasks such as navigation and efﬁcient search and retrieval of information and controls. (2) That the HSIs adequacy for crew communication and teamwork, stafﬁng (assignment of tasks to crewmembers) and procedures. (3) That the HSIs support situation awareness, and accomplishing critical functions and tasks under an acceptable operator workload, inclusive of any human performance assumptions from Probabilistic Risk Assessment (PRA) and Human Reliability Analysis (HRA). (4) That the automation degree (allocation of functions and the degree of task dependence on procedures). (5) That the tolerance to human error and system faults. That the HEDs are identiﬁed and documented during the validation process.

270

X. Ming et al.

2.2

ISV Platform

Normally in projects, ISV is performed upon a Full Scope Simulator (FSS), a complete integrated platform with high degree of functional ﬁdelity and environmental ﬁdelity with the Main Control Room (MCR). 2.3

ISV Scope

As the target of ISV is the entire HSI in main control rooms, the HSIs have been developed for the full scope simulator and included in MCR panels would all be validated, including: • The Computerized Operator Working Position (COWP) with the operating displays and computerized alarms, • The Backup Panel (BUP) and the components, • The large display panel, • The overall arrangement of control room, • The environment of control room. 2.4

ISV Items

For a comprehensive evaluation of the crew performance, different dimensions of the process should be validated during ISV as, (1) Monitoring/detection, (2) Situation assessment, (3) Response planning, (4) Response implementation, (5) Secondary tasks and usability, (6) Workload, (7) Situation awareness, (8) Communication, (9) Teamwork. 2.5

ISV Preparation

Before ISV is executed, four initial activities as following need to be implemented as preparation. (1) Operational conditions sampling This process ensures the scenarios which are included in the validation plan cover the representative range of tasks and events under different plant operational conditions, while all of the characteristics of HSIs and environment of the MCR are taken into consideration. The important human actions, systems and accident sequences about safety should also be taken into account for the scenario sampling. In practice, the selected scenarios of operational conditions sampling must contain three kinds of events, • Normal operational events, like start-up, shut-down and load promotion/ reduction, • The failure events occurred in Instrumentation and Controlling (I&C) system and HSIs, • The transient events and accidents events.

Human Factors Integrated System Validation

271

(2) Scenario deﬁnition and documentation This process deﬁnes the selected scenarios, as well as the speciﬁc sequence to be performed in operation. Practically, all the information as follows should also be given in deﬁnition. • • • • •

The initial condition of the scenario, Task support requirement, Crew members requirement, Personal response and Event Form, The terminal condition of scenario.

(3) Participant/crew selection This process accomplish the selection participants for ISV. Two crews would be necessary for one scenario each day, and complete training is required before the crew is devoted into the ISV in order to obtain representative results. (4) Observers team composition & training This process builds and trains the team of observers for the scenarios before the ISV is executed. 2.6

ISV Execution

As illustrated in Fig. 1, the whole process of validation contains following phases: (1) Brieﬁng meeting will take place before ISV execution for informing the crew of both the initial plant status when starting the scenario and the details of the questionnaire at the end of the scenario. (2) Scenario execution (with no freeze points) and data collection during scenario execution. Observers are responsible for collecting data related to crew monitoring/detection, situation assessment, response planning, response implementation, workload, situation awareness, communication and teamwork. Test participants should not be told about the particular scenarios that will be simulated throughout the whole execution. During the execution, observers need to ﬁll out the “Event Form” along with the execution process. (3) Data collection after scenario execution, by means of questionnaires (operators ﬁlling out the “ISV Questionnaire”) or interviews (observers ﬁlling out the “Scenario Assessment Form”) with the crew related to the scenario performed and HSI. Data collection should be present in multi-dimension, to be speciﬁc, time, accuracy, frequency of actions, amount achieved or accomplished, errors, communication, dynamic anthropometrics and crew movement, plant parameters log recorded by simulator or system, ratings from the crew should be taken into account. (4) ISV analysis and report, mainly includes, • Detailed analysis for each scenario, • Composite assessment for all the scenarios, including analysis of questionnaire and interview answers,

272

X. Ming et al. START Briefing meeting Set up the scenario on the simulator Scenario simulation starts Redo preparation

Scenario simulation … Scenario simulation ends Interview and questionnaire N ISV results reflect the operational condition Y END

Fig. 1. The process of ISV execution

• Preparation of the summary report, including summary of scenarios, summary of interview and questionnaire answers, analysis of acceptance criteria fulﬁllment, preparation of validation conclusions, summarizing positive results, recommendations, improvements, modiﬁcations and conclusions, preparation of HEDs that must be tracked. 2.7

ISV Report

As a summary of the validation activity, ISV report must be presented, documenting the whole process of ISV. Basically, the report shows the objectives, scope, comprehensive description of methodology and the summary of the results (positive or negative), fulﬁllment of acceptance criteria and the ISV conclusions. 2.8

HEDs Management

During the whole validation process, HEDs must be identiﬁed, documented, evaluated and resolved and veriﬁed in the end.

Human Factors Integrated System Validation

273

(1) HED identiﬁcation and description In general, HED will be identiﬁed as such a deviation is found from: • A standard or convention of human engineering practice, • An operator preference or need, • An issue that adversely impacts personnel performance of plant related functions or tasks (increases the likelihood of human error), • A non-compliance between the designed display and the implemented FSS display, • Other system design requirement for the capabilities and roles of operator. Once the HEDs is identiﬁed, related description about the reference and the explanation of identiﬁcation process should be given in HED documenting forms. (2) HED prioritization To give a more effective suggestion to HSI designers, the HEDs should be prioritized in the documenting forms into 3 categories: • Priority 1, the HEDs which lead to signiﬁcant safety related consequences, • Priority 2, the HEDs which lead to no signiﬁcant safety related consequences, but has impact on the plant/personnel performance, • Priority 3, other HEDs. (3) HED evaluation and resolving V&V team and HSI designers will cooperate in this process to evaluate the HEDs and to reach the ﬁnal agreement of whether accept it or not. (4) HED corrections veriﬁcation With the third stage of HEDs management is done, all the HEDs should be either corrected with a revise or rejected with an appropriate explanation, which will be veriﬁed by V&V team until all of the HEDs are closed.

3 Summary A comprehensive methodology of human factors integrated system validation for nuclear power plant main control room is given in this paper. Built upon reference from NUREG 0711, this methodology is veriﬁed by several NPP engineering projects in China. The result of ISV has not only served as valuable feedbacks to HSI design, but also give powerful proof of the plant performance. Furthermore, the analysis methods of the data and questionnaire, and the evaluation ways of the crew performance are still waiting to be optimized in the future. With the practical experience grows, the ISV activity is aiming to be play more and more important role in the design of MCR HSIs and the operation of NPP.

274

X. Ming et al.

References United States Nuclear Regulatory Commission: Human Factors Engineering Program Review Model, revision 3 (2012) Ming, X.: HFE Application in Human System Interface Design of Nuclear Power Plant. Springer, Heidelberg (2017)

The Functional Requirement Analysis in Nuclear Power Plant Human System Interface Design Shiguang Deng(&), Xiaoyang Ming, and Guangwei Yu China Nuclear Power Engineering Co., Ltd., No. 117 Xisanhuanbeilu, Haidian District, Beijing, People’s Republic of China {dengsg,yugw}@cnpe.cc, [email protected]

Abstract. Human System Interfaces (HSIs) are the medium through which operators can monitor and control the Nuclear Power Plant. To ensure the Main Control Room (MCR) design is sufﬁcient to accomplish plant goals, main control room designers would conduct functional requirements analysis to identify the information and control requirements. In consideration of validity and rationality, functional requirement analysis would be performed ahead of the design for various human system interfaces. To identify the control and indication information of the HSIs, the operator task requirements analysis could be performed with the function decomposition. In the task analysis, the cognitive and decision making technology/model could be used to deﬁne the information selection criteria. One of the most popular decision making model is the Rasmussen model. The model deﬁnes the 8 steps of the human cognitive procedure which could be used to identify the components and parameters for the HSIs design. Keywords: Nuclear Power Plant Main Control Room Human System Interface Functional requirement analysis

1 Introduction As the brain of Nuclear Power Plant (NPP), Main Control Room (MCR) is an integrated system, which comprises varieties of Human System Interfaces (HSIs), procedures, operators and environment. In of the modern nuclear power plants, the hybridization of both digital and traditional human system interfaces is equipped in the main control room In consideration of validity and rationality, functional requirement analysis would be performed ahead of the design for various human system interfaces. Functional requirements analysis is the identiﬁcation of functions that must be performed to satisfy the nuclear power plant’s overall goals, (1) to ensure the health and safety of the public by preventing or mitigating the consequences of postulated accidents; (2) to generate power (supply electricity to the grid). A frequently used functional requirements analysis methodology is function decomposition which starts at a high level function and continues to lower levels until a speciﬁc critical end item requirement emerges (to the level of equipment or parameter). To identify the control and indication information © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 275–281, 2019. https://doi.org/10.1007/978-3-030-02053-8_42

276

S. Deng et al.

of the HSIs, the operator task requirements analysis could be performed with the function decomposition. In the task analysis, the cognitive and decision making technology/model could be used to deﬁne the information selection criteria. One of the most popular decision making model is the Rasmussen model. The model deﬁnes the 8 steps of the human cognitive procedure which could be used to identify the components and parameters for the HSIs design. This paper has described the methodologies for the application of function requirement analysis in the HSI design.

2 Human System Interfaces in the NPP Main Control Room In of the modern nuclear power plants, the hybridization of both digital and traditional human system interfaces is equipped in the main control room (Fig. 1).

Fig. 1. HSIs in the NPP MCR

The main control means to operate the plant relies on the software-based operating displays on the visual display units (VDUs). Beneﬁts from the popularization of digital technology, the soft operating displays could provide plentiful and effective information and control means to operators, according to which, the safety and performance of the plant would be both improved. Once the main control means is unavailable, the back-up panel (BUP) could be used as a substitute for the plant supervision. BUP is the traditional hardware-based control panel on which the instruments, buttons, switches and other control facilities for the plant operation are installed. While the software based operating displays would be used for all the plant conditions, BUP is mainly used for the normal operation, shutdown and accident mitigation. To fulﬁll the different operating goals, varieties of operating displays are designed: • System operating displays are the basic operating displays that are used for the system control and monitoring of the plant; • Function monitoring displays are used for the high-level function monitoring during different conditions. The operators could ﬁnish the plant inspection as soon as possible with the function monitoring displays during shift of duty and periodic monitoring;

The Functional Requirement Analysis in Nuclear Power Plant

277

• Large screen displays could provide the plant condition information to all the crew in the MCR; • Besides the displays above, the particular display could be designed for a special operating task that is important or frequently used in operation. The software-based operating displays are designed to fulﬁll the operating goals of all the working conditions that include start-up, shutdown, normal operation, transients, malfunction and all the accidents (Fig. 2).

(1)

(2)

Fig. 2. The example of software based displays (1. system operating display; 2. function monitoring display)

Compared to the flexible and plentiful displays, the back-up panel (safety panel) interface is limited which means that the control and indications on the BUP are limited (Fig. 3).

Fig. 3. The example of back-up panel

278

S. Deng et al.

The BUP was designed for the following operating goals: • When Main Control Means (MCM) is available: BUP supplied plant operation support to provide necessary information for the safety assessment of the unit state in the accident conditions. • When MCM is unavailable during the normal operation, operators maintain plant in operation through BUP. When the operation is overtime: Operators would use BUP to take the plant to safety operation mode. • When the MCM is not available during accident: operators should monitor the safety actions through BUP, and take plant to safety operation. The goals above show that the functions of the BUP are focused on the functions that are related to safety and the shutdown of the plant. To design the diverse human system interfaces, the function requirement analysis is necessary. The function analysis could be performed with different levels, scope and methodologies that depend on the design object.

3 Functional Requirement Analysis in the HSI Design The identiﬁcation of functions for the NPP is obtained ﬁrst by deﬁning general or fundamental objectives of the plant operation, i.e. protection of the public from radiological hazards, safe and effective generation of electrical power. Then by breaking down the top-level functions, allowing those objectives to be fulﬁlled into hierarchy of functions until a speciﬁc critical end-item requirement emerges (e.g., a piece of equipment or parameter.). The breaking down methodology is function decomposition, a common function analysis tool. 3.1

Function Decomposition

The function decomposition usually starts at high-level functions where a very general picture of major functions is described and continuing to lower levels. High-level functions are usually accomplished through some combination of lower-level system actuations. Often plant systems are used in combination to achieve a high-level function. The combination of systems used to achieve a high-level function is called a process. There may be more than one possible process that can achieve a given highlevel function. The function decomposition should address the following levels: • • • •

High-level functions (e.g., Control reactivity); The processes, as appropriate, those enable achievement of these functions; Speciﬁc plant systems; Components and parameters.

Usually, the decomposition structure should contain the levels above, as indicated in Fig. 4. The model in Fig. 4 is a decomposition model for the whole plant. However, the analysis scope and hierarchy can be adjusted according to the HSIs to be designed. To design an operating display for the function: control reactivity, the function can be

The Functional Requirement Analysis in Nuclear Power Plant

279

Ensure Safety

Generate Power

Control Reactivity Reactor Trip

RCS Integrity

Residual Heat Removal

Containment

Boration

Volume and Chemical Control System

Safety Injection System

Safety Injection Pump

Suction Valve

Fig. 4. Plant function decomposition and hierarchy model of a commercial NPP

decomposed from the top function: control reactivity. While for the display of the system: safety injection system, the function could be decomposed from the top function: safety injection; the processes are: suction, transport and discharge; and the bottom level would be components and parameters of safety injection system. When the function decomposition structure is constructed, the function of the HSI is deﬁned. Nevertheless, it is not enough to identify all the information that should be included in the HSI. The task analysis could be implemented to identify all the components, parameters, alarms and messages. 3.2

Task Analysis

Task analysis is the identiﬁcation of requirements for accomplishing these tasks, i.e., for specifying the requirements for the HSI, data processing, controls, and job support aids needed to accomplish tasks. Use of explicit task analysis therefore leads to more efﬁcient and effective integration of the human element into the system design. To perform the task analysis, the operating conditions should be analyzed adequately with all the typical conditions and scenarios included. To identify the basic information, the following information should be deﬁned: • • • • • •

Purpose of the function; conditions that indicate that the function is needed; parameters that indicate that the function is available; parameters that indicate that the function is operating; parameters that indicate the function is achieving its purpose; parameters that indicate that operation of the function can or should be terminated.

The operating procedures are important input for the task analysis. The control and indication requirements could be identiﬁed from the action analysis of each step in the procedure. However, it is not possible to acquire the procedure for a new reactor design. For this condition, the cognitive and decision making technology/model could be used to deﬁne the HSI information selection criteria.

280

S. Deng et al. Function Decomposition Condition and Scenario Sampling

Task Analysis

Procedure

Cognitive and Decision Making Model

Fig. 5. The HSI requirements analysis process

The analysis process is illustrated in Fig. 5. 3.3

Cognitive and Decision Making Model: Rasmussen Model

The Rasmussen Step-ladder is a common-used cognitive and decision making model. The model separated the cognitive process into 8 steps: stimulation, observation, identiﬁcation, explanation, evaluation, task deﬁnition, procedure deﬁnition and procedure implementation. The eight step model is a variation on the well-known analogy between information processing and a simple closed-loop controller. The cognitive task analysis is performed by a set of questions derived from the Rasmussen model onto each node in the function decomposition structure, and could produce plant systems and equipment control data requirements. While in the HSI design task analysis, the model could be used with the process in Fig. 6.

Evaluation

Identify

Observation

Stimulation

Parameters/ Components

Parameters/ Components /Messages

Alarms/ Parameters

Parameters/ Components

Task Definition

Parameters/ Components

Task Components Implementation

Validation

Parameters

Fig. 6. The cognitive and decision model in the task analysis

Combined with the function decomposition and working condition results, the cognitive task analysis could identify the components, parameters, alarms and messages for the HSI design.

The Functional Requirement Analysis in Nuclear Power Plant

281

4 Conclusion The paper has described the methodologies for the application of function requirement analysis in the HSI design. However, in the actual design, the function analysis is not enough to identify all the information requirements. The task analysis is an important supplement step for the design process. As a new reactor without enough reference design input, the cognitive task analysis is a useful tool for the preliminary design.

References 1. O’Hara, J.M., Higgins, J.C., Fleger, S.A., Pieringer, P.A.: Human Factors Engineering Program Review Model (NUREG-0711, Rev. 3). Technical report. U.S Nuclear Regulatory Commission (2012) 2. Kirwan, B., Ainsworth, L.K.: A Guide to Task Analysis: The Task Analysis Working Group. CRC Press, New York (1992)

A Preliminary Study on Color and Grayscale Images Object Recognition and Scene Classiﬁcation Tasks on Amazon Mechanical Turk Crowdsourcing Platform Aimee Yun-Fang Lin(&), Shelley Shwu-Ching Young, Harrison Pang-Sheng Lai, and Danna Gurari 1616 Guadalupe Street, Austin, TX 78701, USA [email protected]

Abstract. Nowadays most of the information on the Internet is displayed in color images. When it comes to image-related tasks, it becomes critical for us to understand how efﬁciently the color images are processed for fulﬁlling the targeted tasks in terms of accuracy and time. Thus, in this study we conducted an initial study to explore how color and grayscale images were likely to affect the results of the crowdsourcing works by using the novel crowdsourcing platform, Amazon Mechanical Turk, so as to better improve the efﬁciency of the works for image design and to better utilize database space for enterprises at the same time. Eventually some signiﬁcant ﬁndings, suggestions and suggestions for future study will be presented at the end of this paper. Keywords: Human-Centered design Human computer interaction Object recognition Scene classiﬁcation

1 Introduction A signiﬁcant amount of online crowdsourcing tasks are related to image works that need repetitive activities. We found that, in image crowdsourcing, most of the tasks are with color images; however, what will happen if grayscale images are used in the tasks to replace the color images? Moreover, since the data size of a color image is three times bigger than the grayscale one, will it be an advantage for the requesters and the company to build a crowdsourcing database based on grayscale images? With the Amazon Mechanical Turk (AMT) platform we use in this research, we are able to get data about this particular issue easily. Furthermore, since we are speciﬁcally focusing on improving the quality of AMT workers, conducting experiment in the same environment will help us better understand the quality related to workers.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 282–286, 2019. https://doi.org/10.1007/978-3-030-02053-8_43

A Preliminary Study on Color and Grayscale Images

283

2 Related Work Object Recognition. Object recognition is a process for identifying a speciﬁc object in a digital image or video, and related crowdsourcing work is letting the workers do this kind of identifying tasks online. Previous studies related to object recognition focused on the effects of color and the texture [2–4], there is limited information about the effects of color for online crowdsourcing object recognition tasks, especially on Amazon Mechanical Turks. Additionally, previous experiments about online object recognition often used color images for creating the tasks. However, we aims to explore the differences between color images and grayscale images for object recognition. Scene Classiﬁcation. Crowdsourcing works on scene classiﬁcation require workers to distinguish and label the scene shown on the image. Previous studies give us knowledge about the ﬁeld of scene classiﬁcation. [5, 6] Color and texture information in scene classiﬁcation were also discussed in a patient with visual form agnosia in the past. [7] However, color images were mainly applied on the previous online scene classiﬁcation crowdsourcing tasks; on the other hand, the grayscale images versus color images are not widely explored. Thus, this is our hope for the experiment to take a deeper look on this topic. Psychology. Theory and empirical work related to color and psychological functioning [8–10] have been developed relatively slow and insufﬁcient to draw a conclusion on the basis of evidence. Therefore, looking into this speciﬁc area, especially focusing on grayscale and color images via “online crowdsourcing platform”, is one of our motivations for this study.

3 Experimental Design 3.1

Purpose

The main purpose of the experiment is to test whether the color of the images will affect the testing users on Amazon Mechanical Turk (AMT) when doing crowdsourcing tasks or not. If the grayscale image may not affect crowdsourcing tasks, we can then store more grayscale images in the database considering the fact that the size of the same picture in color scale tend to be three times larger. Moreover, color images contain more information, and in easier tasks, such as object recognition, people don’t need that much information to distinguish the object, grayscale image may shorten the time the users need to ﬁnish the tasks. 3.2

User Interface

The same interface was used for both image recognition and scene classiﬁcation tasks. With the step-by-step instructions on the top, one picture was put in the instructions to help users perceive easily. Furthermore, we designed the rectangle boxes with color indicating direct selection feedback. See Fig. 1.

284

A. Y.-F. Lin et al.

Fig. 1. User interface wireframe

3.3

System Design and Data Collection

For this human-centered study, overall it contained 12 tasks, 6 object recognitions and 6 scene classiﬁcations. Furthermore, two types of color and grayscale versions were presented below each of the object recognition and scene classiﬁcation. Under the color and the grayscale version, there were 3 items (an elephant, a ferry, and a lotus) designed for object recognition and another 3 items (coast, forest, and highway) designed for scene classiﬁcation. For each of the items, 6 related-to-the-item and nonrelated-to-the-item images were presented for selection. On the web, a detailed step-bystep instruction was given to the participants. In other words, we used 2 datasets, Caltech 101 and 8 scene categories, and further designed 2 types of image crowdsourcing tasks, object recognition and scene classiﬁcation, with color and grayscale versions to ﬁnd out more insights. The data collection was conducted in late November and early December in 2017.

4 Experimental Results Consequently, a total of 720 answers were collected for data analysis from the crowdsourcing workers on the Amazon Mechanical Turk platform. We compared the datasets in terms of the percentage of accuracy for the images and the average task complete time. Results of this study indicate that color does play an important role in scene classiﬁcation. On the one hand, in object recognition tasks, the workers perform better in grayscale images than the color images by 1.3%. However, the average amount of time needed for grayscale image tasks is 26.3 s compared to 20.4 s for color images, extra 28% of time needed. On the other hand, in scene classiﬁcation, color version is both better than grayscale version in terms of the percentage of accuracy and the average amount of time needed to ﬁnish the tasks. The tables of the results are in Fig. 2.

A Preliminary Study on Color and Grayscale Images

285

Fig. 2. Object recognition and scene classiﬁcation

5 Conclusion and Future Study Even though lots of research have mentioned the ability of color to affect the quality of perception and help human to recognize objects, the situations with the online crowdsourcing tasks are still quite unique ﬁelds and yet to be discovered. Based on our ﬁndings, when it comes to decide what kind of images to be used for the systems, we can take the importance of the time factor into considerations and make further decisions of using the color and grayscale images. Eventually, discussions based on the ﬁndings and suggestions for future study are presented with the hope to both improve the efﬁciency of the works for image design and better utilize database space for enterprises. In our experiment, we ﬁnd out that we can improve the accuracy of the results by choosing the images that won’t let the workers misunderstand so that we will have better results in the future. Furthermore, adding more samples of images and improving and standardizing the criteria for choosing those images will be the future steps for this research.

References 1. Ostergaard, A.L., Davidoff, J.B.: Some effects of color on naming and recognition of objects. J. Exp. Psychol. Learn. Mem. Cogn. 11(3), 579 (1985) 2. Wurm, L.H., et al.: Color improves object recognition in normal and low vision. J. Exp. Psychol. Hum. Percept. Perform. 19(4), 899 (1993) 3. Bramão, I., et al.: The role of color information on object recognition: a review and metaanalysis. Acta Psychol. 138(1), 244–253 (2011) 4. Rossion, B., Pourtois, G.: Revisiting Snodgrass and Vander- wart’s object pictorial set: the role of surface detail in basic-level object recognition. Perception 33(2), 217–236 (2004) 5. Boutell, M.R., et al.: Learning multi-label scene classiﬁcation. Pattern Recognit. 37(9), 1757–1771 (2004) 6. Li, L.-J., et al.: Object bank: a high-level image representation for scene classiﬁcation & semantic feature sparsiﬁcation. In: Advances in Neural Information Processing Systems, pp. 1378–1386 (2010) 7. Steeves, J.K.E., et al.: Behavioral and neuroimaging evidence for a contribution of color and texture information to scene classiﬁcation in a patient with visual form agnosia. J. Cogn. Neurosci. 16(6), 955–965 (2004)

286

A. Y.-F. Lin et al.

8. Elliot, A.J., Maier, M.A.: Color psychology: effects of perceiving color on psychological functioning in humans. Annu. Rev. Psychol. 65, 95–120 (2014) 9. Oliva, A., Schyns, P.G.: Diagnostic colors mediate scene recognition. Cogn. Psychol. 41(2), 176–210 (2000) 10. Gorn, G.J., et al.: Waiting for the web: how screen color affects time perception. J. Mark. Res. 41(2), 215–225 (2004)

Design of an Integrative System for Conﬁgurable Exergames Targeting the Senior Population Teresa Paulino1,2, John Muñoz1,2(&), Sergi Bermudez1,2, and Mónica S. Cameirão1,2 1

Faculdade de Ciências Exatas e da Engenharia, Universidade da Madeira, Campus Universitário da Penteada, 9020-105 Funchal, Portugal {teresa.paulino,john.cardona,sergi.bermudez, monica.cameirao}@m-iti.org 2 Madeira Interactive Technologies Institute, Universidade da Madeira, Polo Cientíﬁco e Tecnológico da Madeira, floor -2, Caminho da Penteada, 9020-105 Funchal, Portugal

Abstract. Physical activity plays a key role in the prevention and delay of health problems associated with sedentarism in the senior population. Conﬁgurable exergames that can be adapted to ﬁt different needs have been proposed to address this issue because they are fun and enjoyable while promoting physical exercise. This paper describes the design process of an integrative system for Exergaming targeting the senior population that combines humancentred and software engineering methods. An initial requirements elicitation was carried out involving 12 health professionals. This resulted in a list of 110 design requirements, such as customization of games and activities, visualization of historical data, management of plans and sessions, and automatic recommendation of games. The software design was developed using the Uniﬁed Modelling Language (UML). Finally, we present two iterations with humancentred techniques, namely, card sorting and paper prototype evaluations, which will serve as basis for future implementations. Keywords: Human-centred design Integrative system engineering Human computer interaction Design for health professionals Exergaming system

1 Introduction Exercise videogames (Exergames) use physical activity as interface to foster enjoyment while working out [1]. Conﬁgurable bespoke exergames have been proposed as a solution for promoting physical activity among the senior population since its commercially-available counterparts raise many concerns, such as colorful and visually busy game interactions, unsuitable music, not being based on basic exercise principles, and demanding navigation through the user interface [2]. Exergames have been used in nursing homes and rehabilitation centers with similar or better effects compared to traditional forms of exercise [2]. However, conﬁguration of these systems can be © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 287–292, 2019. https://doi.org/10.1007/978-3-030-02053-8_44

288

T. Paulino et al.

difﬁcult and time consuming when faced with complex and a large amount of parameter choices. Previous research has focused on the design process and evaluation of the effects of exergames targeting the senior population [3, 4]. Despite its importance and signiﬁcant evidence of effectiveness, the success and acceptance of such systems is highly dependent on the adoption by health professionals and caregivers [5]. Due to the negative impact that bad interfaces might have on user experience while interacting with conﬁgurable exergaming systems, a carefully designed interface that follows human-centered design guidelines is essential for technology adoption [5]. In this paper we propose an integrative platform which can aggregate independent exergames that are controlled by a common user interface (UI). The system will be able to incorporate data from physical and cognitive assessments from users to provide decision support on the creation of training/rehabilitation plans adapted for each end-user proﬁle. The system will also be able to manage data that will be hosted both locally and in a cloud database to enable visualizing the historical progress of end-users and promoting ubiquitous access to data. Here, we describe the steps carried out to design the integrative system that involved the main prospective end-users of the conﬁgurable UI, namely health and sports professionals.

2 Procedure Our methodology is structured in stages using techniques commonly found in the ﬁelds of systems engineering, human computer interaction, and action research. The output of each stage served as input for the subsequent ones (Fig. 1). The following subsections describe the steps carried out along the system design process.

Fig. 1. Diagram showing the different design processes and participants involved.

2.1

Requirements Engineering

Literature supports that requirements engineering (RE) is a crucial step on the development cycle of a software system [6, 7], which helps to understand the user needs and therefore, assisting on designing a system towards its effectiveness.

Design of an Integrative System

289

2.1.1 Interviews with Health Professionals For requirements elicitation we have chosen a conversational method, under the form of semi-structured interviews. 12 participants were invited to collaborate in the initial process of requirements elicitation: 5 physiotherapists, 2 occupational therapists, and 5 physical training instructors, with average age of 28.9 ± 5.4, and 5.6 ± 4.7 years of professional experience. Interviews were carefully planned to cover most of the topics that could help understanding the needs of the professionals. Questions included categories such as: (i) knowledge and opinions about similar systems, (ii) positive and negative aspects about these type of interactive systems, (iii) daily practices in terms of rehabilitation therapies and exercise prescription, (iv) expectations of the integration of this system including data visualization, technical assistance, concerns, and constraints. Qualitative data was collected under the form of written notes and audio ﬁles recorded along the interviews. Raw data was coded using the qualitive data analysis tool MAXQDA1, resulting in 349 distinct codes with 994 records in all ﬁles. A code could correspond to a feature, a suggestion, a concrete answer, or even an insight that could be a suitable candidate for a requirement of the integrative system. 2.1.2 Requirements Prioritization Requirements prioritization assists in making the right decisions when choosing the most important features to implement on a system [8]. From the data obtained on the previous step, a data cleaning step was manually performed merging similar codes and removing the ones which corresponded to suggestions and opinions that were not considered requirements for the proposed system due to their high diversity. A ﬁnal list of 110 requirements was then shared with 6 exergaming researchers which evaluated each requirement with an eleven-point Likert scale considering two perspectives: (i) its cruciality for the system, meaning: “does this system work without this functionality?”; and (ii) its importance to fulﬁl the objectives of the system. The requirements were then ranked according to 3 results: the evaluation from the cruciality perspective (the mean of all participants), the percentage of interviewees that mentioned each speciﬁc requirement, and the evaluation from the importance perspective. The top ﬁve requirements are presented in (Table 1). Table 1. Top-ﬁve requirements obtained from the prioritization step. Requirement Audiovisual feedback Variety of activities/games/exercises Store historical data Save routines and plans Activities recommendation based on user proﬁle

1

Cruciality (mean) 8.3 8.2 7.5 7.0 6.8

Interviews (%) 69.44% 68.06% 62.50% 58.33% 56.94%

MAXQDA: Qualitative Data Analysis Software (https://www.maxqda.com/).

Importance (mean) 9.2 9.0 7.0 7.8 7.3

290

T. Paulino et al.

2.2

Software Design

We followed an Object-Oriented design approach using UML notation while applying universal principles of software engineering [9]. The software design process started by creating the use cases based on the requirements list. Functional requirements were adapted and rephrased to represent atomic actions which users can perform with the system. Three main actors were identiﬁed: (i) professionals - the person in charge of prescribing activities, (ii) end-users - seniors who will play the exergames, and (iii) administrator - someone with full access to the system, mainly technicians or researchers. Use cases were crucial to the next steps, which assisted in identifying the main entities and their relationships, enabling the creation of diagrams such as highlevel classes, activity, and architecture diagrams. User, session, plan, game, assessment, exercise, music, are some examples of entities identiﬁed along the software design process. 2.3

Human-Computer Interaction

In parallel with the software design process, a low-ﬁdelity paper prototype was built to assist in the design of the software diagrams. This helped in visualizing the possible interaction flows with the users. Health and sports professionals participated in the validation of the initial designs, allowing the identiﬁcation of possible preferences of interaction with the UI. For this, two traditional techniques were chosen and iterated: card sorting and paper prototype tests, which are described in the following subsections. 2.3.1 Card Sorting Card sorting is a technique that assists understanding how information should be organized in the UI. Open card sorting sessions avoid biasing the participants by enabling them to group and label their own categories [10]. Four open card sorting sessions were performed with 2 sports trainers and 2 occupational therapists, who also participated in the requirements elicitation process. 51 topics were chosen, which corresponded to functionalities of the system, or actions that the user can perform with the system (e.g. “Create session”, “Create training plan”, “Deﬁne frequency”, etc.). Each session was video recorded, and the participants were asked to think aloud while they were organizing the topics (Fig. 2).

Fig. 2. Participants organizing topics along the card sorting sessions.

Design of an Integrative System

291

All participants organized cards in a sequential manner. Some of the new topics that participants created have already been considered on the prototype, other topics were used as inputs for modiﬁcations of the current prototype. For instance, the possibility to edit a training plan, duplicate a session, or creating experimental sessions without being associated with any user, are functions that were not considered before. Despite each user have labeled groups in a distinct manner, we observed a pattern in organizing by: “User and assessments”, “Training plan”, “Sessions and games”, and “Settings and help”. 2.3.2 Paper Prototype Evaluation A paper prototype was built with sticky paper enabling interchanging elements during the prototyping and testing, where the features emerging from the card sorting sessions were considered. The low-ﬁdelity prototype consisted of a mix of hand-written text and drawings, with printed elements, mostly icons, to evaluate how intuitive they could be if applied in the ﬁnal UI. Two participants that participated in the previous design processes (one from each ﬁeld: sports and health), were invited and asked to accomplish two tasks and to think aloud while they were performing the actions. Each session lasted about 30 min. Participants managed to successfully accomplish both tasks faster than expected. Overall, the steps were intuitively performed without the need of much support. Some modiﬁcations emerged to integrate in the ﬁnal UI, such as: (i) recommendation of assessments after creating a new end-user, (ii) availability of an history of the settings previously chosen when creating or editing a plan, session, or activity, and (iii) automatic saving of data, without the need of having a save button.

3 Discussion and Future Work The purpose of this work was to design an integrative exergaming system involving the main interactors with the UI, the health professionals. From the requirements elicitation process, we found it sometimes hard to differentiate the exergames from the integrative system. Speciﬁcally, participants tended to always answer considering the effectiveness and concerns of the exergames. However, in the later steps such as card sorting or the paper prototype evaluations, the image of the system and its speciﬁc functionalities became much clearer. As an insight from the paper prototype evaluations, we identiﬁed that there were too many options present on the system. A better approach would have been to start with a very simple interface with the basic functionalities and iteratively increment it along with the needs of the professionals. By involving the health professionals in the design process, we acknowledge that participants felt empowered by taking part of the various stages, and that they understood the importance and impact of their decisions on the envisioned system. As future steps, after implementing the digital interface with basic functionalities and navigation, we will run a usability test, which will serve as a basis for future implementations towards the deployment of a functional integrative system.

292

T. Paulino et al.

4 Conclusion In this paper we presented the steps carried out to design an integrative system for Exergaming targeting the senior population combining human-centered and software engineer methods. We took advantage of techniques such as semi-structured interviews, card sorting, and paper prototyping to involve the main interactors of the envisioned system in the design process which allowed to design with the consideration of their preferences and needs. We believe that the performed steps are fundamental for the future adoption of the proposed system. Acknowledgments. The authors would like to thank the sports and health professionals from Gymnasium Santo António – Funchal, Campus Neurológico Sénior – Lisboa, and NeuroRehabLab, that participated in the design process. This work was supported by the Portuguese Foundation for Science and Technology through the Augmented Human Assistance project (CMUPERI/HCI/0046/2013), Projeto Estratégico UID/EEA/50009/2013, and ARDITI (Agência Regional para o Desenvolvimento da Investigação, Tecnologia e Inovação).

References 1. Laamarti, F., Eid, M., Saddik, A.E.: An overview of serious games. Int. J. Comput. Games Technol. 2014, 15 (2014). Article ID 358152 2. Skjæret, N., Nawaz, A., Morat, T., Schoene, D., Helbostad, J.L., Vereijken, B.: Exercise and rehabilitation delivered through exergames in older adults: an integrative review of technologies, safety and efﬁcacy. Int. J. Med. Inf. 85, 1–16 (2016) 3. Konstantinidis, E.I., Billis, A.S., Mouzakidis, C.A., Zilidou, V.I., Antoniou, P.E., Bamidis, P.D.: Design, implementation, and wide pilot deployment of FitForAll: an easy to use exergaming platform improving physical ﬁtness and life quality of senior citizens. IEEE J. Biomed. Health Inform. 20, 189–200 (2016) 4. Velazquez, A., Martinez-Garcia, A.I., Favela, J., Hernandez, A., Ochoa, S.F.: Design of exergames with the collaborative participation of older adults. In: Proceedings of the 2013 IEEE 17th International Conference on Computer Supported Cooperative Work in Design (CSCWD), pp. 521–526 (2013) 5. Johnson, C.M., Johnson, T.R., Zhang, J.: A user-centered framework for redesigning health care interfaces. J. Biomed. Inform. 38, 75–87 (2005) 6. Zhang, Z.: Effective requirements development – a comparison of requirements elicitation techniques. In: Software Quality Management XV: Software Quality in the Knowledge Society, pp. 225–240. British Computer Society (2007) 7. Ambreen, T., Ikram, N., Usman, M., Niazi, M.: Empirical research in requirements engineering: trends and opportunities. Requir. Eng. 23, 63–95 (2018) 8. Berander, P., Andrews, A.: Requirements prioritization. In: Engineering and Managing Software Requirements, pp. 69–94. Springer, Heidelberg (2005) 9. Bruegge, B., Dutoit, A.H.: Object-Oriented Software Engineering Using UML, Patterns, and Java. Prentice Hall Press, Upper Saddle River (2009) 10. Wood, J.R., Wood, L.E.: Card sorting: current practices and beyond. J. Usability Stud. 4, 1–6 (2008)

Research Based on Product Design: The Example of Spimi Skin Detector Cao Ying(&) and Cao Jing Huazhong University of Science and Technology, Wuhan, China [email protected], [email protected]

Abstract. In many developing countries, the public understanding of physical condition is increasing, and there are already a large number of users who regularly perform physical examinations to observe health conditions. At the same time, in medical treatment, Chinese medicine is often questioned because of its subjective nature. This requires an objective, intuitive and scientiﬁc means to guide and corroborate Chinese medicine. According to the features of the user model and the principles of design aesthetics, the comprehensive principles of the appearance of medical product design are analyzed and summarized in this paper. Keywords: Service design

Cognitive psychology Medical equipment

1 Introduction This article takes the design process of the SPIMI non-intuitive imaging equipment product series jointly researched and developed by the Cosmos team and Hubei Instrument Chang electronics Co., Ltd. as an example. It discusses the teaching practice with the actual combat of the project, and intends to use design knowledge to instruct the project. The results of the practice will provide more convenient and precise facial health and beauty services to more individuals.

2 Product Research According to the team’s research statistics, 32% of the individuals in daily life do not understand their skin conditions, and 68% of users use makeup and maintenance products that are inconsistent with their skin all the year round, which often leads to various types of skin allergy symptoms or Skin diseases. In the design, the team attached great importance to the user-centered design principles, analyzed the user’s thinking and operation process in cognitive psychology, analyzed the characteristics of the user model based on the analysis, and analyzed and summed up medical theories based on the principles of design aesthetics. Finally, we draw a conclusion about the principle of appearance of the design of the United States product.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 293–298, 2019. https://doi.org/10.1007/978-3-030-02053-8_45

294

2.1

C. Ying and C. Jing

User Surveys

With regard to the market demand for skin testers, the project team conducted a questionnaire based on the research methods which were taught in product design course. A total of 16 questions was included in the questionnaire, and the number of participants was 170. The main content of the survey is as follows (Tables 1, 2, 3, 4 and 5 for details):

Table 1. User’s satisfaction with facial skin Satisfaction Satisﬁed Acceptable Not Satisﬁed Terrible Level Proportion 17.65% 31.18% 47.06% 4.12% Whether the user understands or has performed facial skin detection Detected Detected Not tested Proportion 24.56% 75.44% How satisﬁed users are with the facial products they purchase Satisfaction Determine the application Should Not applicable Unclear Level apply Proportion 17.06% 50% 3.53% 29.41% User’s conﬁdence in the state of the body’s response to facial skin Trust Very Very much Believe a bit Do not conﬁdent believe believe Proportion 21.18% 51.76% 20% 7.06% User acceptance of facial skin tester ① Will skin tests be performed regularly? Degree Often Sometimes Occasionally Will not Proportion 28% 32% 18% 22% ② Will you accept the maintenance plan given by the tester? Degree Believe Doubt Do not Reservations believe Proportion 36% 41% 18% 5%

In accordance with the results of the survey, it can be seen that economically independent women aged 25–45 years are more valued for skin care, and those who have not done skin tests have a certain interest in this product. Most consumers pay more attention to the following points (Table 2): 2.2

Research on Competitive Status

In the analysis of competition appearance and color matching, through the comparative analysis of the extraction of different outlines and colors schemes in Table 4, it can be seen that the products in the market are dominated by smooth, arc-shaped lines, giving

Research Based on Product Design

295

Table 2. Safety Accuracy Privacy Unique Timeliness Feature Economical

No harm to the body Test results meet scientiﬁc standards The conﬁdentiality of the assessment results Give expert assessments Periodic review can see signiﬁcant differences Can give speciﬁc advice Within the affordable price range

Table 3. Competitor Analysis Product name Produced Main functions and features

Market status

VISIA Skin Tester

CBS Skin Tester

United States It is the instrument that can quantitatively analyze the pathological features of the skin. And there are two international authoritative skin databases from Proter & Gamble, which support a large number of clinical data Industry Standard

Taiwan Spectral-color-level facial scanning can not only detect the troubles that have been exposed on the skin surface, but also can visually display the troubles hidden in the basal layer of the skin through quantitative analysis Cheap, poor function, low-end

Table 4.

OBSERV520 Skin Tester Netherlands These changes can be measured and monitored over time to provide precise analysis and reveal ignore other skin troubles

The product features are incomplete and the market is not open

296

C. Ying and C. Jing

individuals a moderate, safety and health feel to meet the emotional needs of users. The combination of black and white is simple and stable. There is no strong color inclination, color harmony, full sense of science and technology, and it meets the rigorous and professional user’s experience of medical products.

3 Product Positioning and Design 3.1

Enterprise CIS Systems Design

The product name inherited from the company’s logo is SPIMI. To this end, it has designed a system of marking standards for marking. This drawing is a marking speciﬁcation and measures the slope and arc formed by the turning of logos (Fig. 1).

Fig. 1.

3.2

Product Design

The design of SPIMI face detector product design is based on the previous research and data analysis. The features of the desktop participating in the operation type are simple structure, wide vision, and good lighting, and various types of items for operation and use can be placed on the desktop. Such products are used by two individuals, ﬁrstly one person to operate, and test with photographing, analysis, imaging and generation, otherwise one person is tested, and mainly used for the detection of medical cosmetology. Vertical products are used by one person, and the user obtains the qualiﬁcation through the method of online payment, and self-photographing, data generation process, and gets the test results. According to the structure of the optical components and the position of the workbench, a sketch model is established in combination with the application of ergonomics, as shown in Fig. 2. After comprehensive evaluation and cooperation with the partners, after selecting option three, and color scheme was derived, as shown in Fig. 3:

Research Based on Product Design

Fig. 2.

Fig. 3.

297

298

3.3

C. Ying and C. Jing

The Overall Principle of Appearance of Medical Beauty Product Design

After the design practice of the SPIMI face detector, the teaching team summarized several principles similar to the appearance design of medical aesthetic products: Compliance and Follow Ergonomics. From the “machine” factor, the ergonomic design of medical equipment needs to design the interaction interface from the operator and the operator’s dimensions; and from the “human” factor, medical device design. Take full account of user needs and emotions. The emotional design of medical beauty products must begin with user needs and value analysis, including the user’s psychological characteristics and the audience’s psychological activities. Improve User Experience and Data Support. Simulate the user experience process and mine user behavior and psychological needs. It helps designers to design medical beauty products that are more in line with the user’s behavior and thinking patterns.

4 Summary and Prospect of the Project First of all, through detailed investigation of the market to obtain project data, to form favorable understanding of market trends, to get precise control of the marketing rhythm; subsequently through a comparative analysis of a variety of competing products, to determine the innovative creativity lies in the use of comfortable the promotion of degree and its universal promotion, thus establishing the subdivision of operation methods, color materials and functional positioning of the project products, focusing on the desktop’s participation in operational product forms; furthermore, the team designed a package for the promotion of product promotion services. The product’s visual identity system, from the visual to the cultural, integrates the design of the product, and ﬁnally obtains the integrated principles of the initial appearance of the product form and the appearance design.

References 1. Cao, G., Wang, H., Zhang, S.: Hierarchical biometrics driven product emotional design method. Packaging engineering (2017) 2. Gou, R., Wang, J., Xu, L., Leo, S.M.: Research on product design integrated innovation procedures and methods. Packaging engineering (2017) 3. Zhai, Y.R.: Home beauty instrument product design research. orange, design art and theory (2017) 4. Li, Y.: Design and research of wearable intelligent blood oxygen movement loop. South China University of Technology (2017) 5. Wen, J.: Application Research of product design based on health concept (2017)

Development of a Prototype for Non-contact Keyboard Yasushi Kambayashi1(&), Keita Ueda1, Masanari Kasahara1, Tatsumi Kusano1, and Munehiro Takimoto2 1

2

Department of Computer and Information Engineering, Nippon Institute of Technology, 4-1 Gakuendai, Miyashiro-Machi Minamisaitama-Gun, Saitama, Japan [email protected], {c1145123,c1145154, c1145210}@cstu.nit.ac.jp Department of Information Sciences, Tokyo University of Science, 2641 Yamazaki, Noda 278-8510, Japan [email protected]

Abstract. There are certain needs for non-contact input devices such as noncontact keyboard. Medical people such as doctors and nurses cannot use keyboards in the operation room because of the hygienic problem. Even ordinary people cannot use keyboards when they use head mounted display for full VR environment because they cannot see the keyboards. In order to solve such a problem, we propose employing a motion sensor device. A motion sensor device is a device for sensing motions. It detects human gestures such as handshakes and ﬁngertip movements. Recently LeapMotion is gaining popularity as a non-contact input means, but we can observe few implementations of keyboard using LeapMotion. In this paper, we report our experiences of developing a non-contact keyboard system using LeapMotion. Keywords: Human factors Human-systems integration Human computer interaction

User interface

1 Introduction Today, people use a physical keyboard or a touch panel keyboard as a primary input device for computers. Medical people such as doctors and nurses, however, cannot use keyboards in the operation room because of the hygienic problem. Even ordinary people cannot use keyboards when they use head mounted display for full virtual reality (VR) environment because they cannot see the keyboards. In order to solve such a problem, we propose employing motion sensor devices. A motion sensor device is a device for sensing motions. It detects human gestures such as handshakes and ﬁngertip movements. Then what motion sensor device we can employ to implement a noncontact keyboard? We chose a small motion sensor device called LeapMotion for this purpose. LeapMotion is a small USB based motion sensor device designed to capture movement of the ﬁngertips. It is relatively inexpensive and small enough to ﬁt various environments. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 299–304, 2019. https://doi.org/10.1007/978-3-030-02053-8_46

300

Y. Kambayashi et al.

Recently LeapMotion is gaining popularity as a non-contact input device, but we can observe few implementations of keyboard using LeapMotion. In this paper, we report our experiences of developing a keyboard system using LeapMotion. We place LeapMotion in front of the computer when we use it so that we can focus on the position of the motion only at the ﬁngertip to capture the keyboard typing. This method, however, has a big problem. Non-contact type keyboard using LeapMotion is not very accurate. Therefore it is important to increase the accuracy of position recognitions at the virtual keyboard heads in the three dimensional space. In order to overcome the accuracy problem, we have employed Windows Presentation Foundation (WPF) to separate the user interface from the logic of position recognition, and made the recognition algorithm to iterate the comparisons of coordinates. Our non-contact keyboard system provides the users eighty to ninety percent recognition rate average for inexperienced users. The previous researches that use LeapMotion to implement keyboards have required the users the same home positions as they use a traditional physical keyboard when using virtual keyboard, i.e. palms facing down and lifting arms [1, 2]. Actually, this position produces fatigue for the users. Therefore, we set our system’s home position vertical where the palms face parallel to the display screen. The experiments show this home position reduces the users’ fatigue because they can lower their elbows. Thus, we have achieved to reduce the accumulation of fatigue in the arms using our non-contact keyboard system compared to using the conventional non-contact keyboard systems. On the other hand, we have found that our design requires the user larger amount of arm movements compared to the physical keyboard. We have to devise a new design that further reduces users’ fatigue as well as increases the accuracy.

2 Background LeapMotion is a small device that senses the motion of the ﬁngers of the user in the recognition scope. The recognition scope is a three-dimensional reverse pyramidal shape as shown in Fig. 1. In the scope, LeapMotion recognizes four gestures as well as the coordinates of ten ﬁngers. The four gestures are circle, swipe, key-tap and screentap. In this paper, we concentrate key-tap. The LeapMotion SDK deﬁnes a class namely InteractionBox that is a virtual rectangular parallelepiped shown in Fig. 1 [3]. From the viewpoint of the user, x-axis deﬁnes the width, y-axis deﬁnes the height and z-axis deﬁnes the depth in the InteractionBox. Through the class, application programs can detect the coordinates of each ﬁnger as well as the gestures of the ﬁngertips. We focus on the touch emulation. In order to detect typing, LeapMotion deﬁnes a virtual plane as x-y-plane, and implement touch emulation. Touch zone consists of hovering area and touch plane. When a ﬁnger enters to the hovering area of its touch zone, a LeapMotion API in class Leap returns plus one. When the return value is in-between zero through minus one, we deﬁne detecting touching. In order to provide the user feedback, we employ Windows Presentation Foundation (WPF) to display a keyboard on the screen so that we can separate the user interface from the logic.

Development of a Prototype for Non-contact Keyboard

301

Fig. 1. InteractionBox in the scope

3 The System We have implemented the proposed system on Windows 10 using Windows Presentation Foundation (WPF). WPF is a user interface subsystem included in .NET Framework 3.0. The system depict a keyboard layout and input space on WPF. Since LeapMotion tracks the coordinates of the users’ ﬁngertips, the system translates the coordinate data into the coordinates on WPF through InteractionBox, and then draws the positions of ﬁngertips on WPF. We have implemented a virtual keyboard on WPF using C# language with LeapMotion SDK [3]. When startup, the system draws a keyboard on WPF, and then the system also draws the sensed positions of the user’s ﬁngertips also on WPF as shown in Fig. 2. Since we set the virtual keyboard parallel to the display screen, we gave up recognizing the thumbs. The user does not use thumb to type. In order to determine the coordinates on WPF, the system utilizes InteractionBox to translate the coordinates of the ﬁngertips tracked by LeapMotion into the coordinates on WPF. Even though LeapMotion SDK has API KeyTapGesture and ScreenTapGesture in order to recognize the gestures of ﬁngers, their recognition rates are not high enough for our application. Therefore, we have made the system to compare the coordinates of multiple frames to obtain the precise recognitions of the gestures of the user’s ﬁngertips. When the system recognizes the positions of the ﬁngertips, it gives a feedback to the user by changing the color of the keys.

Fig. 2. The keyboard

302

Y. Kambayashi et al.

Figure 3 shows the traces of y-coordinates (i.e. heights) of right hand ﬁngers, when a user were performing typing gestures. In the experiments, the user has pushed keys ﬁve times with the index ﬁnger, four times with the middle ﬁnger, two times with the ring ﬁnger, and two times with the little ﬁnger. As shown in the ﬁgure (right hand side), the ring ﬁnger and the little ﬁnger accidentally follow the action of the middle ﬁnger and the ring ﬁnger respectively. We have found that misrecognitions occur when hands are too close to the sensor, and when the roots of ﬁngers go beyond the x-y plane. Especially we have observed that the user moves his ﬁngers parallel to the z-axis (i.e. toward the display screen). In order to mitigate the misrecognitions, we have changed the typing gestures. In addition, we have implemented three steps screening to recognize valid gestures while preventing misrecognitions.

index

middle

ring

little

Fig. 3. The characteristics of y-coordinate of each ﬁnger

The ﬁrst screening is to compare the amount of movement of the target ﬁnger with the average amount of those of other ﬁngers, and to ﬁgure out whether that is a conscious movement or accidental movement. The system measures the average differences of x, y, and z coordinates in 0.4 s and compare the maximum movement and other. If the difference is smaller than a predeﬁned threshold, the system judges that movement is an accidental one and ignores it. If the difference is bigger than a predeﬁned threshold, the system proceeds the second screening. The second screening is to compare the moving distance of y-coordinate of the target ﬁnger and those of the other ﬁngers. If the difference is more than the predeﬁned threshold, we estimate the ﬁnger performs typing. We may misrecognize, however, multiple key strokes for one key typing. We need to proceed the third screening. The third screening is to observe the lifting movement of the target ﬁnger. When a user types, he or she must lift the ﬁnger right after push the key. When these two actions occur continuously, the value of y-coordinate ﬁrst decreases and then increases. Then the system checks the real-time frame 0 through 3, and determines the continuous actions mean a key typing. Any actions pass these three stages of screening; the system appreciates that is a typing and accepts the character input.

Development of a Prototype for Non-contact Keyboard

303

Even though these screenings works well, the rapidity of frame update makes the system misrecognize multiple input characters for just one key stroke. In order to mitigate this situation we make the system not accept input for 0.5 s after one key input.

4 Experiments In order to demonstrate the feasibility of our system, we have employed ﬁve subjects to measure the recognition accuracy of the system. Even though the average recognition rate is 84.4%, many subjects have failed to input “b” and “c”; both are on the bottom line of the virtual keyboard (see Fig. 4). We have found that LeapMotion tend to misrecognize the gestures when the ﬁnger close to the sensor. We may need to reconsider the position placing the sensor. We have also found all the subjects start typing with eight ﬁngers (i.e. excluded the thumbs) as they type conventional keyboard, but after certain period, they start to use only index ﬁngers and middle ﬁngers. We conjecture that they had hard time to type using ring and little ﬁngers. We may need to adjust keystroke for evaluating typing gestures for each ﬁnger.

Fig. 4. Virtual keyboard in use

During the experiments, all the subjects had claimed that this experiment made them tired. Actually, our original design was the same as previous researches, i.e. the same home positions as they use a traditional physical keyboard and palms facing down and lifting arms. In order to ameliorate the fatigue problem, we set our system’s home position vertical where the palms face parallel to the virtual screen. The experiments show this home position reduces the users’ fatigue because they can lower their elbows. Thus, we have achieved the accuracy through the reduction of fatigue accumulation on the arm compared to the conventional system. On the other hand, we have found that our design requires the user larger amount of arm movements compared to the physical keyboard, and this leads to misrecognition and reduces the recognition

304

Y. Kambayashi et al.

rate. We have to devise a new design that further reduces users’ fatigue as well as increases accuracy. We have also observed that the tracking accuracy of LeapMotion is depending on the position of the hands and arms. One particular subject has produced much false detection because of his particular arm positioning. We expect the further study of the arm positioning provides the improvement in accuracy.

5 Conclusion and Future Directions We have designed and implemented a non-contact virtual keyboard so that users can input characters without touching any physical devices. This non-contact feature is especially useful in medical environments as well as VR environments. We chose LeapMotion as a base system and implemented the virtual keyboard with WPF on Windows 10. Even though LeapMotion SDK has API KeyTapGesture and ScreenTapGesture in order to recognize the gestures of ﬁngers, their recognitions are not very accurate and sometime provide misrecognitions. Therefore, we have installed three steps screening to exclude ambiguous similar-to-keystroke-but-not-keystroke gestures. Previous non-contact keyboards have required the users to make their palms face down, but we have found the positioning makes the users tired. Therefore, we have made our system’s home position vertical and make their palms face parallel to the display screen. This design reduces the users’ fatigue, but does not contribute accuracy. We must pursue the further study about the arm and palm positioning in order to obtain more accurate with less fatigue recognitions. We had implemented virtual multi-mouse input system using Kinect [4]. The purpose of the system was to encourage brainstorming, reporting and discussing situations in a shelter for natural disasters. The multiple pointing devices are useful to indicate dangerous areas on a projected large map. The system, however, lacks to put short description what had happen at the pointed place. We are planning to combine the current non-contact keyboard system and the previously built non-contact multi-mouse system into one practical system so that multiple users can freely discuss and write ideas on a virtual whiteboard.

References 1. Hoshino, G.: Development of a Virtual Keyboard AirKey Using Leapmotioncontroller. Bachelor’s Thesis, Kyushu Institute of Technology, In Japanese (2014) 2. Yi, X., Yu, C., Zhang, M., Gao, S., Sun, K., Shi, Y.: ATK: enabling ten-ﬁnger freehand typing in air based on 3D hand tracking data. In: 28th Annual ACM Symposium on User Interface Software and Technology, New York, pp. 539–548. ACM (2015) 3. LeapMotion Inc.: LeapMotion C# SDK v3.2 Documentation. https://developer.LeapMotion. com/documentation/csharp/index.html. Accessed 2017 4. Onodera, Y., Kambayashi, Y.: Vision-based user interface for mouse and multi-mouse. In: Yoshida, T., Kou, G., Skowron, A., Cao, J., Hacid, H., Zhong, N. (eds.) 9th International Conference on Active Media Technology, LNCS, vol. 8210, pp. 14–23. Springer, Heidelberg (2014)

IHSED 1: Virtual Reality and Usability Evaluation Applications

Augmented Reality in the Context of Naval Operations Mário Marques1, Filipe Elvas1, Isabel L. Nunes2,3, Victor Lobo4, and Anacleto Correia1(&) CINAV – Portuguese Navy, Alfeite, 2810-001 Almada, Portugal {simoes.marques,mendes.elvas, cortez.correia}@marinha.pt 2 Faculty of Sciences and Technology, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal [email protected] 3 UNIDEMI, Department of Mechanical and Industrial Engineering, Faculty of Sciences and Technology, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal NOVA IMS, Universidade Nova de Lisboa, 1070-312 Lisbon, Portugal [email protected] 1

4

Abstract. As technology evolves, the feasibility of implementing Augmented Reality applications on mobile systems has increased signiﬁcantly opening opportunities in many usage domains. For instance, in the ﬁeld of education and training, systems AR provides an easy and safe mode of acquiring knowledge and gaining skills and experience in complex contexts and environments. This paper describes two AR applications built for support of naval operations. The Tactical-AR was intended for training purposes, to demonstrate the maneuvers of ships in formations, according to tactical publications. The THEMIS-AR presents the use of AR in the context of disaster relief operations, where responders receive information on the incidents, tasks to perform, navigation guidance to reach the incidents and advice on procedures to use while performing the tasks. Keywords: Human factors Augmented reality Mobile devices Human-computer interaction User experience Naval operations Learning experiences

1 Introduction Augmented Reality (AR) is a set of technologies that can be applied to extend different modalities of human senses (e.g. sight, sound and/or touch). In the context of this work we address Augmented Reality (AR) in a stricto sensu, i.e. considering only the sight sense. Therefore, here AR is considered as a mode of interaction that dynamically combines real images with virtual objects and information, enriching the visualization context in a way that is intuitive and transparent to the user. This emerging technology has a natural ﬁeld of application in military training and education, considering the complexity, risks, and costs inherent to this domain. This is © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 307–313, 2019. https://doi.org/10.1007/978-3-030-02053-8_47

308

M. Marques et al.

particularly true in the ﬁeld of naval activities and missions, a context where the use of simulators is common, but where there are scarce experience and knowledge on the use of this mode of interaction. AR capable devices (e.g., tablets, smartphones, glasses) are key in the creation of new navy training experiences, which in many cases is still very dependent on conventional media supports (e.g., paper-based) that, although capable of providing detailed information, are not able to give users a contextualized view of the different tactical elements in the surrounding environment. The goal of this paper is to present and discuss current research on AR applications which explore new ways to interact and view content associated with the context of naval operations using mobile devices. The next section, presents two experiences on using augmented reality in the context of naval operations. The last section offers some conclusions and envisions future work.

2 Augmented Reality Case Studies This section describes the cycle of development of two AR applications built to support and replace paper-based activities taken place at the Portuguese navy. 2.1

Tactical-AR

The Tactical-AR application was built to support naval operations training, a complement of traditional materials provided to the students, mostly paper-based. The AR application development adopted Scrum, an agile framework used by developers to address complex adaptive problems, in a productive and creative fashion, in order to deliver valuable products [1]. As recommend by the framework, the Scrum team was supported in the development process by two members with speciﬁc roles: (1) the ScrumMaster, acting as a team’s coach, helping them perform adequately the Scrum process; and (2) the product owner, a representative of the end-users, and the team guide toward the most adequate product. The product owner was the manager of the product backlog, an ordered list of the work to be done in order to create, maintain and sustain the AR application. The Tactical-AR was developed through a series of iterations (aka sprints), following the process depicted in Fig. 1. To keep the development process agile, the daily sprints were timeboxed to two weeks. Each sprint consisted in planning and development of a small set of features of the Tactical-AR. Thus, each sprint consisted of the ideation of the AR functionality, their coding and testing. Once the process was ﬁnished, the new features were integrated into the Tactical-AR application. On the beginning of each sprint session, all team members attended a Scrum meeting, which included the ScrumMaster and the product owner. During this meeting (with a planned duration of 15 min), the team members shared their previous session work, described the work planned for the new session, and identiﬁed any impediments to progress. The discussion on the sprint activity was an effective way of synchronizing the work among different teams. At the end of the sprint session, each team conducted a sprint review during which demonstrated the new functionality to the product owner and end-user’s

Augmented Reality in the Context of Naval Operations

309

representatives allowing them to influence the next sprint, and looking for their validation. This feedback loop, within Tactical-AR development process, often resulted in the identiﬁcation of opportunities to improve the recently developed functionality, and was also an opportunity to revise the requirements listed in the product backlog or to add new ones. Thus, the sprint retrospective meeting organized after each sprint session, is an important activity considered in the Scrum framework, since it involves all team members, the ScrumMaster and the product owner, representing an occasion to discuss the issues faced on the last sprint and to identify new ways of improving the Tactical-AR.

Fig. 1. The Scrum framework [2].

In order to build the product backlog of the Tactical-AR application a requirements elicitation was conducted, supported by two distinct techniques: a brainstorming session [3] and an online questionnaire. The brainstorming was conducted in the following 5 steps: (1) deﬁnition of the problem to be studied and object of the Tactical-AR. The stated problem was how to obtain the best graphical interface to present the content in Tactical-AR in the simplest and most practical way of deﬁning the naval environment and inserting the ships on it; (2) a meeting was conducted with trainees to discuss the problem. All members of the group, due to their academic qualiﬁcations, were aware of the facts and could therefore provide valid opinions; (3) a debate was promoted with simple rules: overlap in participants’ interventions, and critics to ideas of the other participants were not allowed; (4) from this discussion resulted a consensus for a practical and easy to use interface, with natural marks. The prevailed idea was that the trigger for the augment 3D virtual ships and their movements in formation would be a text captured and recognized by the camera. With this approach, it would be possible for a trainee when reading the deﬁnition in a book, to trigger with her hand-held display the AR material in order to better understand it through the illustration with the multimedia object; (5) eventually the brainstorming was closed after a synthesis of the main ideas were endorsed to the project owner to be part of the product backlog. From the online questionnaire another set of requirements were gathered for the Tactical-AR, regarding the following items: number of types of ships’ formation made available; number of ships in each kind of formation; perspectives (angles) to made available for watching the formation and ships’ evolution; availability of different

310

M. Marques et al.

meteorological conditions when watching ships’ formation; way to distinguish the guide ship on the formation; initial formation before the beginning of the maneuvers to attain the ﬁnal formation; type of mark to use to trigger the movements; types of AR devices in which the Tactical-AR is to be made available. Figure 2 shows a sequence of images of the Tactical-AR through the camera of a hand-held device, (a) recognizing a mark with the formation name, (b) the moment when the 3D image is triggered, and (c) with 3D image showing the ships’ formation in a loop.

(a)

(b)

(c)

Fig. 2. Recognition of a mark by the Tactical-AR.

After a usability test conducted with 40 students concerning the efﬁcacy of Tactical-AR, conveying the concepts of tactical maneuvers, the responses from 1 (bad) to 6 (excellent) were summarized in Fig. 3. A hypothesis test allowed to conclude, with 5% signiﬁcance, for not rejecting the alternative hypothesis of the TacticalAR being suitable for training of tactical maneuvers.

Fig. 3. Suitability of Tactical-AR for tactical maneuvers training (1-bad; 6-excellent).

2.2

THEMIS-AR

Disaster relief operations (DRO) are an increasingly common operational scenario for navies, exploiting the beneﬁts of sea-based capabilities (mobility, flexibility and logistic capacity) for the projection of international humanitarian assistance to large disasters in coastal area (for instance, as a consequence of earthquakes, tsunamis or hurricanes). THEMIS-AR is an app designed for mobile devices that was developed to assist ﬁrst responders engaged on DRO. The app was developed under the THEMIS (disTributed Emergency Management Intelligent System) project, whose aim is designing

Augmented Reality in the Context of Naval Operations

311

and implementing an intelligent system to assist Disaster Management activities in a very challenging context, considering the physical, cognitive and emotional dimensions [4]. The entire system is composed by desktop stations (used by disaster managers in Command Posts) and by mobile devices (used by responders operating in the affected area). The system supports training activities, in the preparedness phase (prior to a disaster), and real time disaster management in DRO, i.e. during the response phase (after the onset of a disaster). The georeferenced information compiled from multiple sources is used to generate a common picture which ensures decision-makers an adequate situational awareness. The apps installed in mobile devices support information sharing (e.g., location and characteristics of incidents) among responders and guides them regarding the tasks to perform and procedures to apply. THEMIS-AR is one of these apps, which includes AR functionality designed to leverage responders’ perception regarding the operational context. This is achieved using a layer of digital information (e.g., text, icons, images) which overlaps the image acquired by the mobile device camera, providing additional context to the raw video (for instance, navigation or procedural guidance), as depicted in Fig. 4.

Fig. 4. Example of the THEMIS-AR API.

THEMIS-AR architecture and features were ﬁrst presented in [5] which discussed the preliminary research and implementation work, and later in [6] which focused on the assessment of the prototype usability. As explained in [5] the app was designed for Android devices and implemented in Java and JavaScript. The AR functionality used a database server supported by SQLite [10] and an Wikitude API [9]. The initial prototype is currently being improved, and meanwhile has been used to support THEMIS project tests. The usability testing, discussed in [6], involved ﬁeld tests, with subject experienced and non-experienced in emergency activities, using User Experience Questionnaire

312

M. Marques et al.

(UEQ) [7] and System Usability Scale (SUS) questionnaire [8]. The results of these tests were used to identify areas requiring further improvement. Despite the promising results achieved with initial prototype, which was mainly focused on demonstrating the feasibility of implementing an app to support DRO, there is still a signiﬁcant work to be done regarding the user interaction design. The current work is taking in careful consideration the particular demands and challenges imposed by the responders’ operational environment and the quite challenging requirements deﬁned for the app, which include the creation, edition, and visualization of georeferenced information (e.g. incidents, points of interest, other response teams), exchange of instructions and reports, and access to advice regarding task priority or support regarding task procedures.

3 Conclusions This paper describes two AR applications build for supporting different types of naval operations. The ﬁrst app (Tactical-AR) addresses training activities related with naval operations. The aim of this AR application was to provide virtual representations of ship convoys, allowing users to manipulate them in 3D, see transitions between ships’ dispositions or access relevant multimedia information about the tactical scenario presented. The application raises awareness of less known formations or maneuvers and advises the trainees about transitions that could be made from speciﬁc ships’ arrangement. The second example addresses the THEMIS-AR prototype, which is a proof of concept app that offers to DRO response teams AR-based functionalities through the use of mobile devices. The initial results are very encouraging from a User Experience perspective and allow to conclude that the concept is valid. Building on this experience the THEMIS-AR is undergoing a second spiral of development, incorporating significant improvements in terms of API design and app contents. The AR applications, described in this paper, were designed and tested according to User Experience principles, in order to evaluate their usability and assess users’ learning experience. The initial results from the assessment of these applications proved the feasibility of the concepts experimented and reveal that this new interaction approach increases the motivation on military training and also the situational understanding during the execution of naval operations. The experience acquired by the authors in the design of these apps will be a driver to the development of new tactical-oriented AR applications. Acknowledgments. The work was funded by the Portuguese Ministry of Defense and by the Portuguese Navy.

Augmented Reality in the Context of Naval Operations

313

References 1. Schwaber, K.: Agile Project Management with Scrum. Microsoft Press, San Francisco (2004) 2. Scrum: Scrum (2018). https://www.scrum.org/ 3. Gregersen, H.: Better Brainstorming (2018). https://hbr.org/2018/03/better-brainstorming 4. Simões-Marques, M., et al.: Empirical studies in user experience of an emergency management system. In: Proceedings of Advances in Human Factors and System Interactions, AHFE 2017 Conference on Human Factors and System Interactions, Advances in Intelligent Systems and Computing, 17–21 July 2017, pp. 97–108. Springer (2018) 5. Nunes, I.L., et al.: Augmented reality in support of disaster response. In: Proceedings of Advances in Human Factors and System Interactions, AHFE 2017 Conference on Human Factors and System Interactions, Advances in Intelligent Systems and Computing, 17–21 July, pp. 155–167. Springer (2018) 6. Nunes, I.L., et al.: An augmented reality application to support deployed emergency teams. In: Proceedings of the 20th Congress of the International Ergonomics Association (IEA 2018), Advances in Intelligent Systems and Computing, 26–30 August, vol. 822, pp. 195– 204. Springer (2019) 7. UEQ: User Experience Questionnaire (2018). http://www.ueq-online.org/ 8. SUS: Measuring Usability with the System Usability Scale (SUS) (2018). https:// measuringu.com/sus/ 9. Wikitude: Wikitude (2018). https://www.wikitude.com 10. SQLite: SQLite (2018). https://www.sqlite.org

Strategies and Metrics for Evaluating the Quality of Experience in Virtual Reality Applications Xiangjie Kong(&) and Yuqing Liu National Key Laboratory of Human Factors Engineering, China Astronaut Research and Training Center, Beijing, China [email protected]

Abstract. The development of virtual reality makes it possible to apply the technology in many ﬁelds, but the experience of users received limited attention. Quality of experience (QoE) is an evaluation metrics can be used to evaluate applications and services from the perspective of users. By an intensive analysis of the literature about user experience models and QoE related projects, three strategies for QoE evaluation were put forward: emphasizing user needs, emphasizing system characteristics and emphasizing emotional experience. As an example, a set of metrics for the evaluation of astronaut virtual training system (AVTS) are designed under the guidance of the strategies, which consists of ﬁve ﬁrst-level indicators and ten second-level indicators and can be used to quantify the quality of experience of the astronaut virtual training system. Keywords: Quality of experience Virtual reality

Evaluation strategy Evaluation metrics

1 Introduction Virtual reality (VR) is a promising technology that is valued by the market and industry and growing rapidly in the past few years. Now, the technology has been applied in many ﬁelds such as education, surgery training, military simulation, e-games and entertainment [1]. Although virtual reality has its unique advantages in many ﬁelds, it is still a new industry and is still immature. Relevant standards have not yet been established, and the experience of the end users is not considered adequately so far. Quality of Experience (QoE) is the degree of delight or annoyance of the user of an application or service. It results from the fulﬁllment of his or her expectations with respect to the utility and/or enjoyment of the application or service in the light of the user’s personality and current state [2]. In this paper, QoE is used to evaluating the experience of the VR users. Despite the lack of research on user experience (UX/UE) of VR, user experience has been studied extensively in other community, especially traditional media and the design of website. There were lots of guidance for user center design and models or frameworks for the UX evaluation. In this paper, we proposed our strategies through an intensive analysis of the literature about UX models and QoE related projects. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 314–319, 2019. https://doi.org/10.1007/978-3-030-02053-8_48

Strategies and Metrics for Evaluating the Quality of Experience

315

2 Related Work In the case that the product meets the user’s functional requirements, the user’s emotional experience becomes the most important issue for system developers. So human-centered design is becoming more and more popular and user experience is getting growing attention. The international standard on ergonomics of human system interaction deﬁnes user experience as “a person’s perceptions and responses that result from the use or anticipated use of a product, system or service” [3]. Numerous studies have been done on user experience, including the deﬁnition and influence factors of user experience, measurement and evaluation of user experience, and user experience design. There are many well-established models or framework for UX in these studies. In the Elements of User Experience, Garrett. J.J. breaks down UX into ﬁve component elements, that is, the Surface Plane, the Skeleton plane, the Structure Plane, the Scope Plane and the Strategy Plane [4]. The elements model gives readers the big picture of Web user experience development, from strategy and requirements to information architecture and visual design. Another Well-known UX model is the User Experience Honeycomb by Peter Morville, which divides UX into seven facet, Useful, Usable, Desirable, Findable, Accessible, Credible and Valuable [5]. The honeycomb hits the sweet spot by serving several purposes at once. Like the models mentioned above, there are many UX models about web design explaining UX from different perspectives, Table 1 summarized some well-established models. These models can be used for either UX design or UX test. However, the original intention of these models is about Websites or Applications on mobile phones, which attach importance to interaction with the computer. Table 1. Summary of UX models Model UX elements model UX Honeycomb model 5E Principle [6] UX Hierarchy of Needs model [7] Don’t Make Me Think [8]

Author(s) Jesse James Garrett Peter Morville Whitney Quesenbery Stephen P. Anderson Steve Krug

Description Strategy, scope, structure, skeleton, and surface planes Useful, Usable, Desirable, Findable, Accessible, Credible and Valuable Effective, Efﬁcient, Engaging, Error Tolerant and Easy to Learn Functional (Useful), Reliable, Usable, Convenient, Pleasurable, Meaningful Useful, Learnable, Memorable, Effective, Efﬁcient, Desirable, Pleasurable

When it comes to the evaluation of audio and video, quality of service (QoS) or quality of experience is the more popular concept. QoS is the description or measurement of the network performance, however, it failed to take users’ perception into account. QoE is a concept shifted from QoS, highlighting the subjective feelings about

316

X. Kong and Y. Liu

products or services of the users and evaluating the system from both user perception and the system. A great deal of research has been done on the QoE evaluation of traditional multimedia, such as audio, video and 3DTV. There are quite a few models for QoE evaluation as well. Some models are based on objective parameters, such as Peak Signal to Noise Ratio (PSNR), Visual information ﬁdelity (VIF) [9], structure similarity index (SSIM) [10] and texture similarity [11], these models evaluating the quality through processing the input and output signal. Subjective approach evaluates the QoE from users’ perspective thoroughly based on the users’ rating. Janssen et al. used two metrics, naturalness and colorfulness, to evaluate the visual quality of images [12]. Chen et al. proposed a linear model with three metrics, Image quality, depth quantity and visual comfort, for 3DTV QoE evaluation [12]. These methods or models only focus on the signal itself or the users’ perception, other than both. Virtual reality is a new medium and brings the users novel experience. Unlike traditional media, which just shows the user audio or video with little interaction, virtual reality brings a new way of human-computer interaction. Some research has been done on VR QoE. Keighrey et al. studied the QoE of VR and AR with both subjective and objective parameters, and the ﬁndings demonstrate similar QoE ratings for both the AR and VR environments [13]. Hamam et al. studied QoE of the haptic virtual environment with ﬁve metrics: Media Synch, Fatigue, Haptic Rendering, Degree of Immersion and User Intuitiveness, and a mathematical model is constructed with these parameters [14]. Research on VR QoE is just in its infancy and there is no well-established evaluation methods and standards. Despite the fact QoE is important for both developers and users. Under these circumstances, Huawei initiated the QoE for VR (Quality of Experience for Virtual Reality) project proposal at the ITU-T SG in January 2017, which means a signiﬁcant step forward for the community in VR QoE evaluation.

3 Strategies and Metrics 3.1

UX or QoE

As we can see above, the community of website or mobile application design use UX while the network and the multimedia community use QoE. What makes the difference is that website values interaction and information acquisition while multimedia pays more attention on the sensor ﬁdelity. As a new media, VR is designed to provide a more natural way of interacting and immersive experience, so the evaluation of VR experience attaches importance to both interaction and sensory ﬁdelity. In this paper, we use QoE to evaluate the VR application following the Huawei’s proposal, on one hand, this paper is mainly the evaluation of users’ perception, on the other hand, the interaction with virtual environment (VE) is not completely same with that with website.

Strategies and Metrics for Evaluating the Quality of Experience

3.2

317

Strategies

A number of existing models of UX and QoE have been listed in the previous section. Although these models are not intended for VR applications, there are something in common and can be used in VR applications under applicable revision. Our strategies for QoE evaluation are derived from the shared properties of these models and the characteristics of the virtual reality. Three evaluation strategies were put forward: emphasizing user needs, emphasizing system characteristics and emphasizing emotional experience. Emphasizing user need: As described in the deﬁnition, QoE results from the fulﬁllment of the users’ expectations, so when we evaluating a VR application, we must focus on the needs of users and the degree to which user expectations are met. In addition, we can see the parameters useful/functional and pleasurable in the UX models which described the user need. Emphasizing system characteristics: As we can see in the literature, sensor ﬁdelity is an important aspect of traditional media evaluation. and it is the same with VR. VR technology has the characteristics of multi-perception, presence, interactivity and autonomy. But the importance of each characteristic varies in different application. We believe that speciﬁc characteristics of a particular application should be given more attention while other less influential features may be omitted. Emphasizing emotional experience: utility is the basic need of the user, while emotional experience determines whether users are willing to use the application. In the UX models, parameters such as convenient, efﬁcient, meaningful are all related with users’ emotional experience. In this section, we propose our strategies for the QoE evaluation of VR application. The most common parameters in these models are abstracted and then reduced to our strategies, and the characteristics of VR are fully considered in the progress. These strategies can be used as a reference for indicator design. 3.3

QoE Metrics for Astronaut Virtual Training System

Astronaut virtual training system is a VR application developed for astronaut training, such as navigation in the virtual space station and virtual assembly. No doubt the VE has its advantages on astronaut training, but the influence of the new technology on astronaut’s quality of experience is not considered adequately. Following the mentioned strategies, a set of metrics consisting of ﬁve parameters was proposed for the astronaut virtual training system. The parameters for QoE evaluation of AVTS are Utility, Presence, Usability, Side Effect and Work Load. Among these parameters, Utility and Work Load is responsible for the strategy of user needs. Obviously, the trainee of the AVTS wants a practical system which helps them improve their skills while too difﬁcult work will frustrate the trainees and lead to low QoE. Presence and Side Effect is used for evaluating the characteristics of VR. Presence helps improve the experience but the side effects are reversed. However, side effects are common in virtual reality. Usability is about easy to learn, easy to use and the efﬁciency, and these characteristics greatly affect the user’s emotional experience. User will take great delight in using a simple and efﬁcient system.

318

X. Kong and Y. Liu Table 2. AVTS QoE evaluation metrics The higher-level metrics The lower-level metrics QoE Utility Utility Presence Sensory ﬁdelity Involvement Interaction Usability Easy to use Accessible Side effect Fatigue Cybersickness Work load Physical load Cognitive load

These parameters are too abstract for the user to understand and were decomposed into several second-level indicators, the complete metrics is showed in Table 2.

4 Conclusion In this paper, three strategies for QoE evaluation are proposed through literature analysis and ﬁve metrics for the evaluation of astronaut virtual training system is designed based on the strategies. System characteristics and the features of VE are fully considered so that the metrics is applicable for the QoE evaluation of the AVTS. In the future, the weight of the parameters will be determined and an experiment will be conducted to complete the evaluation of the system. Acknowledgement. This work was supported by the Foundation of National Key Laboratory of Human Factors Engineering, Grant NO. SYFD160051802.

References 1. Qinping, Z.: Review virtual reality. Sci. Chin. Ser. F-Inf. Sci. 39(1), 2–46 (2009). (in Chinese) 2. Qualinet White Paper on Deﬁnitions of Quality of Experience (2012). European Net-work on Quality of Experience in Multimedia Systems and Services (COST Action IC 1003), Patrick Le Callet, Sebastian Möller and Andrew Perkis, eds., Lausanne, Switzer-land, Version 1.2, (2013) 3. International Organization for Standardization. Ergonomics of human system interaction Part 210: Human-centered design for interactive systems (formerly known as 13407). ISO F ± DIS 9241-210 (2009) 4. Garrett, J.J.: Elements of User Experience: User-Centered Design for the Web and Beyond. Pearson Education, London (2010) 5. User experience design. http://semanticstudios.com/user_experience_design/ 6. Quesenbery, W.: The ﬁve dimensions of usability. In: Content and Complexity, pp. 93–114. Routledge (2014)

Strategies and Metrics for Evaluating the Quality of Experience

319

7. Anderson, S.P.: Seductive Interaction Design: Creating Playful, Fun, and Effective User Experiences. Pearson Education, London (2011) 8. Krug, S.: Don’t Make Me Think! A Common Sense Approach to Web Usability. Pearson Education India, Delhi (2000) 9. Sheikh, H., Bovik, A.: Image information and visual quality. IEEE Trans. Image Process. 15(2), 430–444 (2006) 10. Wang, Z., Bovik, A.C., Sheikh, H.R., et al.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004) 11. Zujovic, J., Pappas, T.N., Neuhoff, D.L.: Structural texture similarity metrics for image analysis and retrieval. IEEE Trans. Image Process. 22(7), 2545–2558 (2013) 12. Chen, W., Fournier, J., Barkowsky, M., Le Callet, P.: Quality of experience model for 3DTV. In: Woods, A.J., Holliman, N.S., Favalora, G.E. (eds.) SPIE Stereoscopic Displays and Applications XXIII, 82881P (2012) 13. Keighrey, C., Flynn, R., Murray, S., Murray, N.A.: QoE evaluation of immersive augmented and virtual reality speech and language assessment applications. In: Ninth International Conference on Quality of Multimedia Experience. IEEE (2017) 14. Hamam, A., Saddik, A.E.: Evaluating the quality of experience of haptic-based applications through mathematical modeling. In: IEEE International Workshop on Haptic Audio Visual Environments and Games, pp. 56–61. IEEE (2012)

An Initial Design of the Mei Garden Augmented Reality Tour-Guide System Based on the Needs Analysis Hung-Yeh Lin(&) and Shelley Shwu-Ching Young Institute of Learning Sciences and Technologies, National Tsing Hua University, No. 101, Section 2, Kuang-Fu Road, 30013 Hsinchu, Taiwan (R.O.C.) [email protected], [email protected]

Abstract. In this study, we aim to build an AR tour guide system for visitors to support just-in-time experiential learning based on the needs analysis resulted from an online survey. Totally we collected 261 valid responses for analysis and then we modify the original design of the system to meet the target users’ exceptions. Consequently, the highlights of the system are proposed below: 1. A panda avatar will be designed into this system as the tour guide. 2. Some gamiﬁcation will be designed into the system. 3. There would be an online community developed. We aim to connect the real world and the virtual one to provide more vivid information and thus the users can have fun to interact with virtual objects to enhance their visiting experiences and interests. Keywords: Augmented reality tour-guide system Mei garden

Just-in-time learning

1 Introduction In recent years, novel applications of Augmented Really (AR) bring new aspects for learning and lives as well. Related studies indicate AR, if well designed, could enhance informal learning experiences and bring new learning opportunities that would not be possible before. Thus, by taking the advantages of AR indicated above, in this study, we’d like to design and develop an AR tour-guide system to support informal learning to enhance digital humanity of a signiﬁcant location called “Mei Garden” built to commemorate the ﬁrst president of National Tsing Hua University (NTHU), Dr. Yi-Qi, Mei, a highly respected scholar and educator between China and Taiwan. The Garden, bearing important image of the two Tsing Hua Universities between the Taiwan Straits, has many historically signiﬁcant architecture and is full of natural beauty, such as the Plum (pronounced as Mei in Chinese) trees symbolized the remarkable character of President Mei and many evergreen trees on it so that the Garden attracts many visitors from different parts of the world. The issue for us to tackle in this study is that, Mei Garden was built more than ﬁfty years ago, and very limited information is available on site. To date, most people do not know the signiﬁcance of this place when they come for the visit. Therefore, we aim to design and develop the Mei Garden AR tour-guide system to support visitors to grasp © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 320–325, 2019. https://doi.org/10.1007/978-3-030-02053-8_49

An Initial Design of the Mei Garden Augmented Reality Tour-Guide System

321

its signiﬁcance of this place from the aspect of digital humanity for that we will design into the system based on the rationale of just-in-time learning, and furthermore, an online community will be established accordingly for strengthening the relationship of the visitors and Tsing Hua alumni all over the world. This study is conducted in two phases: 1. Design and related information collection, and 2. System design and implementation. We are now in phase 1. To better build the system, we have conducted the needs analysis survey that mainly includes three parts: basic information of Mei Garden, digital Mei Garden system and design of the mobile guide AR system for the target users. Eventually, totally 261 valid online questionnaires were collected in December 2017 for data analysis. In summary, with the results of needs analysis, we use the state-of-art technology to connect the real world and the virtual one to provide more vivid information and thus the users can have fun to interact with the virtual objects to enhance their visiting experiences and interests. Ultimately, it is our hope to implement the system to support the visitor’s just-in-time learning of the Mei Garden, and meanwhile, an online community will be developed to strengthen the Tsing Hua ties across time and space in the future.

2 Literature Review 2.1

Digital Navigation

With the help of the tour guide, the impression of the visitor could be enhanced [1, 2]. Many places, such as monuments, museums, farms, galleries and so on, have guided tours for visitors, and even visitors will be willing to pay for the service [3]. The tour guide is a medium for tourists and places. Through the tour guide, visitors can enhance their experiences, expand their visions, and learn more about the place [4]. However, the explanation is different from each tour guide, when there is lack of human resources, therefore, a voice navigation service is use to support visiting experience. It records learning content ﬁrst, and then broadcast to visitors via the devices. Because of the rapid technology development, navigation contents are not only presented by voice but also by graphics and videos. The digital navigation enhances the visitor’s motivation and strengthens their learning impression. This study will also design a learning companion as a tour guide to increase visitors’ impression and connection between knowledge and place, to enhance visitors’ motivation via videos and games. Therefore, visitors can get better learning effectiveness via this system. 2.2

Augmented Reality (AR)

Azuma [5] argued that AR is a kind of “virtual environment (VE)”. It means a variety of “virtual reality (VR)” we usually called. However, AR allows users to see both real world and virtual objects. AR is a tool that supported reality. It presents to the users an environment with virtual and real things at the same time. If “reality” and “virtual” as two places, AR can be regarded as a bridge between the two places. It exists both “reality” and “virtual” at the same time on the bridge. Therefore, AR can integrate more

322

H.-Y. Lin and S. S.-C. Young

extra information or media supported navigation into real world. It is suitable for learning in informal spaces, such as museums, art centers, parks and so on. In the study of Sommerauer and Müller [6], they compared learning in exhibition with and without AR. The study showed that people have better learning effectiveness and higher learning motivation when they used AR to learn in the exhibition. In another study showed that using AR in the ﬁeld teaching of elementary students could increase students’ motivation to learn [7]. When they used AR at the ﬁeld, they could immediately get the information that they learned in the course at school. It can help students’ research in the ﬁeld and provide more images and knowledge. This study showed that students took the initiative to learn and have better learning effectiveness. In summary, related studies indicate AR, if well designed, could enhance informal learning experiences and bring new learning opportunities that would not be possible before.

3 Needs Analysis Survey The purpose of this study is to develop an AR tour-guide system for Mei Garden. It aims to help people know the knowledge of Mei Garden while they are visiting it. Using digital communication technology and smart devices can give user instant feedback, so users can get more information about the Garden. To better build the system, we have conducted the needs analysis survey that mainly includes three parts: basic information of Mei Garden, digital Mei Garden system and design of the mobile guide system for the target users. The survey participants are Tsing Hua university’s students, professors, staff, and people who lives near campus. Totally 261 valid online questionnaires were collected in December 2017. The highlights of the Mei AR Tour System, based on the main results of the needs analysis, are proposed below. 1. A panda avatar will be designed into this system as the tour guide, because up to 99.23% people expect to use an avatar and 60.92% people expect the avatar to be the mascot of NTHU—panda. The avatar will lead visitors by orally presenting the historical stories. In addition to the oral presentation, more multimedia information, such as texts, pictures, and animation will be included to enrich the presentation. 2. Some gamiﬁcation will be designed into the system because the participants expect to interact with the virtual plum trees by playing games while they are using this system. Thus, some basic information of the virtual plums in different periods of growth will be designed. While they are interacting with the virtual trees, they can have fun to learn more about the history and humanity aspect of the Mei Garden and Tsing Hua University as well, and meanwhile, can help the virtual trees grow and show their concerns from all over the world. 3. Users can not only interact with the virtual trees, but also explore the Garden. There are several signiﬁcant characters, such as the president Mei and fairy of the plums, in the Garden. When users encounter these characters, they may challenge them for the questions competition, just like famous puzzle game App—Quiz Ranger, of Mei Garden. If users can defeat the characters, they can earn virtual money or tokens that

An Initial Design of the Mei Garden Augmented Reality Tour-Guide System

323

they can use to buy some items, such as water or fertilizer to grow the trees, or some clothes that users can change their avatar’s appearance. Furthermore, the games have ranking functions. Users’ virtual money they have earned, rate and quantity of correct questions they answered, and so on will show in the system and in the online community. 4. In addition to the tour-guide system, there would be an online community constructed. Users can share their photos, videos, or creations, such as journals and music, of Mei Garden and Tsing Hua University onto the online community via the tour-guide system while they are physically visiting the Mei Garden. They also can use their own smart devices to access to the online community anytime, anywhere. With this community, we hope to strengthen the ties of alumni and visitors.

4 System Design With the input of needs analysis results, we modify original design of the system to meet the target users’ exceptions. Therefore, according to the highlights of need analysis proposed above, we improve function design and develop the system. The system will be designed as an Android app (Fig. 1) ﬁrst because 69.4% of the respondents use android smartphone.

Fig. 1. Schematic diagram of the system structure

By AR, and network, visitors can use their own smartphones to access more information for the signiﬁcant scenes of Mei Garden visit in a short time; even can see the different views of four seasons of Mei Garden while they are visiting. The main progress of the tour-guide by the system is indicated in Fig. 2. 1. Choose the guide method you want; 2. Scan QR code of each scene; 3. System shows the simple introduction; 4. Choose to read extra-information or not. After viewing some introductions, system will unlock new functions.

324

H.-Y. Lin and S. S.-C. Young

Fig. 2. Schematic diagram for progress of using the system

The panda avatar will lead visitors to guide the Garden visit. It can present the histories of the Garden and guide the visitors where to go. Furthermore, visitors can dress their own avatars, or change the characters that they like. According to the results of needs analysis, we design the game in two parts (Fig. 3). One is the Q&A puzzle game. This game asks visitors the knowledge of Mei Garden and Tsing Hua University. If visitors answer more quickly and correctly than NPC (Non-Player Character), they can win some tokens. They can use the tokens to buy some items and new characters. The second game is virtual tree. The virtual trees are connected to the real trees. Every plum tree will be adopted by the community, such as classmates of departments, colleagues of one ofﬁce, or same year alumni, and users can use items to care of trees.

Q&A Game

Virtual Tree

Fig. 3. Schematic diagram of the game

An Initial Design of the Mei Garden Augmented Reality Tour-Guide System

325

5 Conclusion and Future Study This study is in the preliminary stage, and the results of the needs analysis provide us useful information about the target users’ exceptions for the development of the AR tour-guide system. We hope to connect the real world and the virtual one to provide more vivid information and thus the users can have fun to interact with virtual objects to enhance their experiences and interests. Thus, the system will be able to support the visitor’s just-in-time learning of the Mei Garden, and meanwhile, an online community will be developed to strengthen the Tsing Hua ties across time and space in the future. Acknowledgments. Thanks for the grant support from the research the Ministry of Science and Technology (MOST), R.O.C., under the grant no. 105-2420-H-007-011-MY3.

References 1. Huang, J.F., Cheng, J.C., Pu, C.C., Yeh, S.C., Huang, H.T.: Visitor survey of national science and technology museum. Technol. Mus. Rev. 3(1), 52–69 (1999) 2. Young M.S.: Interpretation Education. Yang-Chih, New Taipei City (1999) 3. Chen, T.S., Lu, T.: Visitor demand and willingness to pay for interpretive service: an example of the botanical garden. Museol. Q. 20(4), 7–25 (2006) 4. Li, I.Y., Lin, Y.N.: Study on the satisfaction of exhibition guide service and willingness to revisit among museum visitors: an example of the shihsanhang museum. Museol. Q. 22(1), 93–106 (2008) 5. Azuma, R.T.: A survey of augmented reality. Presence: Teleoper. Virtual Environ. 6(4), 355–385 (1997) 6. Sommerauer, P., Müller, O.: Augmented reality in informal learning environments: a ﬁeld experiment in a mathematics exhibition. Comput. Educ. 79, 59–68 (2014) 7. Kamarainen, A.M., Metcalf, S., Grotzer, T., Browne, A., Mazzuca, D., Tutwiler, M.S., Dede, C.: EcoMOBILE: integrating augmented reality and probeware with environmental education ﬁeld trips. Comput. Educ. 68, 545–556 (2013)

Designing Virtual Reality to Enhance Spatiality Youngil Cho(&) and Suehusa Mamoru Hiroshima Kokusai Gakuin University, 6 Chome-20-1 Nakano, Aki-ku, Hiroshima 739-0321, Japan {YoungilCho,SuehusaMamoru}@edu.hkg.ac.jp

Abstract. The presence of windows initiates a sense of freedom in the inhabitants of the given space. The present paper is on the study aiming to focus on enhanced spatiality based on the previous researches with twofold: (1) to verify not only the influence of virtual windows but also the influence of the formation elements of virtual windows; (2) to clarify that whether the gender differences were considerable subjective factor or not in the influence of the virtual windows. The ﬁndings of the current study clarify that women tend to be stressful in rigid spatial feelings than men, and also that the spatial freedom of women is related to emotional feeling. On one hand, the gender differences in spatial freedom emerge to enable new forms of spatial design in virtual reality, while on the other hand it also calls for further veriﬁcation and validity of evaluation values. Keywords: Evaluation methodology Affective design Design for variability

1 Introduction How much do people yearn for freedom in space? Human beings sense and have an impression from external features not only from visual stimuli, but also from the assimilated data which consists of other senses, such as auditory, tactile, olfactory, and gustation. In the spatial distribution of limited room, all perceived information affects interactively in the conscious and unconscious state of mind. Feeling the freedom depends on the balance of the feelings among ﬁve sensors [1]. Human being tends to be stressful in rigid spatial feelings in spaces with no windows, and it is hard to be relaxed. Therefore, interior decoration could be a signiﬁcant element to reduce the stress felt from spatial limitation [2]. The activity of the parasympathetic system increases by the feeling of freedom coming from the spatial extension just with windows inclusive [3]. What is openness? Openness is deﬁned as the spatial feeling obtained through sight [4]. People who exist in conﬁned spaces without any windows, make themselves comfortable by resorting to interior design by decorating the space with portraits of scenery, interior planting, etc., to secure spatial freedom. Although this is obligatory in case of rooms that come under the Building Standard Act, it is optional for ofﬁce and underground spaces. Besides, spaces that cannot afford windows or those that have windows that are limited to ventilation purpose are on the rise. Previous research regarding spatial freedom in virtual reality [5], has been focused on the influences of © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 326–331, 2019. https://doi.org/10.1007/978-3-030-02053-8_50

Designing Virtual Reality to Enhance Spatiality

327

the virtual windows. The present study aims to focus on enhanced spatiality based on the previous researches with twofold: (1) to verify not only the influence of virtual windows but also the influence of the formation elements of virtual windows (2) to clarify that whether the gender differences were considerable subjective factor or not in the influence of the virtual windows by experimental method.

2 Methods 2.1

Stimuli and Subjects

From the experiments (1) the emotional reaction to the presence or absence of the virtual window is veriﬁed, and (2) the influence of the shape and size of the virtual windows is clariﬁed. Nineteen university students (nine males and ten females) participated in the experiments. The average age of the participants was 21.7. 2.2

Procedure

In the experiments, evaluation was performed using thirty-one evaluation values, which were selected from previous studies on stress. Virtual windows were used as stimuli which depends on three levels in two factors each: (1) factor 1: the shape of windows (round, arched, quadrangle) (2) factor 2: the size of windows (small, medium, large) [Fig. 1]. With the 2 factors 9 types of stimuli were produced, i.e., Small/Medium/Large Arched, Small/Medium/Large Round, Small/Medium/Large Quadrangle.

Fig. 1. The examples of stimuli

The subjects stood by the screen which displayed the stimuli (virtual window), and observed the virtual window for ten seconds, before evaluating their subjective feelings [Fig. 2] on thirty-one evaluation values such as, Relieve tiredness, Relax, Mood changing, Eye strain relief, Visual stimulation, Spatial changing, Pleasant view of the window, Feeling healthy, Feeling ﬁne, Bright atmosphere, Oppressed feeling, Calmness, Safeness, Concentrativeness, Sensibility to freedom, Sensibility to weather, Sensibility to time, Sensibility to space, Easy going with work, Silentness, Worried about everything, Feeling nervous, Feeling undeserving of praise, Demotivated due to disappointment, Easily frustrated, Feeling troubled after being annoyed, Energetic, Vigorousness, Fed up, Forgetful, Can work efﬁciently. The thirty-one evaluation values were selected from the previous researches on usability of virtual window [5] and psychological and physiological reaction on stress [6].

328

Y. Cho and S. Mamoru

Fig. 2. Experiment conditions

2.3

Analysis and Results

The results presented in Table 1 shows that there is a statistical signiﬁcance between male and female subjects to virtual windows. While intuitive feelings such as Relief tiredness, Bright atmosphere, Sensibility to time, Can work efﬁciently, are in relation to Small Round windows; spatial feelings such as Special changing, Concentrativeness, are in relation to Large Quadrangle. However, intuitive feeling on spatial such as Safeness, Sensibility to special, Vigorousness, are relation to both shapes, Round and Quadrangle. Table 1. The results show a statistical signiﬁcance in the size-shape combinations by gender difference. Evaluation value Relieve tiredness Spatial changing

Level of the factors in window P value Small Round .0495 Medium Round .0086 Large Quadrangle .0209 Bright atmosphere Small Round .0121 Safeness Small Round .0345 Small Quadrangle .0495 Concentrativeness Large Quadrangle .0461 Sensibility to time Small Round .0278 Sensibility to space Small Round .0193 Large Quadrangle .0454 Feeling undeserving of praise Small Arched .0352 Large Arched .0286 .0201 Vigorousness Small Round Small Quadrangle .0386 Forgetful Large Quadrangle .0294 Can work efﬁciently Small Round .0024

Designing Virtual Reality to Enhance Spatiality

329

There was no statistical signiﬁcance by the shape of windows in male subjects to virtual windows. The results presented in Table 2 shows that there is a statistical signiﬁcance by the size of windows in male subjects to virtual windows. Sensibility and intuitive feelings are prominent variables to male subjects in spatial openness by the size of windows. Table 2. The results show a statistical signiﬁcance by the size of windows in male subjects. Evaluation value Comparison Mood changing Medium vs Small Round Spatial changing Large vs Small Quadrangle Pleasant view of the window Large vs Medium Arched Feeling healthy Large vs Small Arched Oppressed feeling Small vs Medium Arched Small vs Medium Round Sensibility to freedom Large vs Small Quadrangle Medium vs Small Quadrangle Large vs Small Arched Medium vs Small Arched

P value .0430 .0019 .0463 .0377 .0084 .0062 .0095 .0412 .0033 .0055

There was no statistical signiﬁcance by the shape of windows in female subjects to virtual windows as well. The results presented in Table 3 shows that there is a statistical signiﬁcance by the size of windows in female subjects to virtual windows. While male subjects focused on sensibility and intuitive feelings through windows, female subjects focused on relax and relief feeling from the existence of windows.

Table 3. The results show a statistical signiﬁcance by the size of windows in female subjects. Evaluation value Relieve tiredness

Comparison P value Large vs Small Quadrangle .0268 Large vs Small Arched .0001 Medium vs Small Arched .0236 Relax Large vs Small Quadrangle .0179 Large vs Small Arched .0005 Mood changing Large vs Small Quadrangle .0112 Large vs Small Arched .0040 Eye strain relief Large vs Small Arched .0053 Pleasant view of the window Large vs Small Arched .0075 Feeling healthy Large vs Small Quadrangle .0154 Large vs Medium Quadrangle .0478 Large vs Medium Arched .0287 Feeling ﬁne Large vs Medium Arched .0496 Bright atmosphere Large vs Medium Quadrangle .0275 (continued)

330

Y. Cho and S. Mamoru Table 3. (continued) Evaluation value Oppressed feeling Safeness

Sensibility to freedom

Comparison Medium vs Small Round Large vs Medium Arched Large vs Small Quadrangle Large vs Medium Arched Medium vs Small Round Large vs Small Quadrangle Large vs Small Arched Medium vs Small Arched Large vs Medium Arched Medium vs Small Round

P value .0075 .0329 .0130 .0247 .0357 .0024 .0001 .0007 .0484 .0080

3 Consideration and Conclusion There is a statistical signiﬁcance between male and female subjects in their evaluation of virtual windows. While intuitive feelings such as Relief tiredness, Bright atmosphere, Sensibility to time, Can work efﬁciently, are in relation to the windows which are small and round-shaped; spatial feelings such as Special changing, Concentrativeness, are in relation to the windows which are large and quadrangle-shaped. However, intuitive feeling on spatial such as Safeness, Sensibility to special, Vigorousness, are relation to both Round and Quadrangle shapes [Table 1]. From the results, shape of window is irrelevant to gender difference; whereas size of windows is relevant to feelings from the windows by gender differences. As the following result was presented, there was no statistical signiﬁcance by the shape of windows in both male and female subjects to virtual windows. There is a statistical signiﬁcance by the size of windows in both male and female subjects in their evaluation of virtual windows. While sensibility and intuitive feelings are prominent variables to male subjects in spatial openness by the size of windows [Table 2], relax and relief feeling are prominent variables to female subjects through the existence of windows [Table 3]. Considering the results, the size factor of the virtual windows is more influential on the evaluation of spatial freedom than that of the shape factor. The male group showed a signiﬁcant difference in the evaluation values categorized as intuitiveness, whereas the female group showed a signiﬁcant difference in the evaluation values categorized as emotion and intuitiveness. It presents that the characteristics of the evaluation values of the influence of virtual windows is more complicated in female subjects than that of male subjects. The ﬁndings of the current study clarify that women tend to be stressful in rigid spatial feelings than men, also the spatial freedom of women is related to emotional feeling. While the gender differences in spatial freedom paves way to enable new forms of spatial design in virtual reality, there is also a requirement for further veriﬁcation and validity of the evaluation values.

Designing Virtual Reality to Enhance Spatiality

331

References 1. Shimizu, T.: Interior and landscape. Special Issue of JSSD, vol. 2, no. 2, pp. 50–51 (1994). in Japanese 2. Sato, M.: The influence of closed space on human physiology, psychology, behavior. J. Soc. Instrum. Control Eng. 33(12), 1028–1031 (1994). in Japanese 3. Yamada, T.: A study on mental fatigue focused on the influence of opening windows. J. Jpn. Soc. Physiol. Anthropol. 15(2), 49 (2010). in Japanese 4. Inui, M., Miyata, T., Watanabe, K.: Evaluation of spaciousness. J. Architect. Plan. Environ. Eng. 192, 49–53 (1972). in Japanese 5. Miki, M., Sato, T., Yoshimi, Y.: Forum on Information Technology of Information Processing Society of Japan, vol. 10, no. 3, pp. 657–658 (2011). in Japanese 6. Suruga, E.: Evaluation of psychological and physiological reaction of reflexology on postpartum stress. J. Jpn. Soc. Nurs. Res. 35(1), 89–98 (2012). in Japanese

UNICAP Virtual: User Experience for a VR Application in Brazilian University Christianne Soares Falcão(&) and Breno Carvalho Master Program in Creative Industries at Catholic University of Pernambuco, Rua do Principe, 526 – Boa Vista, Recife, PE, Brazil [email protected], [email protected]

Abstract. Over the last few years, there has been an increase in the number of glass wearable devices for virtual reality experiments and applications in gaming industry. Universities have been engaged in creating virtual environments and applications with a focus on entertainment industry and scientiﬁc applications, either to evaluate human factors aspects from a user experience perspective, or to check hypotheses about human-computer interaction with low-cost immersive virtual reality devices. This paper aims to present research work on the use of VR applications in which the user walks through and interacts with architectural elements in the campus of Catholic University of Pernambuco, located in northeastern Brazil. An application called UNICAP3D was developed for smartphones with Android operating system, to be used with VR glasses in the Card Board model. The survey was conducted using 114 subject user who after interacting with the application, answered a questionnaire about their experience with the virtual system and navigation devices. Based on the data collected researchers concluded that many users have not previously experienced this emerging technology, but have enjoyed the navigation form and visual feedbacks of the digital artifact. Keywords: Wearable technologies

Game design Emerging technologies

1 Introduction Interactive virtual reality applications in mobile devices are increasingly utilized in technology-based projects applicable to digital games and for education, health, marketing and entertainment industries, pointing out to new demands for research. In this context, companies have invested in devices like Oculus Rift and Google CardBoard to support Virtual Reality as an economical and effective technology in terms of user experience and be more accessible to society [1]. Virtual environments correspond to interactive, multisensory and computersynthesized environments that, when manipulated by the participants, enable them to immerse themselves in a computer simulation [2]. According Sherman and Craig [3], the virtual environment corresponds to an instance of a virtual world presented in an interactive medium such as virtual reality. Virtual Reality technology (VR) aims to simulate a reality, and vision. For most people, according to Gutiérrez et al. [4], this corresponds to the dominant perception and means to the information acquisition. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 332–337, 2019. https://doi.org/10.1007/978-3-030-02053-8_51

UNICAP Virtual: User Experience for a VR Application

333

However, a full VR simulation is not only limited to what we can see, but also involves other components that integrate perceptual experience, such as hearing, tactile feedback, smell, and taste. Although hearing, smell and taste plays important role in this experience, these sensory are poorly explored in VR environments due to the complex technological requirements [4, 5]. With the development of these multimodal and multisensory environments, it is possible for the user to interact with the digital model in a similar way as he/she interacts with a physical model [6], this can occur in two situations: interactions with systems and interactions with virtual objects. In system interactions, multimodal interfaces, including vision, stereo sound, voice input, gesture recognition, and so on, must be integrated with the VR test bed. In the interaction with virtual objects, in addition to a realistic visual feedback, the movements and behaviors must also be simulated [7]. However, the user experience results in virtual environments need to be validated, the simulated environment must replicate the real environment, and this capability is closely tied to the performance capability of both hardware and software technologies [6]. The development of technologies in virtual environments, speciﬁcally virtual reality, has led to the innovation of more intuitive interfaces for the design, allowing users navigation from the interaction with the elements of a virtual environment. This paper aims to investigate the user experience factors associated with VR application, in which the user walks through and interacts with architectural elements in the campus of Catholic University of Pernambuco (UNICAP). An application, titled UNICAP3D developed for smartphones with Android operating system, to be used with a VR glasses in the Card Board model. 1.1

Goals

This paper aims to report on a pilot study of a project focusing on virtual reality (VR) application development, which has been designed to promote the academic community and visitors know the university campus facilities from a playful and interactive interface. In this study, the main focus was to test the user experience with regards to the virtual system and navigation devices, checking if it is capable to involve participants and investigate their interests. For future studies, the intention is to continue this application by inserting more information about the architectural elements existing in the Campus, such as open spaces, furniture, signs, among others, as well as inserting a greater possibility of routes for navigation in the built environment by users. 1.2

The University Campus

The Catholic University of Pernambuco (UNICAP) is located in the city center of Recife, northeastern region of Brazil. It was originated from the Faculty of Philosophy, Sciences and Letters. Manoel da Nóbrega, established it on April 18, 1942 and in this year it begins the celebrations of 75 anniversary dedicated to education, research and extension. Currently the university has approximately 11,000 students in 38 undergraduate and postgraduate courses. The UNICAP campus is formed by buildings distributed in the central area formed by two blocks, in addition to other buildings in nearby areas, adding a constructed area of 67,646.58 m2.

334

C. S. Falcão and B. Carvalho

2 Methodology 2.1

Game Apparatus

The UNICAP3D application was developed by volunteer students of the undergraduate courses in Digital Games and Computer Science under the coordination of the interactive solutions agency Combogó Unicap. After campus floor plan analysis, a route was deﬁned to be carried out by the user from the main entrance to the campus to the building G, where the tests with the application would be happening. To model the virtual environment with the architectural elements that form the campus, a scenario was used based on 3D Max and Maya software, Adobe Photoshop software and UNITY® game engine to produce the application. After this step, from the Low poly process with the use of few polygons, a test was performed in the Unity engine to verify possible errors in the polygon mesh of the created objects. The application was developed for smartphones with operating system Android 7.0 and VR Box oculus. According Fig. 1, the navigation starts at the campus main entrance, near the Cultural Space Padre Tavares de Bragança (left side) and the chapel (right side), and continues through the parking area, passing through buildings A and B until reach the building G (back court). Before starting his journey across campus, the user would see a screen with information for navigation with the application. Throughout the course, when passing the campus main points, the user received information about the equipment, courses and sectors of the university from sign totems inserted in the application (Fig. 2). It was also possible to obtain information about cultural spaces with historical importance to the university (Fig. 3). 2.2

Participants

The application was tested in October 2017, during an important event held every year by the university, the week of integration between UNICAP and society, with activities that integrate the academic community and residents of the region. The test with UNICAP Virtual had 114 volunteer participants among students, staff, professors and visitors, of these 16 were visitors and 98 belonged to UNICAP community. 2.3

The Survey

After navigating in the application, participants answered a questionnaire using a tablet. The questions were placed in the Google Sheets tool and contained the following regarding the app experience: 1. 2. 3. 4.

Have you already known the Virtual Reality before using this application? Have you tried this technology before this application? How did you like to walk on the UNICAP Virtual Campus? What do you think about the interaction and information showed in the application?

UNICAP Virtual: User Experience for a VR Application

Fig. 1. The campus map with the buildings views from the application.

Fig. 2. An example for a totem with campus information.

335

336

C. S. Falcão and B. Carvalho

Fig. 3. An example of the application informing about the sculpture located on the library garden.

3 Results and Discussion In the ﬁrst question, 23 participants reported that they do not know what is virtual reality, while 91 said they knew the technology. The answers to the second question presented a signiﬁcant number of users that have tried VR before, a total of 53, while 61 have never tried before. In question three, participants expressed satisfaction with the application from positive responses like “very good”, “interesting” and “very cool”. However, one of the participants reported feeling dizzy and other one could not adjust the lens of the glasses. The results of the study show that users had a facility of interaction with both the system and VR glasses device, even those who didn’t know or haven’t tried the VR technology before. They give an appropriate and enthusiastic feedback with answers for question 4 likes: “It was a very well planned experience and there was no difﬁculty in the application. It was pleasurable” and “Very well done, with no difﬁculty in assimilation and adaptation to the real world”. Furthermore, some comments for question 4 emphasized the importance to have a greater amount of detail to get closer to the real environment and improve the quality of the images.

4 Conclusion In this paper, the developing process for a Virtual Reality application was presented in order to allow users to obtain information about university campus. Results showed that the VR software and tool presented good interaction with the user. Based on this, the next step is to improve the graphical quality of images to allow the addition of more useful information about the campus.

UNICAP Virtual: User Experience for a VR Application

337

In conclusion, the results of the presented study contribute to the development of user friendly applications that help in the recognition of architectural spaces in an interactive and dynamic way.

References 1. Carvalho, B., Soares, M.M., Neves, A., Lins, A., Soares, G.: The virtual reality devices applied to digital games a literature review. In: Marcelo, M.S., Francisco, R. (eds.) Ergonomics in Design: Methods and Techniques (Human Factors and Ergonomics), 1st edn., vol. 1, pp. 125–140. CRC Press, Boca Raton (2016) 2. Barﬁeld, W., Nash, E.B.: Virtual environments. In: Karwowisk, W. (ed.) International Encyclopedia of Ergonomics and Human Factors, 2nd edn. Taylor & Francis Group, Boca Raton (2006) 3. Sherman, W., Craig, A.: Understanding Virtual Reality: Interface, application and design. Elsevier Science, New York (2003) 4. Gutierrez, M., Vexo, F., Thalmann, D.: Stepping into Virtual Reality. Springer, London (2008) 5. Fortineau, E., Arnaud, S., Nahon, S.: Simulations immersives en réalité virtuelle: une innovation technologique au service de l’intervention ergonomique?. In: Proceedings of SELF 2013 - 48ème Congrès de la Société d’Ergonomie de Langue Française, 28–30 August 2013, Paris, Panthéon – Sorbonne (2013) 6. Ferrise, F., Bordegoni, M., Graziosi, S.: A method for designing users’ experience with industrial products based on a multimodal environment and mixed prototypes. Comput.Aided Des. Appl. 10(3), 461–474 (2013) 7. Wan, H., Luo, Y., Gao, S., Peng, Q.: Realistic virtual hand modeling with applications for virtual grasping. In: Proceedings of ACM SIGGRAPH International Conference on Virtual Reality Continuum and its Applications in Industry, New York, pp. 81–87 (2004). ISBN: 1-58114-884-9

State of the Art and Future Trends in the Usability of Patient Monitors Evismar Andrade1,2(&), Leo R. Quinlan2,3, Richard Harte1,2, Dara Byrne5,11, Enda Fallon4, Martina Kelly4, Paul O’Connor5, Denis O’Hora6, Michael Scully7,8, John Laffey7,8, Patrick Pladys9,10, Alain Beuchée9,10, and Gearoid ÓLaighin1,2 1

Electrical and Electronic Engineering, School of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland [email protected] 2 Human Movement Laboratory, NUI Galway, University Road, Galway, Ireland 3 Physiology, School of Medicine, NUI Galway, University Road, Galway, Ireland 4 Mechanical Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland 5 General Practice, School of Medicine, NUI Galway, University Road, Galway, Ireland 6 School of Psychology, NUI Galway, University Road, Galway, Ireland 7 Anesthesia, School of Medicine, NUI Galway, University Road, Galway, Ireland 8 Department of Anaesthesia and Intensive Care Medicine, University Hospital Galway, Newcastle, Galway, Ireland 9 Centre Hospitalier Universitaire de Rennes (CHU Rennes), rue Henri Le Guilloux, Rennes, France 10 Faculté de Médicine de l’Université de Rennes, Rennes, France 11 Irish Centre for Applied Patient Safety and Simulation, NUI Galway, University Road, Galway, Ireland

Abstract. According to the recent literature, approximately 250,000 deaths occur annually in U.S. hospitals resulting from medical error, making it the 3rd leading cause of death. One of the most commonly used devices in hospitals is the Patient Monitor (PM), a device which constantly monitors the vital signs of the patient. This paper reports on a review of the scientiﬁc literature on the usability of PMs in critical care. A detailed analysis of the data reveals that: (i) PMs are undergoing a slow, but continuous process of evolution with new advances focusing on enhancing the interaction between the caregivers and the PM, (ii) the usability of PMs is beginning to receive particular attention as usability is now considered to be strongly associated with patient safety. The data from this study will be used to carry out further investigations into the usability of PMs and to inform the design of future PMs. Keywords: Patient monitor Physiologic monitor Human factors Ergonomics Usability User experience Critical care

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 338–344, 2019. https://doi.org/10.1007/978-3-030-02053-8_52

State of the Art and Future Trends in the Usability of Patient Monitors

339

1 Introduction There are approximately 250,000 deaths annually in U.S. hospitals resulting from medical error, making it the 3rd leading cause of death; after heart disease and cancer [1]. The U.S Food and Drug Administration (FDA) reports that about one-third of adverse incidents involving medical devices relate to usability issues. Good usability in medical device design is clearly important in preventing potential risks related to user error. The industry is guided in this domain by the regulatory standard IEC 62366-1. In this context, usability is deﬁned as the extent to which the Patient Monitor can be used by nurses and physicians to achieve their patient care goals “with effectiveness, efﬁciency, and satisfaction.” The Patient monitors (PM), one of the most commonly used devices in hospitals, constantly monitors patient vital signs, such as heart-rate and SpO2 (peripheral capillary oxygen saturation, an estimate of the amount of oxygen in the blood). This device is critical for patient care, as regular assessment of multiparameter vital signs has been shown to be important in identifying patients at risk for serious adverse events, thus allowing time for clinical intervention. Here we report on the scientiﬁc literature relating to usability of (1) PMs used in hospital settings and (2) novel PM prototypes designed to overcome identiﬁed problems with current PMs used in critical care.

2 Methodology The literature search includes data up 2018 and was not limited historically. The search was carried out using the Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) guidelines and included all relevant databases. The resulting search yielded 135 publications from which 23 papers were extracted for the ﬁnal review after abstract and full paper analysis. The review was carried out by a multidisciplinary team of engineers, health scientists, anaesthetists, human factors specialists and paediatric consultants.

3 Results 3.1

Usability of Commercial Patient Monitors

Only 5 papers explicitly investigated the interaction of nurses and physicians (primary users) with commercial PMs by observing and/or interviewing users in the context use. This may be due to the difﬁculty in getting access to these environments and/or simulating a high ﬁdelity critical care scenario. Cook and Woods [2] observed the interaction of practitioners with new PM in the context of anaesthesia. They noticed that, although the design of new PMs reflected the cutting-edge technology at the time, it did not support practitioners to perform their tasks according to their typical cognitive strategies. In response to this mismatch, they adapted the PM to operate in a similar manner to the previously used monitoring equipment. Where that was not possible, they had to change the way they previously performed tasks in order to adapt to the

340

E. Andrade et al.

new technology. Almost 20 years later, the poor usability of some PMs persists. Fidler et al. [3] observed the interaction of nurses and physicians with a PM that they were familiar with. Although participants could easily perform the most common tasks, they struggled to perform more advanced tasks and thought that some aspects of the PM should be changed to improve the usability. Despite the authors conclusion that the usability and interaction design of PMs should be improved, it is not known if users of different makes and models of PMs have a similar experience with the PM tested in that study. For instance, Sanderson et al. [4] evaluated 3 different PMs using cognitive engineering methods before choosing which model to introduce to the hospital. They found that the test methods that had direct user involvement, such as usability questionnaires and usability tests, had the highest influence when deciding which system should be eventually acquired. This study shows the importance of investigating the usability of different commercial PMs. Since the nurse is a primary user of PMs in critical care, two studies sought to identify the user needs of Intensive Care Unit (ICU) nurses when interacting with the PM and how these devices could be improved to better support their tasks. Some of the issues identiﬁed in these studies included false alarms, difﬁculty in accessing desired features, poor data visualization, and lack of integration with other hospital systems [5, 6]. These studies suggest that these problems may also present when considering other makes and models of PM on the market. 3.2

Usability of Novel Interfaces for Patient Monitoring

The studies presented below have sought to solve some of the problems highlighted previously, by redesigning the traditional PM or developing completely new PM interfaces with the aim of enhancing PM usability for nurses and physicians. Graphical Displays. In complex, high risk and data-rich environments such as commercial aviation control, nuclear power plant control and military intelligence operations, there is a high volume of critical and variable data being displayed, hence accurate and efﬁcient interpretation of the data is paramount to avoid tragedies. Considering the obvious similarities, some authors argue that the critical care of multiple patients could therefore beneﬁt from the techniques used in other high-risk environments. One such technique for data visualization is the use of graphical interfaces, which integrate data in a graphical manner to improve situation awareness and enhance speed and accuracy of decision making. In the context of patient monitoring, this corresponds to synthesizing data from discrete vital signs into one or multiple graphs to facilitate the assimilation of the patient’s current state. Seven studies designed and used graphical interfaces as an alternative to the “single-sensor-single-indicator” approach utilised in traditional PMs. In two studies, the speed and accuracy of ICU nurses was found to be higher with the newly design graphical displays [7, 8] while in another study, the integrated graphical display was signiﬁcantly superior only in terms of user satisfaction when compared to a discrete bar graph display [9]. In three studies, authors found that the performance of anaesthesiologists also improved signiﬁcantly with the use of graphical displays [10–12], while in another study, anaesthesiologists were more effective using a more traditional PM display [13].

State of the Art and Future Trends in the Usability of Patient Monitors

341

Integration of Multiple Devices. The integration of multiple devices and systems is a reality of daily life. Using a smartphone, for example, provides access to a myriad of systems on one screen. Some authors evaluated the effects of data integration from multiple devices into a PM to improve situation awareness and reduce cognitive load. In two studies, researchers combined elements from the PMs such as numeric values, trends and alarm status with infusion pump information and therapy support indicators on one screen, to be viewed from a distance of 3 to 5 m. The results indicate that, when using the integrated prototype display, nurses and physicians improved their speed and accuracy; nurses experienced a lower level of frustration when identifying which patient needed priority [14, 15]. Similar results were observed when nurses interacted with a novel PM prototype, which combined the key elements of the PM main screen (traces and numerical values) and data from an even wider range of medical devices such as infusion pumps, ventilators and electronic medical records [16, 17]. The limitation for this approach is the requirement for manufacturers of medical devices and systems to provide a means for compatible integration, such that this integration can occur seamlessly across the devices from different companies. Tactile Displays for Anaesthesiology. The traditional PM displays information to nurses and physicians in two different ways; the visual display (patient monitor screen) and the auditory display (patient monitor alarm system). Four studies assumed that the use of just these display mechanisms was not sufﬁcient to optimize the work of anaesthesiologists. The authors argued that the anaesthesiologist is constantly required to switch attention between the screen and the patient because the alarm system fails to clearly indicate which vital sign is generating the alarm. The alternative presented consists of using small devices (tactiles) that utilise vibratory sequences to transmit patient information to the anaesthesiologist during the anaesthesia procedures. In this way, anaesthesiologists would be constantly informed of the patient state and the cause of alarms without having to regularly remove their eyes from the patient during a procedure. The location of the tactile on the body indicates the vital sign and the pattern of vibration indicates its level (e.g. increasing, decreasing). Ng et al. [18] found that mechanical vibration (vibro-tactile) was more appropriate than electrical stimulation (electro-tactile) to convey information. Different locations for the stimulus were compared; wrist, forearm [18], around the waist and on the back of the user [19, 20], but no signiﬁcant difference was found in terms of usability. When compared to the traditional PM, Ferris and Sarter [21] found that the performance of anaesthesiologists was signiﬁcantly improved by the vibro-tactile display. However, despite these positive results, participants felt that the traditional PM display better supported their multitasking performance. The authors suggest that this may be due to the novelty of the display and the fact that humans tend to prefer interfaces with which they are already familiar over new ones. Wearable Devices for Patient Monitoring. Brockman et al. [22] examined the acceptance of Google Glass as a patient monitoring device in a paediatric anaesthesia context. Anaesthetists who agreed to participate in the study were asked to wear the device in the operating theatre. Most participants rated Google Glass as comfortable and easy to use. In the same year, Iqbal et al. [23] tested not only the acceptance, but the feasibility of Google Glass by asking urologists to use it under simulation

342

E. Andrade et al.

conditions. It was observed that the time taken to respond to changes in vital signs was reduced and participants felt that their performance improved when using the new display. However, the author pointed to the current limitations of this device and the barriers for its adoption in real procedures. These limitations include battery life and comfort issues for users who already use normal glasses. McFarlan et al. [24] tested the usability of smartwatches, used by nurses when monitoring multiple patients. In this study, participants undertook highly realistic multitasking within a simulated clinical unit using patient mannequins. The nurses using the smartwatch signiﬁcantly improved their response time with no negative effects on their other duties. The wearable devices such as Google Glass and Smartwatches also beneﬁt from the fact that potential users are familiar with watches and glasses, therefore simply replacing them with enhanced devices should not cause discomfort.

4 Discussions and Conclusion Enhancing the performance of nurses and physicians by improving the usability of the PM is not necessarily a straightforward task, especially because its use and usefulness have been reﬁned over the last 60 years. A change in the PM may require changing the way critical care nurses and physicians perform their tasks, thus it is necessary to provide extremely solid evidence that any changes made to the interface will improve efﬁciency and effectiveness without side effects for the user or the patient. Perhaps, this might explain why some of the revolutionary interfaces presented in this paper have not yet been widely adopted in critical care at this point in time. For example, although modern PMs allow the users to visualize a patient’s vital signs in different formats such as trends, histograms and tables; graphical displays, which have been tested within the research domain for more than 20 years, do not appear to have been adopted by the manufacturers of PMs. Also, the volume of studies addressing graphical displays for PMs has declined in the last 10 years. In addition to efﬁciency and effectiveness, satisfaction is also a pillar of usability; hence good user-experience and device comfort are now essential qualities for new technology adoption by manufacturers and users. This might explain why, despite the positive performance observed in vibro-tactile displays for anaesthesia, they have not yet been widely accepted. Based on the novel prototype PM displays reviewed in this study, those that support integration of multiple devices and those that make use of well-known wearable devices, showed the most positive impact in terms of usability. The reason for this appears to be that these devices enhanced the performance of nurses and physicians without imposing a massive change in the way they perform their task and had a low impact on their current cognitive strategies. These prototypes also made use of technologies and concepts which were already familiar to most users. However, to enable the adoption of these technologies in a real world critical care environment, it is still necessary to overcome some of the current technological barriers, such as battery life and robust communications between multiple systems. It is also ultimately the responsibility of designers and researchers to demonstrate their novel solutions and generate solid evidence that their solution enhances usability and thereby improves patient care.

State of the Art and Future Trends in the Usability of Patient Monitors

343

References 1. Makary, M.A., Daniel, M.: Medical error—the third leading cause of death in the US. BMJ 353, i2139 (2016) 2. Cook, R.I., Woods, D.D.: Adapting to new technology in the operating room. Hum. Factors 38(4), 593–613 (1996) 3. Fidler, R., et al.: Human factors approach to evaluate the user interface of physiologic monitoring. J. Electrocardiol. 48(6), 982–987 (2015) 4. Liljegren, E., Osvalder, A.-L.: Cognitive engineering methods as usability evaluation tools for medical equipment. Int. J. Ind. Ergon. 34(1), 49–62 (2004) 5. Drews, F.A.: Patient monitors in critical care: lessons for improvement. In: Advances in Patient Safety: New Directions and Alternative Approaches, Performance and Tools, vol. 3. Agency for Healthcare Research and Quality (US) (2008) 6. Koch, S.H., et al.: Intensive care unit nurses’ information needs and recommendations for integrated displays to improve nurses’ situation awareness. JAMIA 19(4), 583–590 (2012) 7. Effken, J.A., et al.: Making the constraints visible: testing the ecological approach to interface design. Ergonomics 40(1), 1–27 (1997) 8. Drews, F.A., Doig, A.: Evaluation of a conﬁgural vital signs display for intensive care unit nurses. Hum. Factors 56(3), 569–580 (2014) 9. Effken, J.A., et al.: Clinical information displays to improve ICU outcomes. Int. J. Med. Inf. 77(11), 765–777 (2008) 10. Blike, G.T., et al.: A graphical object display improves anesthesiologists’ performance on a simulated diagnostic task. J. Clin. Monit. Comput. 15(1), 37–44 (1999) 11. Zhang, Y., et al.: Effects of integrated graphical displays on situation awareness in anaesthesiology. Cogn. Technol. Work 4(2), 82–90 (2002) 12. Agutter, J., et al.: Evaluation of graphic cardiovascular display in a high-ﬁdelity simulator. Anesth. Analg. 97(5), 1403–1413 (2003) 13. Jungk, A., et al.: Ergonomic Evaluation of an ecological interface and a proﬁlogram display for hemodynamic monitoring. J. Clin. Monit. Comput. 15(7–8), 469–479 (1999) 14. Görges, M., et al.: A far-view intensive care unit monitoring display enables faster triage. Dimens. Crit. Care Nurs. DCCN 30(4), 206–217 (2011) 15. Görges, M., et al.: Evaluation of an integrated intensive care unit monitoring display by critical care fellow physicians. J. Clin. Monit. Comput. 26(6), 429–436 (2012) 16. Koch, S.H., et al.: Integrated information displays for ICU nurses: ﬁeld observations, display design, and display evaluation. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 54, no. 12, pp. 932–936 (2010) 17. Koch, S.H., et al.: Evaluation of the effect of information integration in displays for ICU nurses on situation awareness and task completion time: a prospective randomized controlled study. Int. J. Med. Inf. 82(8), 665–675 (2013) 18. Ng, G., et al.: Optimizing the tactile display of physiological information: vibro-tactile vs. electro-tactile stimulation, and forearm or wrist location. In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 2007, pp. 4202–4205 (2007) 19. Ng, G., et al.: Evaluation of a tactile display around the waist for physiological monitoring under different clinical workload conditions. In: Proceedings of Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 2008, pp. 1288–1291 (2008)

344

E. Andrade et al.

20. Barralon, P., et al.: Comparison between a dorsal and a belt tactile display prototype for decoding physiological events in the operating room. J. Clin. Monit. Comput. 23(3), 137–147 (2009) 21. Ferris, T.K., Sarter, N.: Continuously informing vibrotactile displays in support of attention management and multitasking in anesthesiology. Hum. Factors 53(6), 600–611 (2011) 22. Drake-Brockman, T.F.E., et al.: Patient monitoring with Google GLASS: a pilot study of a novel monitoring technology. Paediatr Anaesth. 26(5), 539–546 (2016) 23. Iqbal, M.H., et al.: The effectiveness of Google GLASS as a vital signs monitor in surgery: a simulation study. Int. J. Surg. Lond. Engl. 36(Pt A), 293–297 (2016) 24. McFarlane, D.C., et al.: Defeating information overload in health surveillance using a metacognitive aid innovation from military combat systems. J. Def. Model. Simul. 14(4), 371–388 (2017)

Usability Evaluation of a Public Transport Mobile Ticketing Solution Daniel Meireles de Amorim(&), Teresa Galvão Dias, and Marta Campos Ferreira FEUP – Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal {daniel.amorim,tgalvao,mferreira}@fe.up.pt

Abstract. Many public transport mobile ticket solutions have arisen but new technologies can face difﬁculties when launched for a number of reasons that are directly addressed by usability evaluation – a central process in developing usable, hence useful high quality systems with which humans can interact. This paper presents a usability testing performed in context of use of a recently developed public transport mobile ticketing solution for the Metropolitan Area of Porto, Portugal. The objective was to determine if the application is easy to use and navigate for the average user and to generate valuable suggestions for making the app more user-friendly and tasks such as using public transport services and understanding fares quicker and less complicated. Along with problems detected, a list of recommendations by order of importance for improving usability aspects of the mobile application was provided. This study aims to inform developers about usability practices for making mobile ticketing solutions easier to use and stimulate its wider adoption. Keywords: Usability evaluation Mobile payments

Usability testing Public transport

1 Introduction Pressure for higher proﬁts, better service offerings and lower costs affect most, if not all, industries, and that includes public transport industry. Therefore, Public Transport Operators (PTOs) look for an answer in mobile ticketing solutions, as they make use of the customer’s own mobile device to deliver an innovative, ubiquitous and engaging service, while also minimizing investment costs for the PTO [1]. Among other beneﬁts, [2] indicate that the main reasons to adopt forms of mobile payments in public transport are convenience and the time-saving prospect, and that the main concerns in adopting it are privacy, interaction and reliability. The aforementioned reasons and concerns are directly addressed in usability evaluation, so its outcomes can be used to increase the adoption and usage of a mobile ticketing solution. Usability evaluation is a central process to develop usable, useful and high quality systems with which humans can interact. In the Metropolitan Area of Porto, Portugal, the complementary group of companies Intermodal Transports of Porto (TIP), the Faculty of Engineering of University of © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 345–351, 2019. https://doi.org/10.1007/978-3-030-02053-8_53

346

D. M. de Amorim et al.

Porto (FEUP), and technology companies jointly developed a mobile ticketing application, Anda, that allow users to ride buses, light rail and heavy-rail trains in the extensive public transport network of Porto without anything but their mobile devices. It is based on a check-in/be-out scheme, requiring an intentional user action when entering the vehicle and the alight station is automatically detected by the system, as well as intermediary stations along the trip. In the speciﬁc case of Porto, the technologies required for the app to work are Nearﬁeld communication (NFC), Bluetooth and Global Positioning System (GPS), but other solutions for different metro areas have been proposed by researchers or are already being implemented [3–5]. To start a trip with the Anda application, the customer taps the mobile phone on a NFC-enabled validator. During the trip the mobile phone interact with BLE beacons installed in metro and train stations and inside buses, through Bluetooth connection, to locate the customer along the transport network. Besides the complete dematerialization of the public transport tickets, another innovation was the post-paid billing system that optimizes the customer’s monthly tariff. That is, based on customer’s usage, invoices may come in the form of a monthly pass or individual tickets, whichever is the most economical for customers. Regardless of the technologies employed in mobile ticketing solutions, having usability consistency and harmony is essential for systems’ acceptance. However, there have not been found usability evaluation studies for public transport mobile ticketing solutions, due to being a technology newly launched in this peculiar context. Therefore, this project proposes a usability evaluation applicable to a mobile ticketing application in public transport, in the form of usability testing applied on Porto public transport system’s recently launched solution, Anda. As a result, an extensive list of recommendations for the app was created as well. This paper outline is organized as follows: a literature review on usability and public transport mobile ticketing solutions; explanation of the methodology used; description of the usability test; evaluation and discussion of results; and ﬁnal considerations.

2 Literature Review The International Organization for Standardization deﬁnes usability as the “extent to which a system, product or service can be used by speciﬁed users to achieve speciﬁed goals with effectiveness, efﬁciency and satisfaction in a speciﬁed context of use.” According to Nielsen’s [6] classical deﬁnition, usability regards all features and prospects of a system that can be interacted with, and usually it is identiﬁed by ﬁve attributes: learnability of the system, efﬁciency of use, easiness to remember (memorability), ability to recover from errors and subjective satisfaction. In the internet age, usability is a condition for survival. If users ﬁnd it hard to understand how to use a system (e.g. a mobile application or a website), they simply do not use it, there are plenty of other options available for them to try. The user will not persist in a system when facing difﬁculties unless they really must do so, such as in the case of employees and company’s intranet [7].

Usability Evaluation of a Public Transport Mobile Ticketing Solution

347

Usability tests are a form of evaluation in which users are asked to interact with a prototype and give feedback about their understanding of it and easiness to use. The goal is to collect data empirically to make the prototype match the user’s mental model [8]. As a result, Usability tests can beneﬁt both users and companies greatly, by making products easier to use and, hence, improving proﬁtability [9]. It is fundamental to plan beforehand to develop a Usability test. The key elements include, but are not limited to: a concrete purpose and problem statement; identiﬁcation of a representative sample of users; selection of test environment; clear deﬁnition of how test sessions will be conducted; collection of data; and report of ﬁndings and recommendations [9]. Public transport mobile payment is one of the innovative uses of mobile applications, relying solely on customers’ mobile devices to purchase and validate tickets. Users beneﬁt from more convenience in purchasing and validating tickets, not needing to carry cash around and saving time by avoiding queues. PTOs beneﬁt from less operational and maintenance costs. However, users can be concerned about privacy, reliability and interaction issues of such solution [1, 2]. In [10], it is explained that new m-payments systems usually fail to achieve sufﬁcient consumer adoption in order to last long enough because, among other reasons, they are not easy to use. Mobile payment solutions for Public transport differ from the ones for general uses because value co-creation occurs in dynamic context – since customers are in motion –, affecting service experience, hence the need to evaluate usability in context of use. Thus, usability concerns are intrinsically related to the success of public transport mobile ticketing solutions, either to improve user interaction and to help the mobile payment/ticketing system to acquire critical mass of consumers so that it becomes relevant and longevous.

3 Methodology Usability test guidelines were based on previous tests performed on Anda and publications of [9, 11, 12]. Guidelines are available in https://goo.gl/q67FJS and comprise questions regarding the user experience with Android OS – necessary for using Anda –, Porto public transport network and Anda; tasks concerning user’s most frequent and likely actions within Anda application; and a ﬁnal short open interview. The criteria for selecting participants were: (1) having a smartphone with NFC technology and Android Operating System, version 5.0 or higher; (2) residing in the Metropolitan Area of Porto; and (3) being a user of public transport. According to [11], between 5 and 8 users are sufﬁcient for a test to have meaningful results. Thus, 8 users were selected for testing, 4 being ﬁrst-time users of Anda and the other 4 were participants of a pre-deployment ﬁeld trial, already familiar with the app. Users provided feedback on their perceived comprehensibility and utility of app functions in a scale from 1 to 4, being 1 the lowest and meaning “Not understandable” and “Useless”, and 4 the highest and meaning “Very clear” and “Very useful”. Scale had an even number to inhibit midpoint answers as a way of avoiding decisions.

348

D. M. de Amorim et al.

Tests were conducted in context of use, in Porto’s light rail stations, and the results went through a deep analysis. Based on usability evaluation results, far-reaching recommendations were made for Anda.

4 Usability Test 4.1

The Test

The purpose of the usability testing was to determine if the Anda application was easy to use and navigate for the average user. The requisites for participating in the test are described in Sect. 3. A sample of eight users was selected, being four novice and four experienced users. Novice had no previous experience with the application. The sample is described in Table 1. Table 1. Demographics of test participants. Age 18–29 30–44 45–59 60 or more

Novice users (M/F) Experienced users (M/F) 1/2 0/1 1/0 0/0 0/0 2/0 0/0 0/1

Since a public transport mobile ticketing solution was being evaluated, the tests were performed in context, while travelling. In all cases, the test has started in a light rail station previously chosen by the user and ended in another light rail in downtown Porto. The test administrator was responsible for presenting the application to users, explaining the test purpose and how it would go. First, participants’ information was gathered. Then, users were asked to perform a set of tasks while thinking aloud. They were timed on each task to conﬁrm that tasks did not exceed a reasonable maximum duration. After ﬁnishing a task, participants were asked to assess the function and they made comments about it, positive or negative. Finally, there was an interview with focused questions and open answers to extract participants’ overall opinions and ideas from users regarding the app usability. 4.2

Test Results

In total, 16 functions were assessed: Start the app; Register; Start trip; Follow the trip (a) and (b); Check Inspections screen; Automatic trip ending; Check Fares screen; Check History screen; Navigate in History screen; Change Personal info; Change Password; Insert Payment methods; Check Payment method chosen; Log out of the app; and Log in the app. The results that averaged below 3 are presented in Fig. 1. Function “Follow the trip” was divided in two subfunctions: the ﬁrst one (a), related to tracking the user’s route, was considered somehow useful and quite understandable among participants, but a ﬁrst time user raised concerns about the inability to

Usability Evaluation of a Public Transport Mobile Ticketing Solution

349

Fig. 1. Perceived utility and comprehensibility of three application functions

distinguish between this being their route or just their current location; the second (b) related to the white rectangles that indicate public transport close to the user via Bluetooth, was viewed as barely understandable and all users agreed that this function was either useless or barely useful. Because Porto public transport has a Check-in Be out system, when users ﬁnish a journey it is detected via Bluetooth, automatically ending it in the app. Test participants found this functionality to be quite useful as it maintains Public transport network logic, but the lack of explanations can confuse users, especially new ones, as it is not clear in the app what has happened to the journey. During the task to Change password, six participants felt uneasy by the fact that their previous password was not asked before changing into a new one. Anyhow, participants evaluated the task as easy to understand and very useful. Function to Insert payment method was viewed as essential and understandable but not completely obvious because, among other reasons, there should be a Help button.

5 Evaluation and Discussion 5.1

Usability Problems Detected

After reviewing Usability tests results and comments provided, 34 Usability problems were detected, 2 of them were considered critical to the application’s Usability: (1) Anda does not have a “Help” function; (2) When changing the password, the previous password is not required. Other 15 problems were considered to be worrisome and the remaining 17 to be trivial. Problems were considered critical if seriously decreased usability; worrisome if decreased in some level; and trivial if hazarded minor aspects. 5.2

Recommendations for Anda

After analyzing usability problems detected along this study, a far-reaching list of 51 recommendations for Anda organized by order of importance was proposed in https:// goo.gl/ShsvTr. A sample with some recommendations is displayed in Table 2.

350

D. M. de Amorim et al. Table 2. Sample of usability recommendations for Anda. # 1 2 5 18 24 51

Recommendation Add a Help function. Other help information can be added along speciﬁc screens Demand for the previous password before allowing to change into a new one Place tabs on top or a more visible location Display error messages for a reasonable time or keep all of them until dismissed Enable to swipe between tabs Inform users how personal data is protected and used

6 Conclusion A sum of 34 usability problems was identiﬁed by Usability tests, 2 of them being considered critical, 15 worrisome and 17 trivial and a comprehensive list of recommendations for Anda was forged based on the aforementioned issues. At the end of this research, from observing results, it became evident that public transport mobile ticketing solutions ﬁeld is fast-growing and can beneﬁt a lot from usability research. This work aimed to contribute to better usability evaluation that would meet speciﬁc requirements of public transport mobile ticketing solutions and, as a consequence, make it easier developing and improving such solutions; and to detect Anda’s usability problems and enhance its overall usability through recommendations, thus increasing user experience and satisfaction with the app, helping to upsurge its initial adoption and to maintain the user base after it reaches a well-established position in Porto public transport system. In view of the study carried out and the results and artifacts obtained, it was possible to develop information regarding usability practices in context of use for making mobile ticketing solutions easier to use and stimulate its wider adoption. Future works can beneﬁt from the knowledge deepened in this research by applying this methodology in other public transport mobile ticketing applications.

References 1. Campos Ferreira, M., Nóvoa, H., Galvão, T., Falcão e Cunha, J.: A proposal for a public transport ticketing solution based on customer’s mobile devices. Procedia Soc. Behav. Sci. 111, 232–241 (2014) 2. Fontes, T., Costa, V., Campos Ferreira, M., Shengxiao, L., Zhao, P., Galvão, T.: Mobile payments adoption in public transport. Transp. Res. Procedia 24, 410–417 (2017) 3. Campos Ferreira, M., Galvão, T., Falcão e Cunha, J.: Design and evaluation of a mobile payment system for public transport: the MobiPag STCP prototype. In: MOBILITY 2014: 4th International Conference on Mobile Services, Resources and Users, Iaria, Paris, pp. 71– 77 (2014) 4. Rodrigues, H., José, R., Coelho, A., Melro, A., Campos Ferreira, M., Falcão e Cunha, J., Monteiro, M.P., Ribeiro, C.R.: MobiPag: integrated mobile payment, ticketing and couponing solution based on NFC. Sensors 14(8), 13389–13415 (2014)

Usability Evaluation of a Public Transport Mobile Ticketing Solution

351

5. Campos Ferreira M., Dias T.G.: How to encourage the use of public transport? A multiservice approach based on mobile technologies. In: Lecture Notes in Business Information Processing, vol. 201, pp. 314–325 (2015) 6. Nielsen, J.: Usability Engineering. Morgan Kaufmann Publishers, Waltham (1993) 7. Nielsen, J.: Usability 101: Introduction to Usability. Nielsen Norman Group (2012) 8. Rosenzweig, E.: Successful User Experience: Strategies and Roadmaps. Morgan Kaufmann Publishers, Waltham (2015) 9. Rubin, J., Chisnell, D.: Handbook of Usability Testing: How to Plan, Design, and Conduct Effective Tests. Wiley, Indianapolis (2008) 10. Slade, E.L., Williams, M.D., Dwivedi, Y.K.: Mobile payment adoption: classiﬁcation and review of the extant literature. Mark. Rev. 13, 167–190 (2013) 11. Dumas, J.S., Fox, J.E.: Usability testing: current practice and future directions. In: Sears, A. L., Jacko, J.A. (eds.) The HCI Handbook, pp. 1129–1149. Taylor and Francis, USA (2008) 12. Prescott, J., Crichton, M.: Usability testing: a quick, cheap, and effective method. In: Proceedings of 27th ACM SIGUCCS: Mile High Expectations, Denver, pp. 176–179. ACM (1999)

Investigation of the Human Factors, Usability and User Experience of Patient Monitors used in a Hospital Setting Evismar Andrade1,2(&), Leo R. Quinlan2,3, Richard Harte1,2, Dara Byrne5,11, Enda Fallon4, Martina Kelly4, Paul O’Connor5, Denis O’Hora6, Michael Scully7,8, John Laffey7,8, Patrick Pladys9,10, Alain Beuchée9,10, and Gearoid ÓLaighin1,2 1

Electrical and Electronic Engineering, School of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland [email protected] 2 Human Movement Laboratory, NUI Galway, University Road, Galway, Ireland 3 Physiology, School of Medicine, NUI Galway, University Road, Galway, Ireland 4 Mechanical Engineering, College of Engineering and Informatics, NUI Galway, University Road, Galway, Ireland 5 General Practice, School of Medicine, NUI Galway, University Road, Galway, Ireland 6 School of Psychology, NUI Galway, University Road, Galway, Ireland 7 Anesthesia, School of Medicine, NUI Galway, University Road, Galway, Ireland 8 Department of Anaesthesia and Intensive Care Medicine, University Hospital Galway, Newcastle, Galway, Ireland 9 Centre Hospitalier Universitaire de Rennes (CHU Rennes), rue Henri Le Guilloux, Rennes, France 10 Faculté de Médicine de l’Université de Rennes, Rennes, France 11 Irish Centre for Applied Patient Safety and Simulation, NUI Galway, University Road, Galway, Ireland

Abstract. According to the recent literature, approximately 250,000 deaths occur annually in U.S. hospitals resulting from medical error, making it the 3rd leading cause of death. One of the most commonly used devices in hospitals is the Patient Monitor (PM), a device which constantly monitors the vital signs of the patient. In this study, nurses and physicians who regularly interact with patient monitors were surveyed on their perceptions of the usability of the PMs they use on a regular basis. Results indicate that clinicians appeared to be mostly satisﬁed with the general usability of the monitors, particularly in terms of the information being presented and how it is presented. However, participants pointed out problems with the menu navigation during moments of high stress and the high frequency of false alarms. Also, participants expressed the desire to see additional information displayed on screen. Keywords: Patient monitor Physiologic monitor Human factors Ergonomics Usability User experience Critical care © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 352–357, 2019. https://doi.org/10.1007/978-3-030-02053-8_54

Investigation of the Human Factors, Usability and User Experience

353

1 Introduction Medical error accounts for approximately 250,000 deaths annually in U.S. [1]. According to the FDA approximately one-third of adverse incidents involving medical devices are related to usability issues [2]. In this context, usability is deﬁned as the extent to which the Patient Monitor (PM) can be used by nurses and physicians to achieve their patient care goals “with effectiveness, efﬁciency, and satisfaction.” PMs continuously monitors patient vital signs, such as heart-rate, oxygen saturation, carbon dioxide levels and respiratory rate for example. Regular assessment of vital signs is important in identifying patients at risk of adverse health events, allowing time for interventions to take place. Many studies have focused on the detrimental effects on patient care of PM alarm fatigue within the user population and indicate that excessive false alarms can contribute to caregivers not recognising adverse events when they are actively occurring [3, 4]. Despite the focus on PMs from the point of view of alarm fatigue, only a limited number of published studies have investigated the general usability and human factors of PMs from the perspective of clinicians. This is surprising, given: (i) the criticality of PMs in the clinical workflow of patient care, (ii) the important relationship described between usability/human factors engineering and risk management for medical devices in the regulatory standard IEC 62366-1/2 and, (iii) given the increased importance attributed to usability engineering in the design of medical devices by the FDA. In this study, nurses and physicians’ perceptions of the usability, human factors and user experience of the PMs they regularly use was examined using a survey method.

2 Methodology A 40-item questionnaire was designed by a multi-disciplinary team consisting of engineers, human factors specialists, health scientists and clinicians with experience in the use of Patient Monitors. The questionnaire was distributed to clinical centres in six university hospitals in France: Rennes (10 participants), Nantes (10 participants), Angers (5 participants), Poitiers (3 participants), Brest (1 participants) and Tours (3 participants). The questionnaire consisted of ﬁve distinct groups of questions: (i) Demographic and Experience (ii) General Questions (iii) Main Screen Presentation and Usability (iv) Navigation (v) Alarm System.

3 Results Participants. From the 32 completed questionnaires, two participants reported that they did not interact with the PM in their workplace and were therefore not included in the analysis. The remaining 30 participants, nurses (17) and physicians (13) worked in a Neonatal Intensive Care Unit (NICU). The majority of participants had more than 5 years’ experience with patient care (93%) and 80% of participants reported having at least 2 years’ experience using a PM. Interestingly, most participants (57%) reported

354

E. Andrade et al.

not having any ofﬁcial training in the use of the PM. All participants were considered to have at least an acceptable level of knowledge of how to use the PM. The Patient Monitor. All participants used the Philips IntelliVue PMs from the MP and MX series. These PMs are widely used in the hospitals around the world in different levels of care. 93% of participants knew the manufacturer of the PM, but only 33% could recall the exact model found in their units. The models reported to be in use were: Philips IntelliVue MP5, MP50, MP60, MP70, MX400, MX450, MX 700 and MX800. Although the PMs within the same series are different in terms of size, number of waveforms displayed simultaneously, portability etc. they have the same look and feel (Fig. 1).

MP70

MX800

Fig. 1. Example of Philips IntelliVue MP and MX series.

Tasks performed with the PM. The importance of task analysis within usability engineering is described in the IEC 62366-1/2, therefore we asked PM users how often they performed a series of tasks with the patient monitor such as: admit/discharge patient, change alarm limits, visualize trends etc. The three tasks performed most frequently by both, nurses and physicians, were: silencing alarms, event surveillance (i.e. viewing history of events) and monitoring the patient’s current vital signs (traces and values). Most participants reported to perform each of these tasks at least once a day, particularly silencing alarms (see breakdown of the three main tasks in Table 1). Results also indicate that the nurses perform a number of tasks with the PM more frequently than physicians, for example 47% of the nurses reported silencing alarms at least once every hour while only 15% of physicians reported the same frequency. Main Screen Presentation and Usability. Participants were satisﬁed with the way data is presented on the main screen. 97% of participants reported it to be “easy” or “very easy” to ﬁnd PM elements such as: buttons (physical and touchscreen), alarm limits, menu options and patient information. Yet, some participants rated the visibility of menu options (13.79%) and patient information (10%) as difﬁcult to read. Participants also found the way vital signs are displayed to be user-friendly, were satisﬁed with the colour-coding of elements and the way the traces and numerical values are displayed, even during moments of high-stress.

Investigation of the Human Factors, Usability and User Experience

355

Table 1. Frequency of the 3 tasks most performed by nurses (N) and physicians (P). The values indicate the percentage of participants who reported to perform each task at the speciﬁed frequency. Very rarely or never N P 0% 8% 0% 0%

At least once per month N P 12% 0% 12% 0%

At least once per week N P 12% 31% 29% 31%

At least once per day N P 29% 46% 41% 61%

At least once per hour N P 47% 15% 18% 8%

Silence alarms Event surveillance Monitor vital signs 6% 8% 12% 8% 25% 31% 19% 38% 38% 15%

The two main data visualization elements used to convey the current state of the patient in the main screen are the traces/waveforms and the numerical values of the vital signs. For the majority of vital signs that can be presented in both formats, most participants reported that the traces and numerical values have the same importance in identifying changes in the patients’ state. When asked if they would like to see additional information to be presented on screen, 60% reported that they would. The additional information participants would like to see on the screen were: vital sign trends for a longer timescale (40%), EMR/EHR information (17%), treatment suggestions based on the vital signs (17%) and drug information (6%). Almost all participants (93%) agreed that there is no unnecessary/excessive information displayed by the PM. Navigation and Interaction. The navigation system was found to be easy to use, even in moments of high-stress by 74% participants, the remaining participants rated the difﬁculty of navigating the system during moments of high-stress as “neutral” or “difﬁcult”. One participant commented on the difﬁculty of navigating to a particular feature during adverse events, “we often try to put the chronometer (stopwatch function) in emergency cardiac arrest and it is not intuitive”. Another participant commented on the care group functionality which allows the display of alarms from other patients to appear. This feature can potentially block relevant information with false alarms, causing moments of stress if the caregiver cannot close the pop-up window quickly enough. The preferred method of interaction with the PM was via the touchscreen interface (93%), nonetheless, among the users of older MP series, participants reported to also use physical buttons (26%) and the knob control (32%). Only 1 participant reported that they use a mouse for navigation. Alarm System. All participants agreed that false alarms are a problem with the PM. Approximately 23% of physicians consider false alarms to be a serious problem, and this number rises to 47% for nurses. Most participants reported a high frequency of false alarms, for example: low-risk false alarms were reported to happen every few hours by 57% of participants and high-risk false alarms by 20% of participants. Despite the number of false alarms, 87% of participants reported to be satisﬁed with the alarm

356

E. Andrade et al.

system, even during moments of high-stress. High levels of satisfaction were also reported by the participants for each element of the alarm system such as: buttons, alarm information and volume. General Impressions of PMs. When asked about their overall level of satisfaction with the PM in their unit, almost all participants reported to be “satisﬁed” (77%) or “very satisﬁed” (20%). 70% of participants disagreed with the statement “The lack of usability of patient monitors have contributed to clinical challenges during patient care”.

4 Discussion and Conclusion Although several past studies have sought to develop and test new interfaces for vital sign monitoring in critical care, very few studies have actually investigated the usability, human factors and user experience of the PMs used in the hospital. This is a risk, as introducing a technology without fully understanding the needs of the users, the context of use and the tasks which need to be supported by the new technology, can often be the cause of usability problems, particularly in critical care [5, 6]. This study sought to understand how nurses and physicians perceive one of the main patient care monitoring devices in the hospital. Results indicate that in general, the users are satisﬁed with the PM in terms of data presentation, means of interaction and the alarm system. This is a positive ﬁnding for the PM and may be due to the evolution of the PM over the past 60 years and the increased importance attributed to usability engineering in the design of medical devices by regulatory organizations such as the FDA. Despite the positive user experience reported by nurses and physicians in this study, there are clearly still areas that need improvement in physiologic monitoring. For example, most participants would like to see additional information presented in the PM interface. This issue was identiﬁed previously by Koch et al. [7] who observed that nurses often need to gather information from different devices, which consumes time and increases cognitive load. Not many studies have sought to enhance the PM user experience by integrating disparate devices [7–11]. Another well-known problem is false alarms. According to the literature, false alarms are still a major usability problem for patient monitors [3]. Our current study conﬁrms this ﬁnding; however, to our surprise, participants reported a high level of satisfaction with the alarm system despite acknowledging the criticality and high frequency of false alarms in their units. Although the results indicate that nurses and physicians are at a minimum “satisﬁed” with the PM used in their clinical centre. The PMs tested were all from Philips, more research is necessary to verify to what extent these ﬁndings translate to PMs from other manufactures. For this reason and also to get a multinational perspective on the usability of PMs, we are expanding this research to hospitals in Ireland and Brazil. This study provides important information in supporting the design of future monitoring equipment.

Investigation of the Human Factors, Usability and User Experience

357

References 1. Makary, M.A., Daniel, M.: Medical error—the third leading cause of death in the US. BMJ 353, i2139 (2016) 2. Wiklund, M.: Fitting human factors in the product development process. MDDI Online. https://www.mddionline.com/ﬁtting-human-factors-product-development-process. Accessed 09 July 2018 3. Cvach, M.: Monitor alarm fatigue: an integrative review. Biomed. Instrum. Technol. 46(4), 268–277 (2012) 4. Lawless, S.T.: Crying wolf: false alarms in a pediatric intensive care unit. Crit. Care Med. 22(6), 981–985 (1994) 5. Cook, R.I., Woods, D.D.: Adapting to new technology in the operating room. Hum. Factors 38(4), 593–613 (1996) 6. Drews, F.A.: Patient monitors in critical care: lessons for improvement. In: Henriksen, K., Battles, J.B., Keyes, M.A., Grady, M.L. (eds.) Advances in Patient Safety: New Directions and Alternative Approaches. Performance and Tools, vol. 3. Agency for Healthcare Research and Quality (US), Rockville (2008) 7. Koch, S.H., et al.: Intensive care unit nurses’ information needs and recommendations for integrated displays to improve nurses’ situation awareness. J. Am. Med. Inform. Assoc. JAMIA 19(4), 583–590 (2012) 8. Görges, M., Kück, K., Koch, S.H., Agutter, J., Westenskow, D.R.: A far-view intensive care unit monitoring display enables faster triage. Dimens. Crit. Care Nurs. DCCN 30(4), 206–217 (2011) 9. Görges, M., Westenskow, D.R., Markewitz, B.A.: Evaluation of an integrated intensive care unit monitoring display by critical care fellow physicians. J. Clin. Monit. Comput. 26(6), 429–436 (2012) 10. Koch, S.H., Staggers, N., Weir, C., Agutter, J., Liu, D., Westenskow, D.R.: Integrated information displays for ICU nurses: ﬁeld observations, display design, and display evaluation. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting. vol. 54, no. 12, pp. 932–936 (2010) 11. Koch, S.H., et al.: Evaluation of the effect of information integration in displays for ICU nurses on situation awareness and task completion time: a prospective randomized controlled study. Int. J. Med. Inf. 82(8), 665–675 (2013)

Usability Assessment as a Guide to Improve the System Design of a Corneal Topographer Carlos Aceves-Gonzalez1(&), Carlos D. de Leon-Zuloaga1, Zuli T. Galindo-Estupiñan1, and Citlali Díaz-Gutierrez2 1

Centro de Investigaciones en Ergonomía, Universidad de Guadalajara, Guadalajara, Mexico [email protected], [email protected], [email protected] 2 Bleps Vision, Metepec, Mexico [email protected]

Abstract. The study aimed to assess the efﬁciency, efﬁcacy and satisfaction of a redesigned portable corneal topography system to identify design solutions for system improvement. Twelve participants had to complete different tasks interacting with the software and the other elements of the system in a controlled scenario. After, a Retrospective Think Aloud was carried out focusing on the most challenging tasks during the system interaction. Finally, a satisfaction questionnaire was used to assess user perception. The most frequent errors and difﬁculties were identiﬁed along with the participants’ insights into the design elements for improvement of the system. Design solutions were generated to address software deﬁciencies and to improve the design of all aspects of the system. Usability analysis of software showed high relevance in the correct usage of the entire system, and along with the quick guide provides most of the feedback to understand the proper use of the handheld device. Keywords: System Usability

Medical devices Retrospective Think Aloud

1 Introduction Usability assessment of medical devices is so important during the design process since it helps to reduce potential troubles before the product arrives at the market [1, 2]. Most of the medical devices require using software to record and process patient information. This dependency makes the device and software part of a product-system including the user manuals and other technical information [3]. Iterative usability assessments can provide answers to speciﬁc questions about design decisions and help to identify potential problems that the user may encounter in the interaction with the system [4, 5]. With this in mind, a new portable corneal topography system named TOCO is currently under development by Bleps Vision. Recently, the system was re-designed considering results from a previous study [6]. The latest version of TOCO system comprises three elements: (1) software to record patient information and to process corneal scan; (2) handheld device to scan the patient’s cornea, and (3) quick guide to provide basic information regarding the handheld device operation. The study aimed to © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 358–364, 2019. https://doi.org/10.1007/978-3-030-02053-8_55

Usability Assessment as a Guide to Improve the System Design

359

assess the efﬁciency, efﬁcacy, and satisfaction of the three elements to identify design solutions for system improvement.

2 Method 2.1

Participants

A total of twelve optometrists from the Bachelor Programme at FESI, UNAM took part in this study. Seven men and ﬁve women aged in average 35.83 years old (±8.35). Half of the participants had a previous experience using TOCO system, and the other half had never used the system. All the participants received information about their participation on the study and signed a consent letter. 2.2

Procedure

Since the software was a prototyped version in Just In Mind® and not all its features were functioning, the test was carried out in two phases: in phase I, users interacted with the software itself; in phase II, users interacted with the quick guide and the handheld. Phase I – Software. The assessment was carried out using a Retrospective ThinkAloud [7] method. The focus was on the actions where the participants had an error with the intention of identifying the design elements responsible for the mistake and the participants´ suggestions for software design improvement. A check-list of the tasks and subtasks was made based on a Hierarchical Task Analysis [8] to facilitate errors recording and to spot the pain points along the interaction. A list of with seven tasks was given to the participants who were instructed to read it. Participants received the instructions of always using the mouse for software interaction, no time limit for the tasks and they will not have help from the researchers. After ﬁnishing, a questionnaire was used to assess users’ satisfaction. It involved three questions using a type Likert scale, being “1 = totally agree” and “7 = totally disagree”, followed by an open-ended question. The afﬁrmations of perceived satisfaction were taken from the Post Study System Usability Questionnaire (PSSUQ) created by [9]. Phase II - Handheld and Quick Guide. The handheld and the quick guide assessment were also carried out with a Retrospective Think-Aloud method. A check-list with the tasks and subtasks of the handheld interaction was made based in a Hierarchical Task Analysis to the record errors, to record how many attempts to scan were made and to indicate the pain points along the interaction. Regarding the quick guide, it was recorded if the participant read it or not. Optometrists were instructed to perform a routine scan twice and conﬁrm if the picture was correct according to their perception. In the end, two questionnaires – similar than that one used for the software - were used to evaluate users’ satisfaction with the handheld and quick guide use.

360

C. Aceves-Gonzalez et al.

3 Results 3.1

Software Assessment Results

The users’ insights about the software design were the following: the icons and labels should be clearer. Feedback is required to indicate if users are processing the scan correctly and but also reference guides for how an accurate scan should be. The screen with results should be improved, and the information must be easier to explore. Results of the software efﬁcacy are presented in Table 1. Table 1. Software efﬁcacy and researcher’s insights. Task - Create a new user (sign up) - Sign in to software - Create patient proﬁle

Interactions with error 5

Completed tasks 11

2

11

5

12

- Record a scan

10

10

- Process scan

12

5

- Show results - Print results

3 8

12 11

- Save results

4

12

Researcher insight/observations The user assumed the account was already set up The user signed in without ﬁlling the user and password boxesa Tried to use the tab function to change input box Wrote the month in text instead of month number Tried to use the search bara Tried to scroll downa The user advanced forward without completing the taska The user advanced forward without completing the taska The user tried to modify the form of the circle Passive feedback helped the user to realise which was the next step Tried to tweak result preferencesa Participant reached the ﬁve chances limit to make a subtask Participant searched print option in the menu option Participant tried to save ﬁrst Participants click in label instead of the icona Participants click in label instead of the icona

- Return to the 4 12 patient proﬁle - Log out 1 12 The user could not ﬁnd log out button a The task was not completed or with error due to prototype malfunction.

Usability Assessment as a Guide to Improve the System Design

3.2

361

Handheld and Quick Guide’s Assessment Results

All participants completed the scans of the two eyes of the patients. Only 6 (25%) scans were successful in its ﬁrst attempt, the average number of attempts on each scan was 2.37 (±1.13) and being ﬁve the maximum number of attempts recorded. Observed behaviours of participants showed the use of two hands to hold the handheld (4), the use of ﬁngers to remove the top or bottom eyelid out of the frame (6) and two participants held their patient’s heads to align the handheld with the patient eye gaze. Participants stated that the handheld should be modiﬁed to provide a point of support, either at the patient’s forehead or in another noninvasive area of the patient (4). The device should indicate the correct distance to the patient’s cornea (4). The action button shape should be changed (1), and it should be like a gun trigger (3). A coloured gaze point should be included to allow the patients a visual reference to place their gaze (2). A reset button to delete a wrong scan should be included (1), as well as a longer USB cable (1) and a wrist leash, was pointed out as being ideal. Regarding the quick guide, participants mentioned that it should include the recommended distance between the handheld and the patient’s eye (4). Likewise, the writing style and terminology should be improved (3) and comprise scan protocol suggestions (3). They also raise the need for warnings on the ﬁrst page (2), and a diagram of the system elements (1). It should indicate the tilt and rotation angle (1) and if it can be sustained on the patients’ face (2). It needs to describe how the turn on feedback looks like in the handheld (1) and, a clear image of how a correct scan should look like (1). 3.3

System’S Perceived Satisfaction

The average score of the system perceived satisfaction was 1.94 (±1.01). Results of average scores of each system element are shown in Table 2. Table 2. Satisfaction questionnaire scores. System element

Afﬁrmations

Software

The organisation of the information in the software was clear I was able to complete all the tasks without errors The interface of the software was pleasant

Score SD for each afﬁrmation

Average score for each afﬁrmation 1.67

0.78

2.17

0.83

1.42

0.67

Average score for systems’ element 1.75

SD for system element score 0.81

(continued)

362

C. Aceves-Gonzalez et al. Table 2. (continued)

System element

Afﬁrmations

Quick Guide

The information provided by the Quick Guide was clear The information in the Quick Guide was effective in helping me to complete the task The Quick Guide has all the elements I need to complete the task I was able to complete all the tasks without errors I was able to complete the task quickly using the TOCO handheld I liked using the TOCO handheld

TOCO Handheld

Average score for each afﬁrmation 2.17

Score SD for each afﬁrmation 1.11

2.25

1.22

2.42

1.31

2.33

1.07

2

0.74

1

0

Average score for systems’ element 2.28

SD for system element score 1.19

1.78

0.93

4 Discussion and Conclusion The study aimed to assess the efﬁciency, efﬁcacy, and satisfaction of the redesign of the TOCO system. The results allowed to identify missing or deﬁcient elements in the software design which have an impact on the usability of the other elements of the system. As an example, labelling on some buttons and comprehensibility of the icons design were identiﬁed as a potential problem when interacting with the software. Correct labelling needs to be part of the software, handheld and to the exploration diagram to be included in the quick guide, this could allow a better understanding and functioning of the entire system [10]. Also, it is needed to consider the use of familiar language for users [4] as some participants stated that terminology on the software was unclear. Results suggest that it is needed a change in the order of sections within the patient’s main page considering the distribution of similar medical devices. This arrangement would help to address the cultural conventions of users making the software easy to learn [11]. Another important ﬁnding was the requirement of including an image of a correct scan, which is in line with the idea of [12] that it is better for comprehension to use shorter connections between the real objects and its graphic representation. Regarding the handheld, participants had to use both hands to hold it and rested it on the patient’s face. These actions are identiﬁed as those allowing participants to have

Usability Assessment as a Guide to Improve the System Design

363

more control over the framing and the scan capturing [13]. Therefore, to achieve such level of control the handheld design should be modiﬁed, especially the grip, holding and support components. This situation shows that users did not have a clear idea of how using the handheld properly and it highlights the need for feedback mechanisms from the system to guide the scanning [4, 14, 15]. Based on the insights obtained from this study, users contributed with possible solutions for TOCO system improvement. Particularly, it was identiﬁed some usability issues regarding software and the quick guide, which are elements that should provide information for creating the mental model of how the system works. The proper integration of the system elements would enhance the system’s usability as exempliﬁed by [16]; if the elements of the system work as a unity, the user creates a mental model of the steps to achieve their goal, which will be translated as an effective, efﬁcient and satisfying interaction to the user. Additional iterations of usability assessment should be made to TOCO system when software is fully working and recommendations obtained from this study taken into account. Acknowledgements. CONACYT supported this work through the Fondo de Innovación Tecnológica FIT [grant number ECO-2016-C01-274898]. We thank Oscar A. Ramos Montes and Esther Alejandra Malagon Rojas from the FESI, UNAM for helping with the process of participant recruitment and managing the facilities to run this research.

References 1. Brooke, R.E., Isherwood, S., Herbert, N.C., Raynor, D.K., Knapp, P.: Hearing aid instruction booklets: employing usability testing to determine effectiveness. Am. J. Audiol. 21(2), 206–214 (2012) 2. Fries, R.C.: Reliable Design of Medical Devices. CRC Press, Boca Raton (2012) 3. Wiklund, M.E., Kendler, J., Strochlic, A.Y.: Usability Testing of Medical Devices. CRC Press, Boca Raton (2015) 4. Nielsen, J.: Usability Engineering. Elsevier, Amsterdam (1994) 5. Adams, A.: Usability testing in information design. In: Zwaga, H., Boersema, T., Hoonhout, H. (eds.) Visual Information for Everyday Use: Design and Research Perspectives, pp. 3–20. CRC Press (2003) 6. Aceves-González, C., Galindo-Estupiñan, Z.T., Landa-Avila, I.C., Díaz-Gutiérrez, C., Prado-Jiménez, S.D.: Usability assessment of a portable corneal topography device. In: International Conference on Applied Human Factors and Ergonomics, pp. 639–650. Springer, Cham (2018) 7. Paz, F., Pow-Sang, J.A.: A systematic mapping review of usability evaluation methods for software development process. Int. J. Softw. Eng. Appl. 10(1), 165–178 (2016) 8. Leventhal, L., Barnes, J.: Usability Engineering: Process, Products and Examples. PrenticeHall, Upper Saddle River (2007) 9. Lewis, J.R.: IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use. Int. J. Hum. Comput. Interact. 7(1), 57–78 (1995) 10. De Leon Zuloaga, C.D.: Evaluación del efecto de tres tipos de interfaz en la usabilidad de una receta en pacientes con diabetes mellitus tipo 2, Tesis de Maestría en Ergonomía. Universidad de Guadalajara, México (2017)

364

C. Aceves-Gonzalez et al.

11. Vu, K.P.L., Proctor, R.W.: Handbook of Human Factors in Web Design. CRC Press, Boca Raton (2011) 12. Hutchins, E.L., Hollan, J.D., Norman, D.A.: Direct manipulation interfaces. Hum. Comput. Interact. 1(4), 311–338 (1985) 13. Myers, B.A., Bhatnagar, R., Nichols, J., Peck, C.H., Kong, D., Miller, R., Long, A.C.: Interacting at a distance: measuring the performance of laser pointers and other devices. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 33– 40. ACM (2002) 14. Jordan, P.W.: An Introduction to Usability. CRC Press, Boca Raton (1998) 15. Saffer, D.: Microinteractions: Designing with Details. O’Reilly Media Inc., Sebastopol (2013) 16. Ganier, F.: Factors affecting the processing of procedural instructions: implications for document design. IEEE Trans. Prof. Commun. 47(1), 15–26 (2004)

Investigation of Usability Issues Through Physiological Tools: An Experimental Study with Tourism Websites Jyotish Kumar(&) and Jyoti Kumar Indian Institute of Technology Delhi, New Delhi, India [email protected], [email protected]

Abstract. Existing usability testing methods using verbal reports and behavioural observations estimate the time consumed in different tasks and errors occurred therein etc., they fail to provide feedback on traits like attraction and excitement felt by users which is required for tourism websites to succeed. Thus there is a need to expand the usability parameters beyond cognitive measures like ease of use and efﬁciency etc. and into affective measures like attraction, enjoyment, excitement, etc. This study reports use of physiological measures within usability testing setup to measure affective parameters by means of galvanic skin response, eye tracker measures and behavioural responses. In light of the ﬁndings, this paper argues use of physiological measures as an objective, low cost and practical addition to usability testing practice to enrich the usability testing ﬁndings and pave way for website designers for affective improvements into their sites. Keywords: Usability testing

Eye tracker GSR, behavioural response

1 Introduction Website design is an iterative process and requires usability testing at various stages. According to the ISO 9241-11 standard, usability is deﬁned as the “extent to which a product can be used by speciﬁed users to achieve speciﬁed goals with effectiveness, efﬁciency and satisfaction in a speciﬁed context of use” (ISO 9241-11 1998). Various usability testing methods like heuristic evaluation, cognitive walkthrough, think aloud, etc. have been used to evaluate HCI systems (Plantak Vukovac, Kirinic, and Klicek 2010). However, prime focus of usability testing has been to measure cognitive aspects of the HCI systems like ease of use, efﬁciency etc. Emergence of advanced HCI technologies like AI, expansion in usages of HCI like persuasive commerce & brand building, changing market conditions and changing user expectation are pushing the relationship between user and computer beyond cognitive and into affective realms [1]. In order to address the affective need of users new HCI systems will have to be designed and evaluated for their ability to motivate, arouse, persuade, attract, engage, satisfy [2, 3]. Hence, there is a need to expand the usability testing beyond the cognitive functions of users and include affective states of users while testing of websites. Available literature in usability has used self-reported or behavioural measures to © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 365–371, 2019. https://doi.org/10.1007/978-3-030-02053-8_56

366

J. Kumar and J. Kumar

investigate usability issues, however, with the advancement of physiological measurement tools like Galvanic Skin Response (GSR) etc. it has become possible now to detect small changes in human physiology which can be used to assess the affective state of users during usage of HCI systems and this data is more likely to be free from observer and reporter biases. This study reports usability testing of three tourism websites of different usability scores where 20 participants completed tasks while their physiological data using Eye Tracker, GSR, behavioural video and screen activities were recorded. After the experimental recordings, six behavioural markers, namely, ‘engagement’, ‘excitement’, ‘confusion’, ‘boredom’, ‘enjoyment’, and ‘frustration’ were used to mark user behaviour on the recorded video by three observers who by consensus decided the marker for observed behaviour. Eye tracker data, namely, ‘ﬁxation count’, ‘ﬁxation duration’, ‘saccade duration’, ‘saccade count’, ‘blink rate’, ‘saccadic amplitude’, ‘scan-path length’, ‘pupil dilation’ was used along with GSR data to measure arousal levels. At the end of use of websites, participants also rated their subjective experience through a questionnaire. Inter-correlation between behavioural markers and physiological data showed statistically signiﬁcant correlation.

2 Related Work Eye trackers have been in use for sometime, mostly for study of visual attention. Some of the most frequent features of eye tracking often used in research are Eye ﬁxation, Saccade Eye blink, Pupil dilation and Scan-path are studies reported in literature. Next paragraphs describe related work done to each of the eye track parameters. Eye ﬁxations are momentary pauses of eye movements within a particular location that extend for a minimum duration [4]. Goldberg and Kotaval reported that smaller average ﬁxations indicate less meaningful elements and repeat ﬁxations show absence of meaningful cues or lack of visibility [5]. Also, it has been reported that with increase in difﬁculty, number of ﬁxations increase [6] and multiple short ﬁxations are observed where expected information is missing [7]. Saccades describe ballistic movement of eye between two successive ﬁxations [8]. Goldberg et al. reported that larger saccades indicate meaningful clues, as attention is drawn from a distance. He also added that more number of saccades indicate more searching [9]. Ehmke et al. reported that saccadic amplitude decreases with increase in task difﬁculty in visual search context [10]. Eye blinks are also considered as potential marker for task difﬁculty, uncertainty and decision making. Boehms and Davis et al. reported that number of eye blinks increase with increase in task difﬁculty [11]. Siegle et al. reported that latency of eye blinks has relation with decision making [12]. Scan-path has been used as a measure of task efﬁciency and uncertainty. Goldberg and Kotaval reported that, longer scan path length indicates less efﬁcient searching (perhaps due to sub-optimal layout) [5]. Jacob et al. suggested scan-path direction as a marker of navigation and search strategy [13]. Galvanic Skin Response (GSR) has been used to study arousal levels in users. GSR is the change in electrical conductivity of human skin caused by environmental events as well as individual’s psychological state. The change is caused by the degree to which a person’s sweat glands are active and is one of the direct assessments of arousal

Investigation of Usability Issues Through Physiological Tools

367

[14]. Skin electrodes, are placed at the palmer surface because of the more involvement of sweat glands in this area [15]. Skin conductivity and workload are directly associated [14], skin conductivity increases with increase in task difﬁculty [16]. In this study, GSR was used to compare the task difﬁculty level of all three websites.

3 Experimental Set up and Procedure Experiments were conducted in a lab environment with one way mirror setup where experimenters could observe the participants across the mirror while the participants could not see the experimenter. Before starting the experiment, signed consent from participants was obtained, task and scenario were explained. The participants’ eye movements were tracked using the Remote Eye Tracking Device (RED) with sampling rate 60 Hz. Simultaneously, the skin conductivity was recorded using GSR sensor with sampling rate of 100 Hz. Six markers namely, ‘engagement’, ‘excitement’, ‘enjoyment’, ‘boredom’, ‘confusion’ and ‘frustration’ were used in behavioural observation software, Morae. In order to obtain equal number of marker count for each participant for ease of analysis, behavioural observations were marked by observers at an interval of 30 s. 3.1

Stimuli Websites

Initially, a total of 10 tourism websites were identiﬁed to create a pool for expert rating on usability parameters. All 10 websites were assessed by three usability experts using System Usability Scale (SUS). Experts performed given tasks on each website like ‘input personal details’, ‘ﬁlter and comparison of tour packages’ etc. before they rated the sites using SUS. 3 out of 10 websites were selected for usability testing. Websites with highest (Make My Trip, SUS = 87.5), medium (Yatra.com, SUS = 72.5) and lowest (Thomascook.in, SUS = 60) usability scores were selected for experiment. 3.2

Task and Scenario

All the 20 participants (mean age 20.4, std. dev. 2.6) were given a scenario to read before performing tasks on websites. The scenario was given on a printed paper and it read “Congratulations! It is our pleasure to announce that you have been selected as lucky winner for our ‘Swipe and Win’ contest held at IIT Delhi during Rendezvous 2016 (Annual cultural festival at IIT Delhi). For the grand prize, you are awarded 3 days and 4 nights holiday package worth of INR 60,000 for two persons in Goa. This offer is applicable only on Make My Trip, Yatra.com and thomascook.in”.

4 Results and Discussion The participants subjective post task response did not show much difference in the three websites on parameters of ‘engagement’ and ‘confusion’ while showed moderate difference for ‘excitement’ and ‘frustration’ and strong difference for ‘enjoyment’ and

368

J. Kumar and J. Kumar

‘boredom’ (Fig. 1a). On the other hand (Fig. 1b) behavioural markers logged by three observers showed statistically signiﬁcant difference in the three websites on all 6 markers at 95% conﬁdence interval. Mouse clicks, mouse movements and time taken to complete task were also signiﬁcantly different in the three websites.

Fig. 1. (a) Post task response (mean) of users, Fig. 1(b) Comparative study of behavioural markers and input device parameters

This difference in subjective rating and objective markers has practical implications in affective testing of websites. Inaccuracy of memory based ratings are well reported phenomena [17]. The ﬁndings here are giving a preliminary caution against using self reported measures in affective testing for website evaluation. Counts of positive markers, namely, ‘engagement’, ‘excitement’ and ‘enjoyment’ were found to have positive correlation with SUS scores of the websites and negative markers ‘boredom’, ‘confusion’ and ‘frustration’ had negative correlation which conﬁrms that more usable websites generated more positive affects. Further, correlation between eight eye tracker measures and six markers were calculated in Table 1. Fixation count was found to have strong negative correlation with ‘engagement’, ‘excitement’ and ‘enjoyment’ and positive correlation with ‘boredom’ and ‘frustration’. This implies that ﬁxation count increased with difﬁculties in website and vice versa. Average ﬁxation duration was found to have positive correlation with engagement and negative correlation with boredom and frustration. This implies that longer ﬁxation duration is a marker of more engagement and vice versa. Saccades count was found to have signiﬁcant correlation with frustration. As discussed in Sect. 2, larger number of saccades count shows more searching which leads to frustration in user. hence, large number saccades count can be indicative as marker of frustration. Total saccadic duration was found to have strong negative correlation with engagement and positive correlation with boredom and frustration. As discussed in Sect. 2, less meaningful clues and more searching in websites lead to less engagement and more boredom and frustration. Average of saccadic amplitude was found to

Investigation of Usability Issues Through Physiological Tools

369

Table 1. Inter-correlation between eye tracker measures and behavioural markers

have strong positive correlation with excitement and enjoyment. A negative correlation was observed between average of saccadic amplitude and boredom. These results show that average of saccadic amplitude increases in website with increase in positive emotions and vice versa. Blink count was found to be positively correlated with boredom and frustration. Blink count increases with increase in boredom and frustration which reflects the task difﬁculty and uncertainty in decision making while using the websites. Scanpath length was found to have negative correlation with enjoyment and positive correlation with confusion. As discussed in Sect. 2, larger scanpath length indicates less efﬁcient website and poor navigation. Correlation results show that scanpath length increases with increase in confusion and decrease in enjoyment. Pupil size was found to have positive correlation with enjoyment and negative correlation with confusion. Pupil size increases with increase in arousal. Website clues and design attracts user’s attention and excites users whereas, a suboptimal layout and poor navigation creates confusion and boredom for users. ANOVA at 95% conﬁdence interval was performed to compare mean of eye tracking measures for all 20 participants across three tourism websites in Table 2. Eye tracking measures were found to have signiﬁcant difference among three websites. Correlation study and ANOVA results together show that users had higher positive experience with higher SUS score sites. Skin Conductance Level (SCL) of Galvanic Skin Response (GSR) was compared using ANOVA at 95% conﬁdence interval. comparative analysis of GSR showed signiﬁcantly higher (f = 16.23, p < 0.05) skin conductance level (SCL) for Thomascook.in (mean = 6.22 µs & std. dev. = 1.32) followed by Yatra.com (mean = 4.32 & std. dev. = 1.63) and (mean = 3.22 µs & std. dev. = 1.12). In light of the results obtained and as discussed in Sect. 4, it is evident that task difﬁculty was higher in thomascook.in followed by Yatra.com and Make My Trip. This paper concludes that, memory based user responses may not be reliable measures of affective experiences of users. Physiological tools like eye tracker and GSR can provide unbiased and useful moment based user’s affective data and can be the future instruments of usability testing wherever affective states of users need to be studied.

370

J. Kumar and J. Kumar Table 2. Comparative study of eye tracker measures for all three websites

References 1. Hudlicka, E.: To feel or not to feel: the role of affect in human–computer interaction. Int. J. Hum.Comput. Stud. 59(1–2), 1–32 (2003) 2. Conati, C.: Probabilistic assessment of user’s emotions in educational games. Appl. Artif. Intell. 16, 555–575 (2002) 3. Norman, D.: Emotion & design: attractive things work better. Interactions 9, 36–42 (2002) 4. Duchowski, A.: Eye Tracking Methodology: Theory and Practice, vol. 373. Springer, London (2007) 5. Goldberg, J.H., Kotval, X.P.: Computer interface evaluation using eye movements: methods and constructs. Int. J. Ind. Ergon. 24(6), 631–645 (1999) 6. Ehmke, C., Wilson, S.: Identifying web usability problems from eye-tracking data. In: HCI 2007, pp. 119–128 (2007) 7. Fiedler, S., Glockner, A.: The dynamics of decision making in risky choice: an eye-tracking analysis: (519682015-072). PsycEXTRA Dataset. APA (2012) 8. Liversedge, S.P., Findlay, J.M.: Saccadic eye movements and cognition. Trends Cogn. Sci. 4(1), 6–14 (2000) 9. Goldberg, J.H., Stimson, M.J., Lewenstein, M., Scott, N., Wichansky, A.M.: Eye tracking in web search tasks: design implications. In: Proceedings of the Symposium on Eye Tracking Research & Applications, ETRA 2002, pp. 51–58 (2002) 10. Phillips, M.H., Edelman, J.A.: The dependence of visual scanning performance on search direction and difﬁculty. Vis. Res. 48(21), 2184–2192 (2008) 11. Boehm-Davis, D.A., Gray, W.D., Schoelles, M.J.: The eye blink as a physiological indicator of cognitive workload. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 44, no. 33, July 2000 12. Siegle, G.J., Ichikawa, N., Steinhauer, S.: Blink before and after you think: blinks occur prior to and following cognitive load indexed by pupillary responses. Psychophysiology 45(5), 679–687 (2008) 13. Jacob, R.J.K., Karn, K.S.: Eye tracking in human-computer interaction and usability research. In: The Mind’s Eye, pp. 573–605. Elsevier (2003)

Investigation of Usability Issues Through Physiological Tools

371

14. Nourbakhsh, N., Wang, Y., Chen, F., Calvo, R.A.: Using galvanic skin response for cognitive load measurement in arithmetic and reading tasks. In: Proceedings of the 24th Australian Computer-Human Interaction Conference on, OzCHI 2012, pp. 420–423 (2012) 15. Lykken, D.T., Venables, P.H.: Direct measurement of skin conductance: a proposal for standardization. Psychophysiology 8(5), 656–672 (1971) 16. Mehler, B., Reimer, B., Coughlin, J.F., Dusek, J.A.: Impact of incremental increases in cognitive workload on physiological arousal and performance in young adult drivers. Transp. Res. Rec. J. Transp. Res. Board 2138(1), 6–12 (2009) 17. Kahneman, D.: Experienced utility and objective happiness: a moment-based approach. In: Choices, Values and Frames, pp. 673–692 (2000)

IHSED 1: Human-Machine Collaboration

The Human-Tech Matrix: A Socio-Technical Approach to Evaluation of Automated Transport Systems Jonas Andersson1(&), Tor Skoglund2, and Niklas Strand3 1

RISE Viktoria, Research Institutes of Sweden, Gothenburg, Sweden [email protected] 2 Sweco Society AB, Gothenburg, Sweden [email protected] 3 VTI, the Swedish National Road and Transport Research Institute, Gothenburg, Sweden [email protected]

Abstract. An automated transport system has the potential to improve trafﬁc safety and flow, but also to the accessibility and comfort for users of the transport system. Realizing the positive effects with automated transport is about shaping regulatory, organizational and technical systems. Here, appropriate evaluation enables steering efforts in the desired direction. The overall aim of this study was to develop a methodological framework that could identify effects of an automated transport system, and outline methods and metrics for evaluation of these effects. We propose a tentative case-based methodology to deﬁne measures of the effects of an automated transport system that will give key stakeholders new possibilities to evaluate research and development projects and efforts connected to automation of the transport system, and thereby manage these in a human-centered direction. Keywords: Human factors

Automated transport systems Evaluation

1 Introduction Automation in the transport domain is argued to have the potential to improve safety and trafﬁc flow as well as decrease the environmental impact [1]. Automation has a great potential to positively influence the transport system, but there are still uncertainties about how society will be affected and what investments and commitments will create the best outcome. Evaluation methods that can merge several perspectives, and transform these into metrics, are needed in order to design efforts in a way that will bring about largest possible net beneﬁts for society at large. Several evaluation frameworks targeting different aspects of evaluation of ITS and automation have been proposed over the years, e.g. [2–7]. As automation affects society as a whole, a combination of technical, social and behavioral perspectives is needed to make more holistic evaluations. The purpose of this study was to develop a methodology for evaluating transport automation from a variety of perspectives since the realization of © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 375–380, 2019. https://doi.org/10.1007/978-3-030-02053-8_57

376

J. Andersson et al.

the desired effects with automated transport systems is about designing regulatory, organizational and technical systems. In our research, an established scientiﬁc framework for the analysis of sociotechnical systems called “The Human-Tech Ladder” [8], was used as a starting point. Vicente (ibid.) describes a sociotechnical system in terms of “hard” and “soft” technology, where hard technology can be compared to traditional technology, while soft technology is man-made structures or processes (e.g. organizational processes, political structures, etc.). Successful implementation of new technologies (hard and soft) requires that human needs are met on multiple levels in the context of their use. The framework allowed for a description of societal needs (such as transport) in ﬁve system levels: physical, psychological, team, organizational and political level. Further, methods and metrics can be directly linked to the analysis. In the research work leading up to this conference publication we have conducted a series of workshops using several examples of automated transport solutions (such as e.g. platooning and low speed distribution pods). The work suggests a path forward in terms of addressing the methodological needs in order to successfully evaluate the effect of an automated transport system in a systematic way.

2 Aim The aim of the research was to systematically identify effects of transport automation and organize these in a structured manner in order to formulate methodological needs for the evaluation of these effects, and where possible: to map out existing methods suitable for evaluation.

3 Description of the Human-Tech Matrix and Procedure Using a socio-technical systems approach as a basis made it possible to identify potential challenges and changes (and relate them to relevant metrics). The methodology consists of a step-by-step procedure using a Human-Tech Matrix to organize the output. The Human-Tech Matrix is a further development of the Human-Tech Ladder as proposed by Vicente [8], where the ladder’s levels (physical, psychological, team, organizational and political) are made into a matrix (Table 2). Also, columns for influencing factors, metrics and methods were added. Using the methodology on a case requires delimitation, a system description, identiﬁcation of methods and metrics, and validation. These steps are further detailed in Table 1. In practice, the “Human-Tech Matrix” is used to map out a particular chain of relations, to be described in the following columns in the same row as the influencing factors are described, see examples in Table 2. In the example, the arrows describe the relationship between the ﬁrst influencing factor (1) and the effects that are, or are expected, to occur (2, 3, 4). Each factor and effect are placed at an appropriate system level depending on categorization. The relationships can look different depending on the effects that can be identiﬁed, e.g. an effect can lead to several other effects and vice versa. Then measurements and methods can be listed for the respective factor and effect

The Human-Tech Matrix

377

Table 1. A step-by-step description of the procedure. Step 1–3: Delimitation 1 Identiﬁcation of case: The area of interest is identiﬁed 2 Data collection: Information on the case is collected 3 Context based problem formulation: General challenges are described considering transport policy objectives and the problem that the automated technology represented in the case is targeting Step 4–5: System description 4 Identiﬁcation of influential factors: Potential factors are identiﬁed using the collected material 5 Hypothesizing plausible effects of the factors: Possible effects are sequenced and mapped in connection to the challenges (step 3) using the human-tech matrix. Continue until saturation is reached Step 6–7: Identiﬁcation of methods 6 Identiﬁcation of potential metrics and methods: Each factor from the system description is connected to appropriate metrics and how to apply them (methods) if possible 7 Identiﬁcation of potential gaps: If no metrics and/or methods can be identiﬁed, these gaps are documented to enable development of relevant methods and metrics Step 8–9: Validation and evaluation 8 Evaluation of the methods used: Each method identiﬁed and included in the matrix can also be assessed regarding its readiness level. A scale based on the technology readiness level (TRL) adapted into methodology readiness level (MRL) is used. This provides the opportunity to judge the degree to which a method is tested and validated 9 Validation of results: Discussion of methodology and preliminary result with external experts in a workshop

(A–F). Where no methods or measurements can be identiﬁed, this is indicated with a question mark. Each method and measure’s “readiness level” can then be judged and highlighted in the matrix (green = authentic, yellow = tested but not authentic, red = not recognized, blue = no method/measure could be identiﬁed).

4 Example: Using the Human-Tech Matrix to Assess the Socio-Technical Effects of Truck Platooning In this example we use truck platooning to exemplify our analysis approach. Due to space limitations the steps in the procedure in Table 1 cannot be described in full detail. Here we focus mainly on the system description using the Human-Tech Matrix. Two influencing factors were chosen for illustration: (a) how truck platooning presupposes that current legislations of permitted time gaps between vehicles in trafﬁc are changed, and (b) how driver work will shift from being an individual task to a task that needs group coordination. Table 3 gives an illustrative example of how these two influencing factors can propagate across socio-technical levels and examples of what methods and metrics that were found in the analysis. It should be noted that Table 3 represents a simpliﬁed example to illustrate the approach. Analyses can be expanded as necessary in spreadsheets depending on the scope and limitations deﬁned in Step 1–3 in the procedure.

378

J. Andersson et al.

Table 2. Human-Tech Matrix with generic examples of how relationships can be described and linked to metrics and methods.

5 Validation of the Human-Tech Matrix Parts of the method described here have been used in several urban development workshops in Sweden, including more than hundred experts from a variety of ﬁelds including academia, transport OEM representatives as well as senior urban and transport planners. The results show that the method facilitates creating a useful basis for understanding effects and challenges. The method has also proven useful in avoiding groupthink and has generated a wide variety of high quality results in very limited time periods (a couple of hours). The use of the method did however require some guidance during the process as it could be perceived as somewhat complex to grasp at ﬁrst sight after only a short introduction. However, in all workshops the participants successfully performed valuable analyses after some guidance from workshop leaders.

6 Discussion Through our platooning example we show how the “Human-Tech Matrix” can contribute to increased possibilities of elucidating whether future research and societal efforts will render expected effects. The matrix representation was found useful for mapping influencing factors, effects, feedback/feedforward loops, and associated methods and metrics. A challenge and limitation of the approach is evidently how to ﬁt the complexity of society into a couple of squares. However, a ﬁnding from the validation workshops with urban planning experts is that the Human-Tech Matrix help people break their old paths of thinking and adopt new perspectives. We believe that this to be an important piece of the puzzle to guide automated transports into a sustainable future.

Table 3. Two examples of how to use the Human-Tech Matrix. Please note that Organizational, Group and Physical levels have been omitted in the example.

The Human-Tech Matrix 379

380

J. Andersson et al.

Acknowledgments. The project was partly funded by Sweden’s innovation agency Vinnova, The Swedish research council for sustainable development and The Swedish Energy Agency, through the strategic innovation program Drive Sweden. We would also like to address a special thank you to Dr. Annika Larsson (Autoliv) and Dr. Johan Strandroth (the Swedish Transport Agency) who made substantial contributions to the presented work during the course of the project.

References 1. Brookhuis, K.A., de Waard, D., Janssen, W.H.: Behavioural impacts of Advanced Driver Assistance Systems – an overview. Eur. J. Transp. Infrastruct. Res. 1(3), 245–253 (2001) 2. Rydmell, C.: PLUTO Evaluation Planning. Technical report 2002:60E, Vägverket, Sweden (2002) 3. Wilmink, I., Janssen, W., Jonkers, E., Malone, K., et al.: Impact assessment of intelligent vehicle safety systems. Technical report eIMPACT Deliverable D4, version 2.0 (2008) 4. Brook-Carter, N., Parkes, A., Ernst, A.C., Jaspers, I., et al.: Development of multiparameter criteria and a common impact assessment methodology. ADVISORS Deliverable D4.1 (2001) 5. Kulmala, R., Luoma, J., Lähesmaa, J., Pajunen-Muhonen, H., et al.: Guidelines for the evaluation of ITS projects. Technical report FITS Publications 4/2002 (2002) 6. Zhang, X., Kompfner, P., White, C., Sexton, B., et al.: Guidebook for assessment of transport telematics applications: updated version. Technical report CONVERGE TR 1101 D2.3.1 (1998) 7. Grane, C.: Assessment selection in human-automation interaction studies. The FailureGAM2E and review of assessment methods for highly automated driving. Appl. Ergon. 66, 182–192 (2018) 8. Vicente, K.: The human factor: revolutionizing the way we live with technology (2010)

Designing and Management of Intelligent, Autonomous Environment (IAE): The Research Framework Edmund Pawlowski(&), Krystian Pawlowski, Jowita Trzcielinska, and Stefan Trzcielinski Faculty of Management Engineering, Poznan University of Technology, Strzelecka 11, 60-965 Poznan, Poland {edmund.pawlowski,krystian.pawlowski, jowita.trzcielinska, stefan.trzcielinski}@put.poznan.pl

Abstract. This paper focuses on designing, building, and developing of Intelligent Autonomous Environment (IAE). IAE is a system of integrated industrial buildings, or a residential district, or a shopping district, etc., deﬁned as an intelligent environment that has all systems of self-steering and adaptation. Adaptation concerns reaction on changes both in external and internal environment. This paper is a part of a larger research project called “Innovative Management System of Designing, Building, Maintenance and Development of IAE”. The paper presents the research framework of the project. The research model assumes: 1. The management system of IAE covers the whole cycle of IAE life (designing, building, maintenance and development), 2. Model of IAE management system includes ﬁve reference models: 1/Functional, 2/Organizational Structure, 3/Processes and Standards of IAE, 4/Methods and Technics of Management, 5/Computer Systems, 3. The management system of IAE is based on Agile methodology with the aim to ensure the agility of the system in all phases of cycle life of IAE. Keywords: Intelligent autonomous system Management

Life cycle Designing

1 Introduction This paper is a part of a larger research project called “Innovative Management System of Designing, Building, Maintenance and Development of IAE”. The project, undertaken at the construction enterprise WPIP Ltd. (Wielkopolskie Przedsiebiorstwo Inzynierii Przemyslowej) and Faculty of Engineering Management of Poznan University of Technology just started in 2018. The aim of the paper is to present the research framework of the project. Originally the project was limited only to building of the system to control the system of zero emission buildings. WPiP Ltd had previously mastered the technology of designing and building of the zero emission intelligent building, controlled by Building Management System (BMS). An idea occurred, followed by a need, to create an IT system managing the entire energetic and © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 381–386, 2019. https://doi.org/10.1007/978-3-030-02053-8_58

382

E. Pawlowski et al.

information system for a group of buildings. The next idea is a further expansion of the integration horizon. It’s an idea of holistically comprehended Intelligent Autonomous Environment integrating the criteria related to technology, construction and architecture, energy and environment protection, information, social communication and safety. Holistic approach also pertains to the full life cycle of such Autonomous Environment, by integrating its designing, building, exploitation and development. Intelligent Autonomous Environment (IAE) is a system of integrated industrial buildings, or a residential district, or a shopping district, etc., deﬁned as an intelligent environment that has all systems of self-steering and adaptation. Adaptation concerns reaction on changes both in external and internal environment, as well as ability to keep up with the changing technology and social progress. IAE combines Smart Economy, Smart People, Smart Governance, Smart Mobility, Smart Environment, and Smart Living. Intelligent Autonomous Environment is a systemically understand environment for people life, both private and social, and business life. Such approach, associating all these criteria and combining them together, creates a totally news perception of the built structures, in which buildings are only a part of IAE. Such holistic view of the building life cycle helps in generating a new offer, which essence is a full systemic functionality for sustainable buildings and their groups. A Management System (MS) for designing, building, exploitation and development of IAE is to offer an intelligent, consistent and holistic design, comprehensive implementation, management of exploitation and development. It mean taking over the responsibility for all phases of IAE life cycle, including the daily maintenance and keeping pace with technological changes, such as further investments and modernization of buildings. Such extensive scale of scope and time of the IAE undertaking creates a massive organizational and managerial challenge. The essence of the project is creating a Management System of the full IAE cycle. The expected result for the project is a management model based on current, modern trend in science and practice of management. Particular elements of the model will be tested practically in an organizational structure of WPiP Ltd and in a number of carried out projects.

2 Model of Integrated Management System of IAE Model of IAE management system (Fig. 1) includes: 1/Reference Functional Model of IAE (Function Tree), 2/Reference Model of Organizational Structure of IAE MS, 3/Reference Model of Processes and Standards of IAE MS, 4/Reference Methods and Technics of Management supporting designers and managers, 5/Computer Systems Support for designers and managers. Reference Functional Model of IAE MS – contains a list of business processes unfolded in form of a function tree diagram. Methodology of function tree unfolding is based on the following assumptions: – Complete unfolding means that the list of functions on the given level of unfolding contains all functions necessary to reach the superior function;

Designing and Management of Intelligent, Autonomous Environment (IAE)

383

MODEL OF INTEGRATED MANAGEMENT SYSTEM OF IAE ORGANIZATIONAL STRUCTURE

List of Business Procesess (Function Tree)

Organizational Charts

AGILE MANAGEMENT

PROCESS MANAGEMENT

Function Deplyment Table

Prcess Description Cards

Business Procesess Maps

Job Description

List of the Process Owners

Work Flow

MANAGEMENT METHODS & TECHNICS

Quality Function Deplyment

Concurrent Engineering

FMEA

Methods of Strategic Analysis

COMPUTER SYSTEMS SUPPORTING MANAGEMENT SYSTEM BALANCED SCORECARD

Fig. 1. Conceptual model of management of intelligent autonomous environment

– Separate unfolding – function occurs only once in the function tree; it means avoiding duplication of the same processes in different functional areas of the management system – Complementary unfolding means showing the connection of the same function with other functional subsystems. Implementing such relations will partially transform the tree diagram into network diagram. – Conditional unfolding – function tree might be subject to changes in parallel with changes in internal and external conditions of IAE. Considering particular phases of the IAE life cycle, the number of functions in the reference tree model will contain approximately 1000–1200 functions. Reference Model of Organizational Structure of IAE MS contains organizational charts, tables of functions deployment to organizational units and managerial positions, job description cards as well as processes’ owners and leaders [1]. From the perspective of a structural solution, on a global level the IAE management system will most likely assume of form of a network/virtual structure, however the organizational structures in particular phases of the IAE life cycle may become various combinations of divisional, matrix, task or network structures. Reference Model of Processes and Standards of IAE MS includes the classiﬁers of processes developed based on the IAE MS function tree, maps of processes, and standards of processes’ realization included in the process description cards and manuals.

384

E. Pawlowski et al.

Reference Methods and Technics of Management Supporting Designers and Managers. Alist of management methods and techniques will differ between the phases of IAE life cycle. Each method and technique should be adjusted to particular needs of management in each phase of IAE lify cycle. Preliminary research already implicates the usefulness of the following methods: Quality Function Deployment (QFD), Concurrent Engineering (CE), Failure Modes and Effects Analysis (FMEA), Methods of Strategic Analysis. Computer Systems Support for Designers and Managers. Designed IT systems are divided into two groups: – Technical support systems (Meta-BMS and other SMART Building management systems) – Support systems for IAE MS in the phase of building design, exploitation and development referring to particular elements of the management system (unfolding of a function tree system, designing and development of the organizational structure, and chosen methods and techniques of management) A reference model of IAE MS consists of the same elements in all phases of IAE life cycle, however the detailed solutions will be individually designed for each phase (Fig. 2). An important methodological and practical problem is moving from one phase to another. It is a change management in an organizational aspect (organization structure, information flow), legal aspect (creating subsidiaries and strategic alliances) in the aspect of human capital management (flow of people and flow of knowledge). The model gets even more complicated when we assume the realization of multiple IAE investments with different levels of complexity, and with different starting date.

PHASES OF THE PROJECT OF IAE MANAGEMENT SYSTEM DESIGNING

Management System of Designing of IAE

PHASE - 1 Methodology of Agility and Metodology of Designinig of Managment System of IAE

Management System of Building of IAE

PHASE - 2 Modeling of Management System of IAE

PHASE - 3 Testing of Subsystem of Management System of IAE

PHASE - 4 Designing of OperaƟonal Management Systems of IAE

PHASE - 5 Designing of Reference Management System of IAE

Supervision Management System of IAE Management System of Maintenance of IAE

Management System of Development of IAE MANAGEMENT SYSTEM OF INTELLIGENT AUTONOMUS ENVIRONMENT

Fig. 2. The frame diagram of the project of IAE management system designing

Designing and Management of Intelligent, Autonomous Environment (IAE)

385

3 Methodological Approach and Framework Diagram of Project Implementation The management system of IAE is based on Agile methodology with the aim to ensure the agility of the system in all phases of cycle life of IAE. The model of an agile enterprise is analysed in four dimensions [2, 3]: 1/. Shrewdness of the enterprise, which is a function assigning to the turbulent environment a string of potential market opportunities. 2/Resource flexibility of the enterprise transforms the string of potential opportunities into a string of resource available opportunities. 3/. Enterprise’s intelligence is an ability to understand situations and ﬁnd deliberate reactions to them, that is to activate proper resource to weaken the threats or use the opportunities. 4/. Smartness of the enterprise is an ability to quickly use the opportunities in a beneﬁt brining manner. Agility should be investigated on four levels: 1/. Reacting for opportunities – identiﬁcation of opportunities by own organizational units is required [4, 5], 2/. Creating opportunity – creative, innovative research and development base is required [6], 3/. Carrying out opportunities using virtual subsystems – it means focus on a score business and using outsourcing, 4/. Carrying out opportunities by using virtual system – it means to become a broker of virtual network [3, 7]. The project is divided into 5 stages Stage 1. Development of the agile methodology and methodology of IAE MS design. The methodology is based on anticipated approach to design of function tree and methodology of agile adjustment organizational structure to the internal and external conditions of IAE MS [1, 8]. Stage 2. Modeling of the IAE MS including – Analysis of external and internal conditions of IAE MS – Development of the IAE MS model – Development of the IT system model Stage 3. Testing of the subsystems of the prototype of the IAE MS in conditions similar to reality. Testing will consist of realization of particular elements of IAE MS in real structure of the WPiP Ltd company and in one selected building investment. Stage 4. Development of operational model of IAE MS. Development of a detailed model of IAE MS considering the results of tests performed in stage 3 as well as testing this model in real conditions in two selected investment projects. Stage 5. Development of a reference model of IAE MS for a full IAE life cycle. This stage consolidates all results of previous stages and prepares a ﬁnal reference version of a IAE MS model.

386

E. Pawlowski et al.

4 Summary and Conclusions Designing and building of smart buildings for a large scale is still on initial phase, but the idea of smart development penetrated public and business consciousness. Investors not always can deﬁne their expectations in such wide aspects, but consciousness and understanding the ideas of IAE still growing. Intelligent Autonomous Environment (IAE) is a system of integrated buildings deﬁned as an intelligent environment that has all systems of self-steering and adaptation. Adaptation concerns reaction on changes both in external and internal environment, as well as ability to keep up with the changing technology and social progress. IAE combines Smart Economy, Smart People, Smart Governance, Smart Mobility, Smart Environment, and Smart Living. The research project called “Innovative Management System of Designing, Building, Maintenance and Development of IAE” is a holistic approach to smart buildings development (designing, building, exploitation and development) and agile approach to management system of IAE. The project is on an initial phase, the ﬁnal result is expected in 2021.

References 1. Pawlowski, E., Pawlowski, K.: A framework of organizational structure designing for agile enterprises. In: Karwowski, W., Salvendy, G. (eds.) Proceedings of 2nd International Conference on Applied Human Factors and Ergonomics, Las Vegas, pp. 1–8 (2008) 2. Trzcielinski, S., Pawlowski, E., Pawlowski, K.: Agile enterprise – some research problems. In: Trends and Ideas in Management, Monograph, Wydawnictwo Politechniki Lodzkiej, pp. 127–136, Lodz (2007) 3. Trzcielinski, S.: Przedsiebiorstwo zwinne, Wydawnictwo Politechniki Poznańskiej, Poznan (2011) 4. Dove, R.: Response Ability. Wiley, New York (2001) 5. Brander, S., Hinkelmann, K., et al.: Mining of agile business processes. In: Artiﬁcoial Intelligence for Business Agility – Papers for AAAI Spring Symposium (2011) 6. Dove, R.: Engineering agile systems: creative-guidance. Frameworks for requirements and design. In: 4th Annual Conference on Systems Engineering Research (CSER), Los Angeles, CA, 7–8 April 2006 7. Trzcieliński, S., Wojtkowski, W.: Toward the measure of organizational virtuality. Hum. Factors Ergon. Manuf. 17, 575–586 (2007) 8. Pawlowski, E.: Organizational structure designing and knowledge based economy. the research framework for the polish enterprises. In: Advances in Human Factors, Ergonomics, and Safety in Manufacturing and Service Industries, pp. 52–61 (2011)

Impact of Machine’s Robotisation on the Activity of an Operator in Picking Tasks Adrian Couvent1(&), Mathieu Dridi2, Nicolas Tricot1, Christophe Debain1, Mahmoud Almasri1, Gil De Sousa1, Gerald Chaloub2, Marie Izaute3, and Fabien Coutarel2 1 IRSTEA, TSCF, Avenue Blaise Pascal. 9, 63170 Aubière, France {adrian.couvent,nicolas.tricot,christophe.debain, mahmoud.almasri,gil.de-sousa}@irstea.fr 2 University Clermont Auvergne, Clermont-Ferrand, France {mathieu.dridi,gerald.chaloub,fabien.coutarel}@uca.fr 3 LAPSCO, Clermont-Ferrand, France [email protected]

Abstract. This paper presents the results of an analysis of the activities during manual and robotised piece picking tasks to know and understand the impact of the robotisation. Here, this task is manually realized ﬁrst and then partially automated with a robot. The activity is described with three indicators. These indicators are computed with image processing and a subdivision of the picking area. The robot has an impact on the activity because its introduction induces a convergence of the subjects to the same activity regardless without of their uses of technologies considering the interaction in this case. Keywords: User-centred systems Usability and user experience

Human-machine systems

1 Introduction The human-machine interaction (HMI) is a key point for a suitable integration. Currently more and more machines need autonomy to work in team with human operators and other robots. Robots can also assist or work with other robots and humans. Therefore, Human-Robot Interaction (HRI) receives considerable attention because humans and robots are increasingly cooperating in many ﬁelds (transport, agriculture, industrial ﬁeld…). Human-computer interaction (HCI) and HRI differ in several dimensions. HRI differs from HCI and HMI because it concerns systems which have complex and dynamic, control systems, show autonomy and cognition, and which operate in dynamic world as in [1, 2].

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 387–393, 2019. https://doi.org/10.1007/978-3-030-02053-8_59

388

A. Couvent et al.

2 Background Regarding the teamwork between robots and humans (seen as agents), it seems important to have an efﬁcient task repartition. It is noted in [3] that human skills are required in robotic systems. The performance approach to evaluate skills level, is used for robot assisted surgery [4] where the learning and the progress of skills are estimated with the performance of the surgeon. In [5], the comparison between manual and assisted work permits to estimate the progress of skills. Another approach is the use of mental models approach like in [6]. Skills estimation is a key to obtain an efﬁcient task repartition. The convergence of automated ﬁelds together with human sciences and the usercentred design described in [7] allows a machine adaptation to the human, the leverage of human skills and capacities, and to answer to the human failure. Many tasks are feasible with human-robot teams. In this situation the robot must understand the operator, as in [8], and adapts its activity. Many interesting surveys have studied the efﬁciency of human-robot teams as in [9]. But in the HRI, the efﬁciency is linked to the potential activities of each agents like described in [3]. Regarding this approach, it is interesting to understand the operator activity evolution without considering the performance. It is important to understand if the robot has an impact on the activity of the operator. To have an efﬁcient HRI and a good acceptability, the robot must not impact the activity but must complete it. The workflow study is important to understand the force and the problem of the HRI as in [10]. It is proved in game theory that the behaviour of the user is influenced by his belief learning [11]. Regarding this, it is interesting to compare if the operator knowledge and technology uses have an influence in the HRI. Considering the previous element with user-centred design in robotic, we hope that it is conceivable to have HRI which do not affect the activity. Regarding this, it is interesting to evaluate if, how and why the robotisation of machines affects the activity of operator.

3 Field Test In order to evaluate if, how and why the robotisation of machines affects the activity of operator, we propose here to compare activity with and without robot thanks to a brief and typical example of picking task. This comparison is achieved after a familiarization phase to get a repeatable and repeated activity. We suppose that this phase is completed when the operators have the same activity to do the same task twice. In this task of picking, the executed strategy is represented by the activity. In addition, another assumption is that the interaction level with the robot is linked with the use of technologies. Considering these elements, an experiment has been developed to evaluate our hypotheses in the case of piece picking. For this purpose, an experimental platform has been designed in laboratory to evaluate the impact of a mobile robot to carry packages.

Impact of Machine’s Robotisation on the Activity of an Operator in Picking Tasks

3.1

389

Experimental Platform and Robot

The experimental platform is composed of four shelves, identiﬁed with letters. On each shelf, there are six packages, identiﬁed with the letter of the shelf and a number according to their position on the shelf. The lateral distance between each shelf is one meter and the shelves measure ﬁve-meters long. To evaluate the impact of the robotisation, a cart is used ﬁrst to establish and end the familiarization phase. The cart and the robot have the same sizes. The robot is the Sherpa by Norcan (depicted in Fig. 1). The robot is able to follow or precede someone thanks to a rangeﬁnder and four other lidars to secure its displacement. The robot interface is composed of two set of four buttons, located at the front and at the rear. Each set is composed of two buttons to choose the mode: follow or precede mode. There is also one button to activate the automated return to the end point, and a red emergency button.

Fig. 1. Sherpa robot by Norcan with its description of element of vision.

3.2

Experiment

The experiment begins with a questionnaire about the use of technologies (smartphones, cars, smart homes and their connections). After an explanation of the task, the operator picks up ten orders of ﬁve packages randomly chosen with the cart. Between each order, the operator makes an audio recorded self-confrontation. After an introduction of the robot, the operator must pick up four packages and ten orders with the robot as in the previous step. During this experiment, all orders are recorded from a camera over the area. Audio recordings are realized for every self-confrontation. Recordings are also synchronized with data of the robot. The panel of subjects is composed of twenty-three voluntary people not familiar with picking tasks to ensure the observation of the phase of familiarization.

4 Deﬁnition and Analysis of Activity To compare the activities of each subject, indicators have been designed. These indicators represent activities of subjects and are independent from the orders. They are used with a program and are based on the subdivision of the picking area depicted in Fig. 3. The ﬁrst indicator describes the independence between the subject and the cart. It means the time rate when the cart is in another area than the subject. In (1), Tinitial is the date at t = 0 s and Tfinal is the ﬁnal date.

390

A. Couvent et al.

P I1 ¼

Timesubject and cart in different area : Tfinal Tinitial

ð1Þ

The second indicator describes the method to pick up packages on shelves. This indicator highlights if the subject picks up packages on one shelf or two shelves at the same time. I2 ¼

Number of area where packages are picked up : Number of shelves with packages which are collected

ð2Þ

The third indicator describes if the subject pushes or pulls the cart/the robot. PNbr transition

I3 ¼

di with d ¼ Number of transition pass by the subject i¼0

1 if the subject push : 0 if the subject pull

ð3Þ

A transition is the passage between two subdivisions. Regarding these three indicators it is possible to know the applied strategy. For example if I1 ¼ 0; I2 ¼ 1 and I3 ¼ 1, the subject pushes and stays with the cart and picks up packages shelf by shelf. Thanks to these indicators, the end of the familiarization phase is recognized when they describe the same strategy twice or more. For example in Fig. 2, the familiarization phase is ﬁnished at the sixth order. After this order, the subject applies the same strategy for every order. The subject leaves his cart ðI1 [ 0:4Þ, picks up package on two shelves in the same area ðI2 \ 1Þ and pushes the cart ðI3 ¼ 1Þ.

Fig. 2. Subject 15 indicators evolution during ten orders.

Using this deﬁnition of the indicators, it is also possible to automate their computation. To do so, a program has been developed to generate automatically the subdivision of the picking area as shown in Fig. 3. This subdivision is based on an image processing of the video recorded during the experiment. The program tracks the subject and detects the cart/robot thanks to theirs color and movement. Then, successive positions are recorded and the indicators are computed at the end of the video. The program gives two results, an image (Fig. 3) and csv ﬁles where indicators and positions are stored. The program computes the three indicators but currently the image processing need to be improved to avoid luminosity problems and false movements detection. To obtain a good description of the strategies chosen by the subjects, the proposed indicators are converted to binaries values. So for I1 , if the

Impact of Machine’s Robotisation on the Activity of an Operator in Picking Tasks

391

Fig. 3. Image result from the program with indicator. Subdivision of the area with blue central areas and red borders area. Blue points are operator position and yellow cart position.

value is superior to 0.15, the operator have the will to leave the cart. Therefore beyond this value, I1 ¼ 1 else I1 ¼ 0. In the case of I2 , if the value is equal to 1, I2 ¼ 1 else I2 ¼ 0. For I3 , if the value is superior to 0.5, the operator mostly pushes so I3 ¼ 1 else I3 ¼ 0. There are eight possible strategies. To quickly identify the strategy, I1 is multiplied by 1, I2 by 10 and I3 by 100. The addition directly highlights the strategy applied by the subject. With this method, the hundred represent the value I3 , the tens the value of I2 and the unities the value of I1 . More than 90% of subjects applied the same strategy at the end of the experiment when using the robot. Nevertheless, it is interesting to compare the results of the questionnaire with the results of the experiment. In this purpose, an indicator, NbComm, is created to estimate the number of collected orders before the end of the familiarization phase. Different variables regarding the use of technologies, smartphones Sm, car Car and domotic Domo, are extracted from the questionnaire. The variable concerning the use of smartphones is composed of four modalities: no smartphones (Sm0), simple use (SmP), classical use (SmM) and intensive use (SmF). The variables Car and Domo are composed of three modalities: no car or domotic (Car0 and Domo0), classical use (CarM and DomoM) and extensive use (CarF and DomoF). Regarding NbComm, this variable is composed of three modalities: need 1,2 or 3 orders to end the familiarization (NbCommP), need more than 3 and less than 6 orders to be familiarized (NbCommM) and need more than 6 orders (NbCommG). Based on these variables and their modalities, a Multiple Correspondence Analysis is achieved with result in Fig. 4. The MCA distinguishes three groups based on the use of the car, smartphones, domotic and the number of orders to be familiarized. Subjects without smartphones, car and domotic rapidly ﬁnish their familiarization (group G0 in red). Subjects, who intensively use each technologic elements (G2), ﬁnish their familiarization between three and six orders. Subjects who classically use technologic elements (G1), ﬁnish their familiarization after six orders. The group (G4 in green) are subjects with an intensive use of technology who rapidly ﬁnish their familiarization. Three patterns of users can be observed here.

392

A. Couvent et al.

Fig. 4. Result of MCA projected on axes 1 (inertia of 21.84%) and 2 (inertia of 20.36%).

5 Conclusion In this paper, the impact of robotisation on the operator’s activity is evaluated. It is clear that the proposed function of the robot (follow mode) influences the strategy of operators. To reduce this, a user-centered design of the function is essential. Another point discussed here is the linked between the use of technology and the interaction level. It is observed that subjects who intensively use technology need more orders to achieve their familiarization than subjects who have a classical use of technology. The robotisation influences operators who choose the easiest way of use. Considering the link between the use of technology and the interaction level, more the subject is comfortable with the use of technology more he will interact with the robot. Acknowledgments. This research was ﬁnanced by the French government IDEX-ISITE initiative 16-IDEX-0001 (CAP 20-25 with the support of the regional council Auvergne-RhôneAlpes and the support with the European Union via the program FEDER).

References 1. Fong, T., Thorpe, C., Baur, C.: Collaboration, dialogue, human-robot interaction. In: Robotics Research, pp. 255–266. Springer, Heidelberg (2003) 2. Goodrich, M.A., Schultz, A.C.: Human-robot Interaction: a survey. Found Trends Hum. Comput. Interact. 1(3), 203–275 (2007) 3. Scholtz, J.: Theory and evaluation of human robot interactions (2003) 4. Chandra, V., et al.: A comparison of laparoscopic and robotic assisted suturing performance by experts and novices. Surgery 147(6), 830–839 (2010) 5. Chang, L., Satava, R.M., Pellegrini, C.A., Sinanan, C.A.: Robotic surgery: identifying the learning curve through objective measurement of skill. Surg. Endosc. Interv. Tech. 17(11), 1744–1748 (2003) 6. Graham, J., Zheng, L., Gonzalez, C.: A cognitive approach to game usability and design: mental model development in novice real-time strategy gamers. Cyberpsychology Behav. 9, 361–366 (2006) 7. Abras, C., Maloney-Krichmar, D., Preece, J.: 1. Introduction and History, p. 14 (2004)

Impact of Machine’s Robotisation on the Activity of an Operator in Picking Tasks

393

8. Nikolaidis, S., Hsu, D., Srinivasa, S.: Human-robot mutual adaptation in collaborative tasks: models and experiments. Int. J. Robot. Res. 36(5–7), 618–634 (2017) 9. Gombolay, M.C., Gutierrez, R.A., Clarke, S.G., Sturla, G.F., Shah, J.A.: Decision-making authority, team efﬁciency and human worker satisfaction in mixed human–robot teams. Auton. Robot. 39(3), 293–312 (2015) 10. Casper, J.L., Murphy, R.R.: Workflow study on human-robot interaction in USAR. In: IEEE International Conference on Robotics and Automation, vol. 2, pp. 1997–2003 (2002) 11. Ioannou, C.A., Romero, J.: A generalized approach to belief learning in repeated games. Games Econ. Behav. 87, 178–203 (2014)

On the Lack of Pragmatic Processing in Artiﬁcial Conversational Agents Baptiste Jacquet1,2(&), Olivier Masson1,2, Frank Jamet1,2,3, and Jean Baratgin1,2,4 1

P-A-R-I-S Association, 25 rue Henri Barbusse, 75005 Paris, France [email protected] 2 Laboratoire CHArt - EA4004 (Université Paris VIII & EPHE), 4-14 rue Ferrus, 75014 Paris, France 3 Université Cergy-Pontoise (UCP), 33, boulevard du Port, 95011 Cergy-Pontoise Cedex, France 4 Institut Jean-Nicod (IJN), Ecole Normale Supérieure (ENS), 29 rue d’Ulm, 75005 Paris, France

Abstract. With the increasing demand for automated agents able to communicate with humans, a lot of progress has been made in the ﬁeld of artiﬁcial intelligence in order to produce conversational agents able to sustain open or topic-restricted conversations. Still, they remain far from the capacity of interaction displayed by humans. This article highlights the challenges still faced in artiﬁcial social interaction regarding the contextualization of utterances within a conversation, either in chatbots or in more complex social robots, through processing of the pragmatic clues of conversations, using current knowledge in psychology and linguistics. It also suggests a number of points of interest for the development of artiﬁcial agents aimed at improving their communication with humans, the relevance of their utterances, and the relationship with the people interacting with them. We believe that in order to be recognized as a social agent, an artiﬁcial agent must follow similar rules humans follow themselves when conversing with each other. Keywords: Pragmatics Natural language processing Conversations Social artiﬁcial agents

Cognition

1 Introduction Artiﬁcial conversational agents, such as chatbots or social robots, are some of the most challenging applications of Artiﬁcial Intelligence (AI), for they rely on both Natural Language Understanding (NLU) and Natural Language Generation (NLG), two ﬁelds current AI programs still struggle with, despite the many breakthroughs of the last decade. While, on the one hand, AI systems have considerably improved in syntax processing and semantic processing, pragmatic processing, on the other hand, remains mostly absent despite being of tremendous importance in human conversations [1]. In this review, we aim to describe the recent improvements in the ﬁelds of NLU and NLG, along with the challenges that are faced regarding pragmatic processing. We also © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 394–399, 2019. https://doi.org/10.1007/978-3-030-02053-8_60

On the Lack of Pragmatic Processing in Artiﬁcial Conversational Agents

395

suggest different ways of investigating the pragmatics of social entities (human or artiﬁcial alike), in order to provide tools to improve the current systems on this speciﬁc aspect. Humans indeed have a keen eye when it comes to differentiating a machine from a human conversational partner, as they will usually use all the clues they can ﬁnd to judge on the humanness of their partner [2–4]. Since pragmatics rely heavily on contextual information, their correct use remains a difﬁcult task for artiﬁcial agents. As their ability to use syntactic and semantic information increases, the main difference between humans and artiﬁcial agents becomes their use of conversational clues, like the tone of a conversation, the mood of their human partner, what they believe their partner should know and what they believe their partner should not know. All of this is even more important with social robots, for in their case, they should be able to understand what their human conversational partner can see or cannot see in a much more explicit way than a chatbot. Indeed, while pragmatic processing is of great importance in written conversations, it becomes absolutely essential in spoken conversations, and involves many more aspects: the modulation of the tone of the voice, gestures, gaze… So essential, in fact, that some elements have even made their way into written conversations, like smileys, which are used to emulate facial expressions, or additional symbols to emulate a modulation of the voice that an exclamation mark or an interrogation mark cannot convey, or even sometimes the voluntary omission of some other symbols, like the period at the end of a message to avoid sounding too abrupt. Recent experiments in the ﬁeld of cognitive psychology indeed highlight the importance of pragmatic processing, in particular regarding the relevance of each statement within conversations between humans, but also between humans and artiﬁcial agents like chatbots and social robots [5]; and the importance of gestures, voice modulation, facial expressions (or emulations of them) and word use for social robots in order to create motivation, involvement, cooperation or competition in the human partner [6, 11].

2 Pragmatics in Chatbots Chatbots were one of the ﬁrst manifestations of AI, and remain in the collective imagination of the public what an AI should look like. The main reason for this is that it is the most visible and easily accessible kind of AI. It has also existed for decades, with the ﬁrst program of this kind being ELIZA [12], which uses keywords within the user’s utterances to reply with open-ended questions in order to keep the conversation going. This rudimentary processing of utterances within a conversation is still at the heart of more complex chatbots, like A.L.I.C.E [13] and Elizabeth [14]. While these AI cannot claim to be actual conversation partners because of their limitations, they can still remain quite useful in some narrow contexts, like in the case of Woebot [15], a chatbot presenting itself as a coach to help users deal with anxiety, depression, procrastination and others. It is indeed not able to participate in open-ended conversations, and the answers the user can give as a reply are in most cases predeﬁned, and in consequence does not involve any text recognition. Tess [16] is also designed to coach people to help them cope with stress, anxiety, depression and burnout (among others) by allowing users to interact in natural language with it. Recently, advances in neural networks has made it possible to develop programs able to generate answers that are

396

B. Jacquet et al.

much more open to any topic of conversation, like Zo [17] which is today one of the most advanced chatbots. Still Zo is far from perfect at keeping track of a conversation. As an example of its interactions, the user can be asking it if it likes the beach, to what it will answer that it hates the beach because of the sand. When the user replied “Yeah sand is annoying” the Zo replied with “Yuo are annoying.”1 showing that the chatbot certainly lost the context of the conversation. It will very often do, in fact, unless the actual context is repeated in almost each utterance. This lack of contextualization is quite frequent in chatbots. While most are pretty good at answering their partner’s utterances, they quickly become unable to answer in a satisfactory way when the user is referring to a previous message, or to the global context of the conversation, without explicitly saying so in their utterance. This detail is far from being anecdotic, since this practice is quite often used in human conversations, for example: A1: Do you have ice cream? B1: I do yes! A2: I’d like some. It is obvious that in A2, the person is referring to the ice cream. Yet many chatbots today would be unable to give an answer to this user, other than a generic one such as “So do I”. This is the main pragmatic issue with today’s chatbots: the relevance of their utterances within the context of the conversation. Indeed, while a generic answer can fool users into thinking that their partner is human, it does not work as well when it happens with almost every utterance. A.L.I.C.E [13] can give the illusion that it understands context for simple questions, like for example when talking about children, and then asking saying “I have two”, the bot will reply with “What are their names?” which is a relevant and expected reply. The problem comes from the fact that it clearly becomes visible that it is not actually a contextualized reply, but indeed a keyword based reply, when the topic switches from children to computers. In this context, when the user says: “I have two”, the chatbot also replies with: “What are their names?” which is certainly not a relevant, expected reply. This lack of contextualization makes current chatbots better at answering series of questions instead of being actual conversational agents, especially in the context of customer service [5], like in the following conversation example: Customer: I am unable to login to my account. I forgot my password. Chatbot: Please click on the ‘‘Forgot Password” link. You will be emailed a link that will help you change your password. Customer: I already tried that. That link takes me to a form to reset password. But it requires me to enter my Customer Relationship Number. I do not have this number. Chatbot: The Customer Relationship Number can be found in the ‘‘Account Information” tab when you log into your account. Customer: But I am unable to login to my account. I do not remember my password. 1

This conversation is from a public message on Twitter, available at https://twitter.com/zochats/status/ 1009141014827761664. It is possible to chat with the bot directly on the same page, through private messages.

On the Lack of Pragmatic Processing in Artiﬁcial Conversational Agents

397

Chatbot: Please click on the ‘‘Forgot Password” link. You will be emailed a link that will help you change your password. Developing the ability of chatbots to use conversation pragmatics thus appears to be one of the most important goal that need to be achieved before such programs can be reliably used as conversational agents, especially for business. Proper testing of the pragmatic competence of chatbots is often difﬁcult and laborious, but poor pragmatic competence can produce feelings of awkwardness for a reader, and if they are not aware of the artiﬁcial aspect of one of the participants in the conversation, might believe that they were mentally ill [2]. A way of approaching the evaluation of chatbots can be to use the framework of the Turing Test, with the addition of recording the response times of the human judge when replying to the evaluated chatbot, as they signiﬁcantly increase following a pragmatic mistake of the chatbot [3, 4].

3 Pragmatics in Social Robots Social robots are certainly going to be more and more common in our daily lives, but more particularly in the lives of elderly people to provide assistance, but also companionship [18, for a review]. Compared to chatbots, the need for the ability to process pragmatic clues, and to generate them, is even greater. For example, being able to infer the user’s mood through the tone of their voice, or their facial expressions. Such clues contribute to the context of the conversation in the same way previous utterances do. In order to provide natural, long lasting interaction with a robot, it is necessary to keep the person motivated by showing them that their interactions matter to the conversation itself, since they will usually quickly lose interest when they notice that the conversation is in fact more like a monologue than an actual dialogue [6]. It should also be taken into account that understanding pragmatic clues is not enough, but they should also be able to generate them if the aim is to let users interact in a natural way. Indeed, neutralizing them in the robot’s behavior produces a barrier against the anthropomorphization of the robot, which is then no longer seen as a social agent, but as a simple machine [7–9]. During face-to-face conversations, the gaze is also a tool for non-verbal communication. Developing a social robot should take into account the different messages a social entity’s gaze can convey to avoid producing undesirable side effects. Indeed, if a user is acknowledged by the robot through its gaze, they will tend to communicate more, have more affection towards it, and also pay more attention to what the robot actually said [10]. Other aspects contribute to the effectiveness of a conversation with a robot, in particular the social norms that are expected of a social entity, like greeting their conversation partner, talking with them in a turn-based conversation, engagement in the sense of getting to know their conversation partner, and ﬁnally a correct handling of the end of conversations involving politeness [6]. When considering facial expressions during a conversation, it is important to understand that they are a tool of communication that is not independent from the

398

B. Jacquet et al.

conversation they belong in. Indeed, understanding the mood a facial expression represents can be difﬁcult for some ambiguous expressions. In order to understand the mood during a conversation, humans use different clues coming from the verbal context and the facial expressions in interaction with each other, in particular since facial expressions during a conversation often act as interjections [19]. Finally, gestures are another important factor in human conversations, in particular for persuasive power. They are again not independent from other clues, such as the gaze. Indeed, [11] has shown that gestures increase the persuasive power of an artiﬁcial agent only when it is combined with correct gaze tracking, and will decrease it otherwise, even though in any cases, gestures give a more lifelike feeling to users.

4 Conclusion We believe that developments in the ﬁeld of artiﬁcial intelligence, applied to natural conversations, should deﬁnitely consider evaluating their agents on their ability to use and generate pragmatic clues whenever their goal is to converse with humans if they want to be recognized as social agents, and not simply as answering machines. This is especially important if they are to be used as a replacement for actual human interaction in the context of business, as they need to provide the same quality of interaction that a customer might expect, but also in the context of care, where bad communication could have consequences for the user’s health.

References 1. Sperber, D., Wilson, D.: Relevance: Communication and Cognition, 2nd edn. Blackwell, Oxford (1995) 2. Saygin, A.P., Cicekli, I.: Pragmatics in human-computer conversations. J. Pragmat. 34(3), 227–258 (2002) 3. Jacquet, B., Jamet, F., Baratgin, J.: Looking for humanity: the influence of inferences on the humanness of a conversational partner in the turing test. In: 3rd Workshop on Virtual Social Interaction (VSI), CITEC, Bielefeld, Germany (2017) 4. Jacquet, B., Baratgin, J., Jamet, F.: The gricean maxims of quantity and of relation in the turing test. In: 11th International Conference on Human System Interaction (HSI), pp. 332– 338 (2018) 5. Chakrabarti, C., Luger, G.F.: Artiﬁcial conversations for customer service chatter bots: architecture, algorithms, and evaluation metrics. Expert Syst. Appl. 42(20), 6878–6897 (2015) 6. Gockley, R., et al.: Designing robots for long-term social interaction. In: International Conference on Intelligent Robots and Systems, pp. 1338–1343 (2005) 7. Masson, O., Baratgin, J., Jamet, F.: NAO robot and the “endowment effect”. In: IEEE International Workshop on Advanced Robotics and its Social Impacts (ARSO), pp. 1–6 (2015) 8. Masson, O., Baratgin, J., Jamet, F.: NAO robot, transmitter of social cues: what impacts?. In: International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, pp. 559–568. Springer, Cham (2017)

On the Lack of Pragmatic Processing in Artiﬁcial Conversational Agents

399

9. Masson, O., Baratgin, J., Jamet, F.: NAO robot as experimenter: social cues emitter and neutralizer to bring new results in experimental psychology. In: IEEE International Conference on Information and Digital Technologies (IDT), pp. 256–264 (2017) 10. Mutlu, B., Shiwa, T., Kanda, T., Ishiguro, H., Hagita, N.: Footing in human-robot conversations: how robots might shape participant roles using gaze cues. In: Proceedings of the 4th ACM/IEEE International Conference on Human Robot Interaction (HRI), pp. 61–68 (2009) 11. Ham, J., Bokhorst, R., Cuijpers, R., van der Pol, D., Cabibihan, J.J.: Making robots persuasive: the influence of combining persuasive strategies (gazing and gestures) by a storytelling robot on its persuasive power. In: International Conference on Social Robotics, pp. 71–83 (2011) 12. Weizenbaum, J.: ELIZA - a computer program for the study of natural language communication between man and machine. Commun. ACM 9, 36–45 (1966) 13. Wallace, R.S.: The anatomy of ALICE. In: Parsing the Turing Test, pp. 181–210 (2009) 14. Elizabeth for Windows. http://www.philocomp.net/ai/elizabeth.htm 15. Woebot Labs Inc. https://woebot.io 16. X2AI Inc. http://x2ai.com/ 17. Microsoft Inc. https://www.zo.ai/ 18. Broekens, J., Heerink, M., Rosendal, H.: Assistive social robots in elderly care: a review. Gerontechnology 8(2), 94–103 (2009) 19. Motley, M.T.: Facial affect and verbal context in conversation: facial expression as interjection. Hum. Commun. Res. 20(1), 3–40 (1993)

Optimal Design of a Robotic Assistant Based on the Structural Study Using Finite Elements Graciela Serpa-Andrade(&), Luis Serpa-Andrade, Vladimir Robles-Bykbaev, and Irene Serpa-Andrade Grupo de Investigación en Inteligencia Artiﬁcial y tecnologías de asistencia GI-IATa, Universidad Politécnica Salesiana sede Cuenca - Ecuador, Cuenca, Ecuador [email protected], {lserpa, vrobles}@ups.edu.ec, [email protected]

Abstract. This project structurally optimizes the robotic assistant that serves as pedagogical support to children with and without disabilities. We began studying of different geometries and materials considering certain aspects tecnic and economic that determine by weighting parameters: the form and possible materials for the design of the new structure. After, be deﬁned all the variables and restrictions affecting it during the therapy, thus establishing the structural model and for its analysis used CAD-CAE computational tools based on ﬁnite elements it can to observe the possible deformations, material, efforts and safety factor to which would be subjected. By last, the results obtained from the prototype and the possible materials for an immediate, medium and long term future are presented that will improve the robot-patient interaction, helping in the visual, sensory and social therapy and in the general learning of the child. Keywords: Robots Special education Children with disabilities Communication disorders Simulation by numerical methods Pedagogical robotics Finite elements Structural design

1 Introduction The UNESCO Chair, Support Technologies for Educational Inclusion and the Research Group on Artiﬁcial Intelligence and Assistance Technologies (GI-IATA) of the Salesian Polytechnic University (UPS), to provide support in the therapy of speak and of the language (SLT) and to the phonoaudiologists (SLP), created the robotic assistant SPELTRA [1–3]. Nevertheless, the robotic assistant SPELTRA has presented failures in its envelope or shell, being necessary to increase the resistance of its structure, without affecting its mechanism. Whit help of programs based on the ﬁnite element method is develops a mathematical model that includes three phases: preprocessing, processing and post processing. To optimize it, is simulated a new a form, material and/or structure that is determined in the preprocessing [4]. After, the results that is obtains is comparators with the results obtained in the simulation the structure old and is presents the prototype and the possible materials for an immediate, medium and long term future assistant. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 400–406, 2019. https://doi.org/10.1007/978-3-030-02053-8_61

Optimal Design of a Robotic Assistant

401

2 Preprocessing The preprocessing phase consists of the recompilation and studies of technical information about the therapies and needs of children with disabilities, therapeutics robots, artiﬁcial intelligence, therapeutics toys, possible shells or form (view Fig. 1) and materials that, meet the requirements to help with the therapies of language and learning. After is analysis and determination the geometry, the material, the loads and the restrictions to that the robotic assistant is subjected and by last, its discretization [5].

Fig. 1. Geometric scheme of: 1. Current structure, 2. Horizontal oval, 3. Vertical oval, 4. Truncated pyramid, 5. Rhomboid, 6. Guitar, 7. Drop and 8. Mixed, [5].

Additional, technical and economic values obtained in the form analysis, a survey was conducted on the shape of the prototype to 40 children from 8 to 12 years old, which provided the children’s preference for the rounded shapes that represent 60% of the total of the survey, predominating the Drop prototype. And, of the 35 materials analyzed technically and economically, we can conclude that the possible materials are ASA-PEEK-SILICONE with a technical value of xi = 0.90% and economic value yi = 0.84%; since they are closer to the point (1, 1) [5].

3 Processing Galileo points out that “the book of nature is written in mathematical language”, conﬁrming the interaction between mathematics and physics [6], that is, we all use mathematics daily. In this case, allowing us to simulate the behavior of the possible materials, whose unknown is the magnitude that characterizes the phenomenon to be simulated numerically [7, 8], facilitating the design and minimizing the experimental study by reducing time, design costs and labor [9, 10]. The models in the science of materials have a very varied mathematical rigor that allows to limit the type of systems according to a given functionality, to understand results and/or to predict new behaviors not yet observed [7]. That is, it is the processing phase where a series of computational processes is carried out that solves the differential equations corresponding to each of the small elements resulting from the discretization taking into account the properties of the materials, geometry, the neighborhood element and the conditions of contour [11, 12]. (View Fig. 2.)

402

G. Serpa-Andrade et al.

Fig. 2. Fuerzas, restricciones y discretización del cuerpo de la estructura actual y del prototipo Gota

Next, Eq. 1 corresponding to the stiffness matrix of a ﬁnite element. ZZZ: ½BT ½D½BdV

½K ¼

O

ð eÞ

Kij ¼ BTi D Bj VðeÞ

ð1Þ

V

Where: BiT is the transposed deformation matrix (3 * 6), D is the constitutive matrix (6 * 6) and B is the deformation matrix (6 * 3) Obtaining by to ﬁnite tetrahedral element a system of 12 12 equations [13]. Finally, we have to be able to recognize whether or not it is an optimal solution or approach us sequentially to an optimum [14]. Consequently, the optimization of the mathematical model seeks three basic aspects: design variables, objective and constraints to form the problem to be simulated, whose process is iterative in which the structure is sought to comply with the resistance to safety manipulation and the least possible cost [15, 16].

4 Post Processing or Results According to the results obtained from the different simulations, conserving the shape only by changing the material, we can point out that the mechanical physical properties of the materials have a considerable influence on their behavior. If the body is will manufactured in PEEK it suffers less deformation, effort and with a minimum safety factor of 6.79. However, the ﬁrst alternative with a similar deformation and stress to the PEEK, but with a minimum safety factor of 3.30 is to print on ASA because in addition to having a considerable safety factor the weight decreases and its cost is lower than the PEEK. Now, the results obtained in the simulations only changing the shape of the quadratic body to the Drop prototype, indicate that: The best alternative to build the body of Drop conserving the extremities and electromechanical components is the silicone with a maximum deformation of 3.15 mm, maximum effort of 4.15 MPa and a safety factor of 15. As a second alternative, we have the ASA material with a maximum deformation similar to the PEEK of 1.76 mm, to have a safety factor of 0.90 because the latter affects only the part bottom of the back in the joint that serves to open and

Optimal Design of a Robotic Assistant

403

close the body of the Drop prototype. Then we have the PEEK, PLA and ﬁnally the ABS. We can also indicate that the ABS material has a better behavior in structures whose shapes are straight remember that the deformations obtained in the simulation of the current shape as Drop body suffers on the sides of the screen whose shape is curvilinear. In addition to changing the shape, decreases its weight by 25.5% with respect to the weight of the old structure that is 3 N to 2,235 N if printed on ASA. And, when printing the body drop shape in PLA has a total weight of 2,331 N, decreasing by 22.3% with respect to the old structure by changing the rounded quadratic form to a truncated drop at the top. Next, the simulation of the Drop prototype is makes with two alternatives (view Fig. 3): 1. Type A: the head and limbs is preserved but the shape, the material the screen, and battery are changed and speaker are increased, the latter will be placed one on each leg. 2. Type B: differs from the previous one in that, in addition to changing the material and internal components, the shape of the head and extremities is modify. • In Drop type A Prototype, the best alternative is: print in PEEK because it would have a maximum deformation of 0.08 mm, a maximum effort similar to 1.39 MPa and a safety factor of 15; which allows us to decrease the volume accordingly its weight. The second alternative could be with SILICONE has deformation of 1.65 mm and with a minimum safety factor to the effort of 15, the same one that indicates that we can decrease in thickness therefore decreases the weight. When printing the robotic assistant in ASA, there are deformations, minor efforts not very signiﬁcant and with a minimum safety factor to the effort similar to 2, so it is a tempting alternative and its cost is lower than SILICONE and PEEK. In addition to being, a thermoplastic that is on the market and its printing is similar to ABS and PLA. As in the previous simulations, the ABS undergoes major deformations with respect to PLA, ASA, PEEK and SILICONE. However, in this case the prototype Drop type A is 2.33 mm almost like that of PLA of 1.93 mm and with similar efforts of 4 MPa meaning that if it is reinforced around the screen and center of the body it is an alternative for the near future.

Fig. 3. Drop type A and type B prototype

404

G. Serpa-Andrade et al.

• The Drop type B prototype with all the materials has maximum deformations around 1, 6 mm and efforts of 4.5 MPa. The best alternative to be printed the robotic assistant is in ASA because it has less weight, easy handling, exists in the market, a safety factor of 11.72 and has a similar price to PLA and ABS. Below is the representation of the results obtained in the simulation of the Drop type A and type B prototype combining the materials: • Drop Type A Prototype In all the combinations we have a minimum safety factor to the effort superior to 2.33 allowing to optimize the volume giving priority to the combination of materials that have less weight and cost. In this case, the combination ASA-PEEK and ASA-ABS-PLA have a weight similar to 2.16 kg, followed by the ASA-SILICONE combination with 2.19 N, reducing the weight by 28%, that is, almost one third of the weight. • Drop Type B Prototype: The best alternative in combination is ASA-PEEK has less deformation of 0.13 mm, with an effort similar to 4 MPa, a minimum safety factor to the effort of 11.72 and a weight of 2.024 N representing two-thirds parts of the current weight of the robotic assistant. That is, has been reduced the 32.5%, representing for 0.976 N less than the old weight.

5 Conclusions For a near future, the Drop type A prototype was simulated with the possible materials ASA, ABS, PLA, PEEK, SILICONE only optimizing its appearance, being the best alternative ASA-PEEK with a deformation of 0.08 mm, minimum safety factor to the effort of 15 and a weight of 2.161 N indicating that the thickness can be decreased. And as a second alternative giving priority to the weight we have the combination of ASAABS with a 2.156 N weight, deformation of 2.33 mm, similar effort to 4.2 MPa and minimum safety factor to the effort of 2.3. For the medium term future, the Drop type B prototype was simulated: a robotic assistant with a caricatured appearance with: bear head, drop body and limbs with joints to facilitate handling, transport, functionality, etc. The best alternative is print it on ASA which decreases the weight by 34% which represents more than a third of the weight, 1,996 N, maximum deformation of 0,13 mm and minimum safety factor to the effort around 12, and can be printed on ﬁlament of 1,75 mm without affecting the safety of the structure. Of all the simulations carried out, the Drop type B prototype according to the results obtained is the one that has less weight, suffers less deformations and stresses with a factor of permissible safety depending on the material, that is, the materials analyzed have a good behavior, is adapt to the geometry of drop type B prototype. In the case of printing in PLA if desired, the arms and hands can be reinforced; and in ABS reinforce the central part of the body For a long-term future, you can take as a basis the study of the materials made and take advantage of their properties. As for example, you can print a graphene battery: that has a longer life, easy to recharge, lighter and less space; lighter servomotors,

Optimal Design of a Robotic Assistant

405

smaller and torque greater. And, in this way would stylize the Drop type B prototype body, decreases volume, weight, improving its appearance and its functionality. According to the annotated, the results obtained in the simulation of the Drop type B prototype in ASA, can be reduced the volume or what is the same, the weight, without affecting the stability and safety of the structure. At the time of possible accidents, the structure would suffer superﬁcial damage, not cracks, much less fractures in the head at the height of the nose or as an additional measure, not necessary, the part can be reinforced Central of the head to decrease the deformation and effort obtained.

References 1. Robles Bykbaev, V., López Nores, M., Ochoa Zambrano, J., García Duque, J., Pazos Arias, J.: SPELTRA: a robotic assistant for speech-and-language therapy. In: LNCS 9177, pp. 525– 534 (2015) 2. Ochoa Guaraca, M., Carpio Moreta, M., Serpa Andrade, L., Robles Bykbaev, V., López Nores, M., García Duque, J.: A robotic assistant to support the development of communication skills of children with disabilities, 24 November 2016 3. Robles Bykbaev, V., Ochoa Guaraca, M., Carpio Moreta, M., Pulla Sánchez, D., Serpa Andrade, L., López Nores, M., García Duque, J.: Robotic assistant for support in speech therapy for children with cerebral palsy, 26 de enero de 2017 4. Rubio Benavides, J.A.: Diseño Y Construccion De Un Robot Interactivo Para El Tratamiento De Personas Con El Trastorno Del Espectro Autista (Tea). Sangolqui: ESPEDepartamento de Ciencias de la Energía y Mecánica, Carrera de Ingeniería Mecatrónica (2016) 5. Serpa-Andrade, G., Robles-Bykbaev, V., Serpa-Andrade, L.: Preprocessing the structural optimization of the SPELTRA robotic assistant by numerical simulation based on ﬁnite elements. In: AHFE 2018, AISC 776, pp. 1–12 (2018). https://doi.org/10.1007/978-3-31994622-1_12 6. Garcia Prada, O.: Matemáticas y Física Teórica, pp. 43–47. www.imdea.org 7. Shackelford, J.F.: Introducción a la ciencia de materiales para ingenieros. PEARSONPrentice Hall, Madrid (2010) 8. de Castro, A.B.: Matematicas e Industria, pp. 13–19 (2007) 9. Dagotto, E.: Complexity in strongly correlated electronic systems 309(5732), 257–262. https://doi.org/10.1126/science.1107559 10. Dennis Jr., J., Schnabel, R.B.: Numerical methods for unconstrained optimization and nonlinear equations. In: Classics in Applied Mathematics, SIAM (1996). eISBN 978-161197-120-0 11. Oñate, E.: Cálculo de Estructuras por el Método de Elementos Finitos, Segunda Edición. Centro Internacional de Métodos Numéricos en Ingeniería, Barcelona - España (1995) 12. Zienkiewicz, F.O.C., Taylor, R.: El Metodo De Elementos Finitos. Formula-cion básica y problemas lineales. Mcgraw-Hill/Interamericana De España, S.A.-CIMNE, BarcelonaEspaña (1994) 13. Shackelford, J.F.: Introduccion a la ciencia de materiales para ingenieros. Prentice-Hall (2010). ISBN 9788483226599

406

G. Serpa-Andrade et al.

14. López Cerdá, M.A.: La optimización y el método cientíﬁco en la toma decisiones, pp. 53–60 15. Martinez Bustamante, M.d.R., Patiño Zarate, D.B.: Simulacion por ordenador median-te el metodo de elementos ﬁnitos y optimizcion por el método de ingenieria robusta, de conectors para la estanteria metálica. Universidad Politécnica Salesiana-UPS, Cuenca- Ecuador (2015) 16. Acosta, L.M.: Ejemplos de Aplicación de la Optimización Estructural mediante Elementos ﬁnitos, vol. 2, p. 3, julio–diciembre de 2011

Human-Autonomous Technology Interaction: A Systemic-Structural Activity Theory Perspective Julian P. Vince1(&) and Gregory Z. Bedny2 1

Defence Science and Technology Group, Department of Defence, 506 Lorimer Street, Fishermans Bend 3207, Australia [email protected] 2 Ergologic, Inc., Louisville, KY, USA

Abstract. This paper proposes that the language of ‘team’ and ‘team-mate’ in the use of autonomous technologies invokes a ‘social technology’ metaphor which arguably confuses the ontological relations between humans and autonomous technologies. Activity Theory, and more speciﬁcally its ergonomic formulation in Systemic-Structural Activity Theory, are argued to offer unique approaches for studying interaction with these emerging technologies. Functional analysis of the model of orienting activity applied to the context of autonomous tool usage is advanced as an approach to understanding the selfregulative mechanisms of the human-autonomous tool system. Keywords: Human-autonomy interaction Human-autonomy teaming Systemic-Structural Activity Theory Human factors

1 Introduction Increasing research effort is being invested in examining how humans can usefully and safely interact with ‘autonomous systems’, ‘machine learning’ and ‘artiﬁcial intelligence’. A signiﬁcant approach to research in this ﬁeld makes use of the metaphor of social interaction [1–3]. From this perspective, human-technology interaction exhibits similar features to that of human interlocutors or actors. Furthermore, this approach makes social psychological concepts such as ‘team’ and ‘trust’ implicit to understanding how humans interact with these technologies. Referring to autonomous technologies as ‘team-mate’ arguably favours a simple, if incomplete, metaphor of anthropomorphised technology over a more complex analysis of the social nature of these tools and their use. This social technology approach arguably marginalises an alternative metaphor proposed by the general form of Activity Theory (AT) [4], and more speciﬁcally Systemic-Structural Activity Theory (SSAT) [5]. General AT grew from the study of psycholinguistic influences on human thinking, and has influenced thought in work psychology and ergonomics. SSAT was developed to better apply AT to the study of human use of tools and technology in work activity. The alternative metaphor presented by SSAT views human-autonomy interaction as an extension of human tool use. In this approach, human-autonomy interaction may © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 407–412, 2019. https://doi.org/10.1007/978-3-030-02053-8_62

408

J. P. Vince and G. Z. Bedny

also be viewed as a technological extension of the mind and body, rather than primarily an interaction with other social actors.

2 Interacting with and through Tools Activity theoretical ideas were arguably part of a broader zeitgeist in scientiﬁc and philosophical thought in the Soviet Union, which focussed on the importance of language as a mediational tool between consciousness and culture, e.g. Vygotsky [6]; as well as the concomitant role of physical tools to the same end, e.g. Rubinstein [7], Leont’ev [4]. Vygotsky proposed that higher cognitive functions are based on semiotic meaning, which is conveyed through the development and use of conceptual tools of signs (as semiotic referents to objects in the world) and then language (symbolic referents). The genesis for the development of the apparatus of meanings is predicated on engagement with the external social world. Rubinstein, on the other hand, emphasised the dual interaction or reciprocity involved in physical interaction with the objective world. For Rubinstein, each human act changes not only the object of the act, but also the subject as well. An individual actively changes the objective world and culture, and thereby changes their self [8]. A parallel example of the broader philosophical resonance in the Soviet Union of the time also includes the work of literary theorist and philosopher, Mikhail Bakhtin [9, 10], whose writings provided a comparable emphasis to that proposed by Vygotsky on transmission of meaning mediated by the use of culturally recognised and refashioned use of language. Bakhtin’s concept of the ‘dialogic’ refers to how language and texts1 are in dialogue with other historical forms of meaning, communication and thought. Words, phrases and texts are products of their historical context of usage while remaining open to the future changing context of their use. Bakhtin’s concept of ‘inner dialogue’ informs the social signiﬁcance of tool use in three chief ways: (1) there is the anticipated or projected social dialogue created through the outcomes of tool use on the object and what this means to others, who in turn respond; (2) there is the dialogue with the tool as a genre of human development, and the use of it which has both historical and immediate dialogical implications in that, through the employment of the tool toward an end, one is engaging inter-subjectively with those involved in the same activity previously; (3) in engaging with the tool there is a form of dialogue which is immediate, in that the human is investing inputs and receiving outputs and adjusts the use of the tool in response to those outputs. Research in humanautonomy ‘teaming’ arguably focusses too narrowly on the last of these forms of dialogue, while largely underappreciating the signiﬁcant influence of the former aspects on the human’s meaning and sense-making with regard to their use of the tool. SSAT provides a multidimensional set of models and methods of analysis which to understand the anthropocentric development and employment of technology. SSAT builds on the activity theoretical premise that the ability to interact with the world

1

The term ‘texts’, for the purposes of the argument in this paper, has been interpreted broadly to include all physical and conceptual tools – not only those pertaining to the literary domain.

Human-Autonomous Technology Interaction

409

conceptually and physically is based on the capacity to develop and use representational systems and physical artefacts in order to satisfy human motives. SSAT does not regard the human as a separate unit to the technological system, but rather views the human as a core organising element of the system – ‘self-regulating’ the humantechnology system towards the realisation of desired human goals [11]. SSAT also proposes an ontological relational structure to human activity. This ontology describes the relationship between subject and object, and subject and subject, and the mediational role of representational and physical tools. The ontological relations that underlie SSAT provide clarity in understanding how human interaction with and through tools aids the evolution of those tools; and thereby the evolution of human creative capacity and culture. The language of ‘team-mate’ and ‘trust’ applied to autonomous tool use arguably mistakes the mediation of subject-object relations in favour of subject-subject relations. Part of what is underemphasised in viewing the use of autonomous tools as subject-subject relations is the reciprocity of modiﬁcation and evolution of both tools and their user through ‘object-oriented’ activity.

3 Interacting with and through ‘Autonomous’ Tools Descriptions in recent human-autonomy ‘teaming’ research of ‘bi-directionality of information flows’ [1] appear to have drawn on cybernetic concepts, however these concepts are not exclusive or unique to an anthropomorphised pair of team-mates – whether they both be human or human and technology. It is arguable that either a socio-technical human-tool based system, or an extended body/mind-technology system metaphor might be just as applicable in understanding the ‘bi-directionality’ of human-autonomy interaction. These alternatives to the human-technology social interaction metaphor may also provide some new perspectives to the study of humanautonomy interaction, that avoid the issue of an inaccurate representation the anthropomorphised technology sets up in terms of operator expectations. SSAT has placed a key emphasis on the role of feedback and feedforward mechanisms in the process of self-regulation, applying this knowledge to ergonomics and the psychology of work. This emphasis is a result of drawing on the work of neuroscientist and physiologist Pyotr Anokhin [12, 13] and psycho-physiologist Nikolai Bernstein [14, 15]. These researchers were well advanced in understanding feedforward/feedback mechanisms of information processing in human and other biological systems well before Western research into cybernetics, such as that of Wiener [16]. Through the influence of Anokhin and Bernstein, SSAT views activity as having a recursive loop structure organised around feedforward/feedback mechanisms which enables the evaluation of performance relative to the desired end. The SSAT approach to understanding human interaction with autonomous technologies suggests that problems of human-autonomy interaction lie with the ability of the human to self-regulate that activity. This seems problematic because many of the mechanisms humans use for self-regulatory processes are at a low level of consciousness or are unconscious. However, these processes involve emotionalmotivational factors important in sense and meaning making. For example, the content of some mechanisms which help regulate the use of a tool towards a desired end

410

J. P. Vince and G. Z. Bedny

will in part draw on personal and/or cultural experiences in order to form the image of the desired end, as well as the means to achieve this through tool use. SSAT provides functional models for examining preferred strategies of task performance which support the study of these processes. For example, Fig. 1 shows the model of selfregulation of orienting activity. The boxes in the model represent function blocks which comprise of integrated cognitive processes which support regulatory functions in the orienting stage of a goal directed activity.

Fig. 1. Model of self-regulation of orienting activity.

The function of orienting activity is to provide a diagnosis about a situation which promotes hypothesis formation about the current and future state of that situation [17]. Orienting activity focuses on identifying the most efﬁcient way of transforming a situation from its current existing condition towards a future desired state. The study of orienting activity may be of particular interest with regards to the use of autonomous tools, due to the reduced opportunity to actively direct the behaviour of such technologies relative to more manual forms of technology. There is also a ‘general model of self-regulation’ [18]. The general model of self-regulation includes further functional blocks associated with the executive components of activity involved in the transformation of a situation and the achievement of the goal of the task.

Human-Autonomous Technology Interaction

411

‘Functional analysis’ is a form of analysis that in SSAT speciﬁcally considers activity as a self-regulative system. The main purpose of functional analysis is to study preferred strategies of task performance. A model of self-regulation such as the model of self-regulation of orienting activity has a number of interdependent windows which may be used to observe human performance. For example, a researcher may focus on a function block called ‘goal’ and study such aspects of activity as goal interpretation, goal signiﬁcance and goal acceptance. The researcher may also focus on the other related function blocks such as ‘assessment of task difﬁculty’ in order to understand motivation with relation to the formation and acceptance of the goal. The interrelation of the function blocks should also be taken into account because the content of linked blocks will influence each other and self-regulative process as a whole. Autonomous system interaction and design research should take advantage of these mechanisms of self-regulation. Technological augmentation of the mind and body through the use of autonomous tools should work in congruence with humans’ evolved mechanisms of self-regulation.

References 1. Chen, J.Y.C., Lakhmani, S.G., Stowers, K., Selkowitz, A.R., Wright, J.L., Barnes, M.: Situational awareness-based transparency and human-autonomy teaming effectiveness. TIES 19, 259–282 (2018) 2. Chen, J.Y., Barnes, M.J.: Agent transparency for human-agent teaming effectiveness. In: IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 1381–1385. IEEE Press, New York (2015) 3. Schaeffer, K.E., Chen, J.Y., Szalma, J.L., Hancock, P.A.: A meta-analysis of factors influencing the development of trust in automation: implications for understanding autonomy in future systems. Hum. Factors 58, 377–400 (2016) 4. Leont’ev, A.N.: Activity, Consciousness and Personality. Prentice Hall, Englewood Cliffs (1978) 5. Bedny, G.Z., Karwowski, W.: A Systemic-Structural Theory of Activity: Applications to Human Performance and Work Design. CRC Press, Boca Raton (2007) 6. Vygotsky, L.S.: Mind in Society: The Development of Higher Order Psychological Processes. Harvard University Press, Cambridge (1978) 7. Rubinstein, S.L.: Problems of General Psychology. Academy of Pedagogical Science, Moscow (1935) 8. Nosulenko, V.N., Barabanshikov, V.A., Brushlinsky, A.V., Rabardel, P.: Man-technology interaction: some of the Russian approaches. TIES 6, 359–383 (2005) 9. Bakhtin, M.M.: The dialogic imagination: four essays by M. M. Bakhtin. In: Holquist, M. (ed.) C. Emerson, M. Holquist (Trans.). University of Texas Press, Austin (1982) 10. Bakhtin, M.M.: Speech genres and other late essays. In: Emerson, C., Holquist, M. (eds.) V.W. McGee (Trans.). University of Texas Press, Austin (1986) 11. Bedny, G.Z.: Application of Systemic-Structural Activity Theory to Design and Training. CRC Press, Boca Raton (2015) 12. Anokhin, P.K.: The Problem of Centre and Periphery in the Psychology of Higher Nervous Activity. Gorky Publishers, Russia (1935) 13. Anokhin, P.K.: Features of the afferent apparatus of the conditioned reflex and their importance in psychology. Probl. Psychol. 6, 16–38 (1955)

412

J. P. Vince and G. Z. Bedny

14. Bernstein, N.A.: The problem of relationship between coordination and localization. Arch. Biol. Science. 38, 1–34 (1935) 15. Bernstein, N.A.: The Psychology of Movement and Activity. Medical Publishers, Moscow (1966) 16. Wiener, N.: Cybernetics: Or Control and Communication in the Animal and the Machine. MIT Press, Cambridge (1948) 17. Bedny, G.Z., Karwowski, W.: A functional model of the human orienting activity. TIES 5, 255–274 (2004) 18. Bedny, G.Z., Bedny, I.: Work Activity Studies Within the Framework of Ergonomic, Psychology and Economics. CRC Press, Boca Raton (2018)

IHSED 1: Design Evaluation, Learning and Assessment

Research Design to Access the Mental Workload of Air Trafﬁc Controllers Thorsten Mühlhausen1(&), Thea Radüntz2, André Tews1, Hejar Gürlük1, and Norbert Fürstenau1 1

German Aerospace Center, Institute of Flight Guidance, Lilienthalplatz 7, 38108 Brunswick, Germany {thorsten.muehlhausen,andre.tews,hejar.guerluek, norbert.fuerstenau}@dlr.de 2 Unit “Mental Health and Cognitive Capacity”, Federal Institute for Occupational Safety and Health, Nöldnerstraße 40-42, 10317 Berlin, Germany [email protected]

Abstract. The German Federal Institute of Occupational Safety and Health in Berlin developed a method for neuronal mental workload monitoring. The socalled Dual Frequency Head Maps (DFHM) method allows deﬁning the workload range of each person individually. The current research project describes the evaluation and condition-related veriﬁcation of the DFHM method in a simulated realistic environment of an air trafﬁc control center. During an interactive realtime simulation at the Air Trafﬁc Validation Center of the German Aerospace Center, the load level for the controllers was varied by means of two independent variables: the trafﬁc demand and the occurrence of a priority request. Dependent variables for registering mental workload were the DFHM index, heart rate, subjective questionnaires, and air trafﬁc performance data. Keywords: Mental workload Air trafﬁc controller

Electroencephalogram (EEG) Simulation

1 Motivation and Goal Present-day work in highly dynamic environments imposes high demands on the cognitive capabilities of employees and is related to negative consequences such as stress-induced diseases. A reliable method to register the mental workload of each employee can help to deﬁne the individual optimal workload range. By working within these boundaries the employee can solve his task most efﬁcient and stay healthy. The German Federal Institute of Occupational Safety and Health in Berlin developed a method for mental workload monitoring by means of the electroencephalogram (EEG). The so-called Dual Frequency Head Maps (DFHM) method was developed and successfully validated in a laboratory setting with 54 test persons [1], i.e. it has reached technology readiness level (TRL) 4 [2]. In order to reach TRL5, the DFHM method has to be validated in a realistic environment. The working position of an air trafﬁc controller (ATCO) was chosen because of two considerations. Firstly, the assigned tasks © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 415–421, 2019. https://doi.org/10.1007/978-3-030-02053-8_63

416

T. Mühlhausen et al.

should be easily modiﬁable to allow for subject’s exposure to different stress situations and thus, induce different workload levels. Secondly, the tasks had to be carried out while sitting in order to minimize artifacts in the EEG. The German Aerospace Center (DLR) was a preferred choice as it provides the requested real-work environment and an optimal suited testbed with the Air Trafﬁc Validation Center. This article introduces the research design, simulation environment and hardware as well as the experimental procedure for the validation of the DFHM method.

2 Research Design The selected environment for the investigation was the radar controller approach position at Dusseldorf airport (EDDL) that was simulated at the Air Trafﬁc Management and Operations Simulator (ATMOS). Boundaries of the simulation were set to one runway available for landings, 5% heavy aircraft, 95% medium aircraft, and an average approach speed of 140kt. Approach controllers were responsible for the aircraft arriving from higher altitudes at the initial approach ﬁx (IAF) until it reached the range of the tower controller on the ﬁnal approach. In Dusseldorf, they took over the control at a distance of roughly 50 nautical miles (nm) from the airport and handed the aircraft over to the tower controller approximately 5 to 10 nm before touchdown. Their main task was the coordination of the arriving aircraft. They gave advisories to the aircraft crews with the aim to guide them safely and efﬁciently to the airport. For variation of the workload, two different parameters were used: trafﬁc intensity and flight with or without priority request. The trafﬁc intensity was varied in the range between 25 and 55 aircraft per hour (ac/h) related to four simulation scenarios. To induce additional workload, we implemented four additional simulation scenarios where one aircraft requested priority due to a sick passenger on board (see Table 1) after the 10th minute of the simulation. Table 1. Simulation scenarios related to the dependent variables (i.e., air trafﬁc volume and priority request). Scenario, Air trafﬁc duration volume 1, 20 min 25 ac/h (low) 2, 25 min 25 ac/h (low) 3, 20 min 35 ac/h (medium) 4, 25 min 35 ac/h (medium)

Priority request No

Description notes The trafﬁc-intensity number reflected a low trafﬁc demand scenario

Yes, after 10 min No The actual coordination value for arrivals at EDDL is regulated by the “FLUKO Flugplankoordination Yes, after Deutschland GmbH” and is set to 33 ac/h [5]. So the value of 35 ac/h reflected roughly this coordination value, 10 min i.e. the maximum number of planned arrivals at EDDL (continued)

Research Design to Access the Mental Workload of Air Trafﬁc Controllers

417

Table 1. (continued) Scenario, Air trafﬁc duration volume 5, 20 min 45 ac/h (high) 6, 25 min 45 ac/h (high) 7, 20 min 55 ac/h (extreme) 8, 25 min 55 ac/h (extreme)

Priority request No

Description notes

With the assumed boundary conditions, the air-trafﬁc value reflected the theoretical maximum number of Yes, after landing aircraft 10 min No This high number of arrivals would lead to an extension of the ﬁnal approach and therefore could only be handled Yes, after during a short period of time 10 min

The recorded data during the simulations included the EEG, heart rate, instantaneous self-assessment (ISA) [3] and NASA-TLX [4] questionnaires, as well as air trafﬁc performance and radio-communication data.

3 Simulation Environment The air trafﬁc management (ATM) simulation environment reproduced a regular ATCO working place at an air trafﬁc control center (see Fig. 1). It represented all typical components such as a radar screen, a weather display, and a voice communication system to talk to pilots. These so-called pseudo pilots sat in an adjacent room. They used the same phraseology as airline pilots when communicating with the controller and entered the given commands in a computer simulation, which generated the trafﬁc and displayed it on the radar screen of the controller. All these functions were activated, controlled, and recorded by a supervisor. For practical reasons, the supervisor was located nearby the controller in order to be able to react immediately on unforeseen occurrences. The EEG was recorded from 25 channels with reference to the Cz electrode (see EEG-channel layout in Fig. 1 and with a sample rate of 500 Hz. We applied a 50 Hz notch ﬁlter and a bandpass ﬁlter between 0.1 and 200 Hz. The heart rate was recorded by means of a pulse plethysmogram and with a bandpass ﬁlter between 0.1 and 30 Hz. We used the g.LADYbird device for EEG and the g.PULSEsensor for heart rate recording. Both devices were manufactured by g.tec and coupled with g.tec’s mobile ampliﬁer g.Nautilus. Logging of simulation-relevant events (see top of Fig. 1) in the bio-signal recordings was done using g.tec’s g.TRIGbox. For bio-signal recording, we used g.tec’s g.Recorder software. The hardware used for bio-signal recording was placed next to the controller position. It included the receiver box for the wireless g. Nautilus device, the recording laptop, and a connection to the ATM simulator via a microcomputer to synchronize and trigger the EEG with the air trafﬁc simulation events.

418

T. Mühlhausen et al.

Made with Fritzing

Fig. 1. Connection pattern between ATM simulation and bio-signal recording via microcomputer (including communication system and event triggering at top left of the ﬁgure). Controller working position with EEG monitoring, EEG-channel layout, and g.TRIGbox are shown at the bottom of the ﬁgure.

4 Trial Procedure All of the investigations acquired were approved by the local review board of the BAuA and the experiments were conducted in accordance with the Declaration of Helsinki. All procedures were carried out with the adequate understanding and written consent of the subjects. The sample consisted of 21 subjects between the ages of 22 and 64 years (2 female, 19 male, mean age 38 ± 11). For getting reliable results, we involved not only DLR internal research controllers but also controllers from the German air navigation service provider (ANSP) DFS and the Austrian ANSP Austrocontrol. The duration of the investigation was set to one day for each subject, starting at noon and lasting until noon of the next day. As the number of scenarios were eight (four trafﬁc intensities with and four without priority request flight for each subject), their duration was set to 20 min for scenarios without priority request and 25 min for scenarios with priority request to make best use of the experimental day. The day started with an introduction about the aims and the experimental protocol of the study, followed by an initial checkup of subject’s current health, mood state, and demographic data (e.g. work experience, qualiﬁcation). Next, the controllers got a brieﬁng and performed a training scenario to get used to the simulation environment. Subjects completed three previously trained cognitive tasks that were presented in counterbalanced order. These were the AOSPAN task for assessing the working memory capacity [6], the Tower of Hanoi as a measure of planning abilities [7], and the 0-back task as an easy baseline task [8]. Furthermore, we conducted rest measurements at the beginning and end of subject’s experimental days. The simulation scenarios started in random order after the cognitive tasks. During each scenario, the controller had to answer the ISA questionnaire every ﬁve minutes

Research Design to Access the Mental Workload of Air Trafﬁc Controllers

419

and rank his perceived workload as (1) under-utilized, (2) relaxed, (3) comfortable, (4) high, or (5) excessive. After each scenario, he also completed the NASA-TLX questionnaire. In parallel, the pseudo pilots rated the controller’s workload in accordance with the ISA categories (see above). After two scenarios there was a small break before starting the last two simulation runs of the day. On the next day, the remaining four scenarios were performed until noon using the same workflow. A detailed schedule is given in Table 2. Table 2. Daily schedule. Day 1 Time Schedule 12:30 Brieﬁng: introduction, questioning of subject’s current health, mood state, and demographic data 13:00 Simulation training

13:20 Training of the cognitive tasks 13:35 Application of electrodes 13:50 Rest measurement, cognitive tasks

Day 2 Time Schedule 9:30 Application of electrodes 10:00 Rest measurement, 2 simulation scenarios 11:05 Break 11.20 2 simulation scenarios 12:25 Rest measurement, end of day 2

14:50 Rest measurement, 2 simulation scenarios 15:55 Break 16:10 2 simulation scenarios 17:20 Rest measurement, end of day 1

5 Observations During the trials, it became evident that the DFMH method could be a major step towards an objective assessment of mental workload. Each controller had a different way to handle the trafﬁc that was possibly related to the different individual experience level. The perception of his own workload according to ISA and NASA-TLX varied strongly and did not always reflect the rating of the pseudo pilots. An objective method would help to identify a more realistic view, additional to these two subjective perceptions. The variation of the trafﬁc volume proofed to be a good instrument for workload manipulation. Nearly all controllers considered the low trafﬁc scenario as under-utilized or relaxed and perceived a workload increase with higher trafﬁc volume. The gradient depended on experience and the resulting operational strategy. The implementation of a flight with a priority request had only limited influence on the perceived workload. Only in high and extreme trafﬁc-load scenarios some

420

T. Mühlhausen et al.

controllers reported increased workload because of the priority request. Hence, the variation between priority and non-priority scenarios was only limited.

6 Summary and Outlook The design used proofed to be suitable for successful workload variation. The induction of different workload levels via trafﬁc volumes worked very well, but the implementation of a flight with a priority request showed only limited effects. Therefore, further research is needed to identify additional parameters that affect workload besides trafﬁc volume. Synchronization between simulation status, events, and brain activity was possible by coupling the air trafﬁc management simulation via a microcomputer with the external EEG system. Currently, analysis of the EEG takes place and ﬁrst results will be published in the near future. We expect that the DFHM-workload index from the EEG will be able to differentiate between the various load conditions that arise from the trafﬁc volumes. However, we assume that it will be only limitedly able to differentiate between load conditions arising from the priority request or from the interaction between trafﬁc volume and priority request. Finally, we expect the DFHM index to show similar tendencies as other workload-relevant parameters from the cardiovascular, subjective, and performance data. The positive validation of the DFHM index based on the proposed design is very promising because an objective method for mental-workload registration anticipates the development of workplaces with appropriate workload. Acknowledgments. We would like to thank Kerstin Ruta for her daily operational support, Emilia Cheladze and Lea Rabe for conducting the experiments and the numerous pseudo pilots for their contribution during the experiments. We would also like to thank Martin Schütte for his general project support. Author Contributions. T.R. initiated the project and was responsible for the overall conception of the investigation. T.R., T.M., A.T., H.G., and N.F. developed the research design of the study. T.M. and A.T. were responsible for the implementation of the simulation scenarios and the technical support. The study was supervised by T.R. The manuscript was written by T.M. and T.R. Final critical editing was performed by A.T., H.G., and N.F.

References 1. Radüntz, T.: Dual frequency head maps: a new method for indexing mental workload continuously during execution of cognitive tasks. Front. Physiol. 8, 1019 (2017). https://doi. org/10.3389/fphys.2017.01019 2. Mankins, J.C.: Technology Readiness Levels – A White Paper. Ofﬁce of Space Access and Technology NASA (2004) 3. Kirwan, B., Evans, A., Donohoe, L., Kilner, A., Lamoureux, T., Atkinson, T., MacKendrick, H.: Human factors in the ATM system design life cycle. In: FAA/Eurocontrol ATM R&D Seminar, Paris, France (1997)

Research Design to Access the Mental Workload of Air Trafﬁc Controllers

421

4. Hart, S.G., Staveland, L.E.: Development of the NASA TLX: results of empirical and theoretical research. In: Hancock, P., Meshkati, N. (eds.) Human Mental Workload, pp. 139– 183. North Holland, Amsterdam (1988) 5. FLUKO Flughafenkoordination Deutschland GmbH Information. http://www.fhkdspeicher.org 6. Unsworth, N., Heitz, R.P., Schrock, J.C., Engle, R.W.: An automated version of the operation span task. Behav. Res. Methods 37, 498–505 (2005) 7. Lezak, M.D., Howieson, D.B., Bigler, E.D., Tranel, D.: Neuropsychological assessment, 5th edn. Oxford University Press, New York (2012) 8. Gazzaniga, M., Ivry, R., Mangun, G.: Cognitive Neuroscience: The Biology of the Mind, 4th edn. W.W. Norton & Company, New York (2013)

Measuring Collaborative Emergent Behavior in Multi-agent Reinforcement Learning Sean L. Barton1(&), Nicholas R. Waytowich2, Erin Zaroukian1, and Derrik E. Asher1 1

Computational and Information Sciences Directorate, U.S. Army Research Laboratory, Adelphi, USA [email protected] 2 Human Research and Engineering Directorate, U.S. Army Research Laboratory, Adelphi, USA

Abstract. Multi-agent reinforcement learning (RL) has important implications for the future of human-agent teaming. We show that improved performance with multi-agent RL is not a guarantee of the collaborative behavior thought to be important for solving multi-agent tasks. To address this, we present a novel approach for quantitatively assessing collaboration in continuous spatial tasks with multi-agent RL. Such a metric is useful for measuring collaboration between computational agents and may serve as a training signal for collaboration in future RL paradigms involving humans. Keywords: Multi-agent reinforcement learning Human-agent teaming Collaboration

Deep reinforcement learning

1 Introduction Reinforcement learning (RL) is an attractive option for providing adaptive behavior in computational agents because of its theoretical generalizability to complex problem spaces [1, 2]. In particular, deep RL recently produced striking results [3]. Extending RL to the multi-agent domain has received an increasing amount of attention as the need for human-agent teams has increased, especially with regards to training agents to behave collaboratively [1, 4–8]. Unfortunately, the nature of multi-agent RL makes guaranteeing collaboration between agents impossible except in limited provable cases, even when these methods yield better task performance [1, 7]. Thus far, evaluating collaboration in multi-agent learning has been accomplished by measuring performance in tasks where coordination is required. While this may be satisfactory for discretized tasks where cooperative policies are provably optimal [1, 6–8] it is not clear that this methodology generalizes well to more complex and continuous tasks (such as those presented in [4, 5] and here). The question at hand is how to assess performance enhancing coordination (or collaboration) between agents in multi-agent RL tasks. Here, we present a method borrowed from the ﬁeld of ecology, called convergence cross mapping (CCM), and show how it can be used to measure collaboration between agents during a predatorprey pursuit task. Additionally, we show a striking result: a state-of-the-art multi-agent © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 422–427, 2019. https://doi.org/10.1007/978-3-030-02053-8_64

Measuring Collaborative Emergent Behavior

423

reinforcement learning algorithm does not exhibit coordinated behavior between agents during a collaborative task even though high task performance is achieved, indicating that performance metrics alone are not sufﬁcient for measuring collaboration.

2 Methods 2.1

Simulation Environment

In a modiﬁcation of the classic predator-prey pursuit task (see Fig. 1A–C), three slower predator agents score points each time they make contact with a prey agent in a continuous bounded 2D particle environment. Predator agents were identical in terms of capabilities (i.e., velocity and acceleration). This simulation environment was made available through the OpenAI Gym network [9] and was developed for the multi-agent deep learning algorithm discussed below [4]. Prey agents were capable of 33% greater acceleration and 25% greater maximum velocity than any predator agent, making capture by a solitary predator extremely difﬁcult. All agents had the same mass, with minimal elastic properties to provide a small bump force upon collisions. Agent positions and velocities were randomized at the start of each episode, and acceleration was initially set to zero. Predator agents all received a ﬁxed reward when any one of them made contact with the prey agent. Prey agents received a punishment when they were contacted.

Fig. 1. Experimental design. (A–C) Yellow circles denote prey, while others represent predator strategies. Black arrows indicate actions. Panels show (A) Chaser, (B) Spring (coupling shown as grey dotted lines), and (C) MADDPG agent conditions. (D) At test time, one predator’s behavior was determined by the physics of a double pendulum (grey circle). Agents coordinated with a modiﬁed predator should have their goal-seeking actions (grey dashed arrow) augmented by the actions of the modiﬁed predator. (E) Schematic of MADDPG algorithm.

424

2.2

S. L. Barton et al.

Agents

In order to evaluate our metric’s ability to estimate collaboration between predators in the predator-prey task, we utilized two types of agents: learning agents and ﬁxedstrategy (non-learning) agents. Learning agents’ behaviors were guided by a multiagent deep deterministic policy gradient (MADDPG) algorithm [4]. In all cases, prey agents were learning agents and thus utilized the MADDPG algorithm independent of predator behavior. Two types of distinct ﬁxed-strategy predators were implemented to demonstrate upper and lower bounds of coordinated behaviors. The ﬁrst (termed ‘Chaser’ predators, see Fig. 1A) naively pursued prey agents by maximally accelerating in the instantaneous direction of the prey relative to the predator’s own position. As such, Chasers were incapable of coordinating their behavior with each other. The second strategy (termed ‘Spring’ predators, see Fig. 1B) also naively minimized distance to the prey, but predator movements were modiﬁed by spring forces which constrained their position and velocity relative to one another. A Spring predator’s movement direction was a sum of the spring forces acting on it and its desired vector of movement towards the prey. In this case, predator actions were explicitly coordinated with their partners’. The MADDPG algorithm (Fig. 1C and E) used to guide learning agent behavior is an extension of a deep deterministic policy gradient (DDPG) algorithm [3] into the multi-agent domain [4]. Like DDPG, MADDPG utilizes an actor-critic model with deep neural networks representing policy and Q-learners (Fig. 1E). Multi-agent capabilities are achieved by passing information about each agent’s state and actions to each critic network. As such, the learning agents are joint action learners as opposed to independent learners, giving them an advantage when coordinating their behaviors [8]. All conditions (Fig. 1A–C) were trained for 100k episodes to ensure model convergence before evaluating collaborative behavior (Fig. 1D). Each episode lasted for 25 time steps during learning and 2000 time steps during evaluation. These intervals were selected to match what was shown in literature for sufﬁcient learning [4] and required for analysis [10]. During testing, 10 episodes were recorded without learning for each predator strategy in order to produce a distribution of agent behaviors with the same level of training over various random starting positions. 2.3

Collaboration Metrics

In order to measure collaboration, we used predators’ positions over time as time-series data to implement a technique called convergent cross mapping (CCM) which examines the causal influence one time-series has on another [10]. The CCM technique embeds a time-series in a high-dimensional attractor space and then uses this embedded data as a model to predict states of the other time-series in its own attractor space. To the extent that this is possible, the original time-series is said to be causally driven by the time-series it attempts to model. Thus, in multi-agent tasks with homogeneous agents, we are deﬁning collaboration to be the amount of causal influence between agents as measured by the CCM.

Measuring Collaborative Emergent Behavior

425

Importantly, this metric can be misleading when two time-series have a mutually causal relationship with a third. For the present task, this is important because the causal influence one predator has on another can be confounded by their mutual relationship with the prey. To address this, at test time we modiﬁed the behavior of one predator agent such that its previous behavior was replaced by a secondary behavior that did not pursue the prey1. If a causal influence exists between predators, the movements of the modiﬁed predator should change the behavior of the other predators, even though the modiﬁed predator no longer pursues the prey (see Fig. 1D).

3 Results Table 1 shows the log scaled mean reward per episode for predator and prey agents in the Chaser, Spring, and MADDG cases. It is clear that MADDPG-equipped predators are betters performers of this pursuit task, achieving roughly an order of magnitude more average reward per trial. Table 1. Performance results for the different experimental conditions N Mean reward SD 95% CI Chaser 10 1.85 0.66 0.47 Spring 10 4.29 0.92 0.66 MADDPG 10 13.70 2.57 1.84

While the capacity for coordinated actions afforded by the MADDPG algorithm coupled with the increased task performance may suggest collaboration between predator agents, the CCM analysis tells a different story (Fig. 2). As shown, there is very little difference, in terms of causal influence, between MADDPG-equipped predators and naive Chasers. The causal influence of the modiﬁed agent might be marginally stronger for MADDPG predators (hinting at the shared action information afforded by this algorithm), but the effect is too weak to make any strong conclusions about the collaboration between the agents in this case. For Spring predators, on the other hand, there is a strong causal influence from the modiﬁed predator. The Spring experimental condition (see Fig. 2) shows an example of coupled behaviors in the predator-prey pursuit task. This provides a strong validation for the CCM technique, and illustrates an ecological upper-bound on coordinated actions between learning agents.

1

In this case, the dynamics of a double pendulum were used to specify the movement of the modiﬁed predator.

426

S. L. Barton et al.

Fig. 2. CCM measure for causal influence of modiﬁed predator behavior on unmodiﬁed predators. True causal influence is indicated by an increase in CCM score (Y axis) as more of a trial (X axis) is used to build the model (so called “convergence” [10]). A clear causal relationship is shown for Spring predators. Minimal causal influence is detectable for MADDPG agents. Coordination for Chaser agents is indistinguishable from zero.

4 Discussion The MADDPG algorithm presented by Lowe et al. [4] is promising, because it suggests that multi-agent deep RL may be able to solve complex problems in continuous tasks involving multiple actors. However, as we demonstrate here the improved performance of the MADDPG algorithm does not necessarily indicate coordinated behavior between agents. Assessing collaboration between agents instead requires a direct measure of the coordination between agent actions. We present such a method here, in the form of CCM, and we validate the metric within the context of a continuous multi-agent task. Though we did not ﬁnd strong evidence of coordination between the actions of learning predators, the marginal increase in coordination over Chaser predators points to potential for collaboration given sufﬁcient learning pressure. Producing collaborative behavior in computational agents is critical for the future of human-agent teams. Human-factors research has long held that when computational systems fail to adapt to human needs, serious issues in performance and human satisfaction can arise [11]. These issues are often best alleviated by promoting a collaborative relationship between humans and computational agents, rather than forcing humans to act as overseers [12, 13]. Measures like CCM, which can be used to assess (and even promote) collaborative behaviors in multi-agent RL, constitute powerful tools for the future of human-computer interactions.

Measuring Collaborative Emergent Behavior

427

Acknowledgements and Disclosure. This research was sponsored by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF-18-2-0058. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the ofﬁcial policies, either expressed or implied, of the Army Research Laboratory or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.

References 1. Matignon, L., Laurent, G.J., Le Fort-Piat, N.: Independent reinforcement learners in cooperative markov games: a survey regarding coordination problems. Knowl. Eng. Rev. 27, 1–31 (2012) 2. Sen, S., Sekaran, M., Hale, J., et al.: Learning to coordinate without sharing information. In: AAAI, pp. 426–431 (1994) 3. Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A.A., Veness, J., Bellemare, M.G., Graves, A., Riedmiller, M., Fidjeland, A.K., Ostrovski, G., Petersen, S., Beattie, C., Sadik, A., Antonoglou, I., King, H., Kumaran, D., Wierstra, D., Legg, S., Hassabis, D.: Human-level control through deep reinforcement learning. Nature 518, 529–533 (2015) 4. Lowe, R., Wu, Y., Tamar, A., Harb, J., Pieter Abbeel, O., Mordatch, I.: Multi-agent actorcritic for mixed cooperative-competitive environments. In: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R. (eds.) Advances in Neural Information Processing Systems, vol. 30, pp. 6382–6393. Curran Associates, Inc. (2017) 5. Foerster, J., Farquhar, G., Afouras, T., Nardelli, N., Whiteson, S.: Counterfactual MultiAgent Policy Gradients. arXiv:1705.08926 [cs] (2017) 6. Matignon, L., Laurent, G., Le Fort-Piat, N.: Hysteretic q-learning: an algorithm for decentralized reinforcement learning in cooperative multi-agent teams. In: IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007, pp. 64–69 (2007) 7. Lauer, M., Riedmiller, M.: An algorithm for distributed reinforcement learning in cooperative multi-agent systems. In: Proceedings of the Seventeenth International Conference on Machine Learning. Citeseer (2000) 8. Claus, C., Boutilier, C.: The dynamics of reinforcement learning in cooperative multiagent systems. In: AAAI/IAAI 1998, pp. 746–752 (1998) 9. Brockman, G., Cheung, V., Pettersson, L., Schneider, J., Schulman, J., Tang, J., Zaremba, W.: OpenAI Gym. arXiv:1606.01540 [cs] (2016) 10. Sugihara, G., May, R., Ye, H., Hsieh, C.-H., Deyle, E., Fogarty, M., Munch, S.: Detecting causality in complex ecosystems. Science 1227079 (2012) 11. Parasuraman, R., Sheriden, T.B., Wickens, C.D.: A model for types and levels of human interaction with automation. IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 30, 286–297 (2000) 12. Rovira, E., McGarry, K., Parasuraman, R.: Effects of imperfect automation on decision making in a simulated command and control task. Hum. Factors 49, 76–87 (2007) 13. Klein, G., Woods, D.D., Bradshaw, J.M., Hoffman, R.R., Feltovich, P.J.: Ten challenges for making automation a “team player” in joint human-agent activity. IEEE Intell. Syst. 19, 91–95 (2004)

Products as Mass Media: Entertainment vs. Ediﬁcation Del Coates(&) San Jose State University, San Jose, CA, USA [email protected]

Abstract. The information inherent in a product’s form—evident as contrast and/or novelty of shapes, colors and any other visible characteristics—communicates meanings to viewers via empathic expression. Indeed, mass-produced products, seen daily by millions in homes, workplaces, stores and streets, constitute more pervasive and arguably more compelling mass media than newspapers, radio, TV and movies. As with other mass media the design of a product either entertains audiences by reinforcing existing mindsets or ediﬁes them by changing them in ways that beneﬁt users and society. Keywords: Information Contrast Novelty Empathic expression Mass media Entertainment Ediﬁcation Stereotype

1 Introduction Products manufactured by the thousands, hundreds of thousands and millions constitute mass media, especially those like automobiles, which are constantly visible to large audiences. They are mass media as surely as newspapers, magazines, radio and TV broadcasts and movies. And like all other mass media, they can entertain and/or educate their audiences. I examine here the process by which they edify or merely entertain their audiences and suggest some ways designers might nudge a product’s audience in more culturally beneﬁcial directions. Most people would not ﬁnd it a reach to consider certain products to be mass media. For example, a product might display written messages or pointers, or it might use chimes or buzzers or even a quasi-human voice. But the argument here is broader. Speciﬁcally, all products “speak” to viewers through all their visual aspects and, moreover, they answer any implicit question a viewer asks them. These assertions are supported by information theory. As originally deﬁned, “to inform” meant “to shape.” Clearly, people shape all products. Manufacturing processes literally inform metal, plastic and other materials to realize the products their designers imagine.

This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 428–432, 2019. https://doi.org/10.1007/978-3-030-02053-8_65

Products as Mass Media: Entertainment vs. Ediﬁcation

429

2 Empathic Expression The information thus embodied in a product’s design “speaks” to viewers through every visual aspect of its “body,” including shapes, colors, and textures. It does so by means of “empathic expression” related to the “body language” of expressions, gestures, and postures we come to understand from infancy. The upshot is that we perceive feelings as occurring within a product, which seems lifelike enough to have its own feelings and will effectively answer any question implicitly put to it. You might wonder which of the watches in Fig. 1 is heaviest, or most accurate, or most expensive—or anything else that comes to mind. One inevitably answers, “I am!”

Fig. 1. Different designs say different things.

The interpretation of empathic expression is simpler than it might seem because crucial principles of information theory are rather straightforward. The acknowledged father of modern information theory, Shannon [1], concluded that information always entails some degree of variance. And Berlyne [2] determined, in his psychological aesthetics research, that only two basic kinds of variance matter, both of which can be measured and tuned: • Contrast (which I call spatial information) is a perceived difference between simultaneously perceived stimulus properties (like black printing on white paper); • Novelty (which I call temporal information) is a difference between a previously formed expectation and present reality. For example, just as a newspaper has more information on ﬁrst reading than on second reading, an unusually shaped car has more information than conventional models.

3 Interpreting Empathic Expression As information, contrast and novelty draw attention to a product, making it interesting and exciting. A design can have too much contrast or novelty or too little. Contrast doesn’t normally change over time. A crease along a car’s side, for example, which constitutes contrast due to the varying orientation of the sheet metal on either side of

430

D. Coates

the crease, doesn’t change. Novelty, however, always changes over time. A design that depends chiefly on novelty for its appeal will lose appeal over time. Novelty typically grabs more attention and stirs more emotions than contrast because it conforms less to expectations, which are determined by a mental model called a stereotype. A new family sedan will seem normal, for example, if it conforms closely to a stereotype representing a blend or average of every family sedan seen before. If it does not, it won’t make sense to the viewer because it is so unusual that it doesn’t resemble its associated stereotype closely enough. It is by deﬁnition too novel. For example, the second generation (G2) Toyota Prius of 2004 at the top of Fig. 2 has a simple shape, typical of low-contrast design. But its overall form, designed to slip through the air with minimal resistance, departed too much from the stereotypical family sedan at the time. It was too novel—and still is for many car buyers, even today.

Fig. 2. Semantic proﬁles of G2 Prius (at top) and G3 Prius compared.

Products as Mass Media: Entertainment vs. Ediﬁcation

431

So designers of its 2010 (G3) successor de-emphasized that novelty. To retain interest and excitement they increased contrast by squaring its corners somewhat, adding a prominent crease along the side and surrounding the headlights with a complex zig-zag pattern instead of the much simpler shape of the 2004 model’s headlights. The 5-spoke wheels of the G3 are also more exciting than the 6-spoke wheels of the G2. The graph below the photo suggests the empathic expressions of the cars. I created semantic proﬁles using the semantic differential survey methodology developed by Osgood [3] that I have used in my research [4]. Subjects marked seven-valued scales anchored by adjectives of opposite meanings (complex-simple, heavy-light, fast-slow, etc.). The vertical ordering of the scales and their left-right orientation were randomized, of course, to mitigate bias effects. Normally, I would have conducted a separate survey for each car. This time I had participants score cars on the same form because I was more interested in how different the cars seemed. Notice the G3 scores on every scale to the right of the G2, where the more active and potent terms appear. This indicates that designers achieved their objective of a car that seems more interesting, exciting and robust than the previous model.

4 Entertainment or Ediﬁcation? When the oil crises of the 1970s led to escalating gasoline prices and long lines at gas stations General Motors decided to create so-called compacts with smaller frontal areas (the car’s proﬁle as seen head-on). These smaller cars would consume less fuel by virtue of less weight and smaller frontal areas yield less aerodynamic drag. A car’s coefﬁcients of aerodynamic drag (CD is calculated by comparing the car’s aerodynamic drag in a wind tunnel to that of a disk equal in area to the car’s frontal area. If they were equal the car would be assigned a CD of 1.0, the worst possible outcome. The typical American car at the time had a CD between 0.55 and 0.6. Chevrolet’s all-new compact, the 1980 Citation shown in the background of Fig. 3, had a commendably better CD of just 0.42. But the “all new” Citation didn’t actually look very new. It corresponded closely to the American automotive stereotype by adhering to the “longer, lower, wider” idiom. The nearly flat nose spanned the full width of the car and featured lots of chrome grillework. Had the nose been rounded to decrease its CD even further the rounded nose would have made it look unfashionably narrow. It otherwise looked quite ordinary for its day except for its “fastback” tail. Shortly after the Citation’s introduction I wrote a magazine article showing how it could have looked newer and more exciting with a shape that was more aerodynamically efﬁcient. I designed the hypothetical Citation using a very accurate method developed by England’s Motor Industry Research Association (MIRA) employed by automotive designers and engineers worldwide (I had used it while working in Ford’s Advanced Vehicle Concepts Dept. during the 1960s). Following MIRA’s methodology I retained the Citation’s desirable fastback form, smoothed and rounded the nose, gave the windshield more curvature and put skirts over the rear wheels. The results yielded the design shown in Fig. 3 with a predicted CD of just 0.27, good for an estimated 10-percent improvement in fuel economy.

432

D. Coates

Fig. 3. 1980 Chevrolet Citation behind a Hypothetical Aero Citation concept

So why did we have to wait a quarter century for cars so efﬁcient when, by today’s standards it looks quite ordinary? But it didn’t in 1980. Despite quite normal looks by today’s standards, it was too novel for the editor of the magazine I designed it for. He called it “the ugliest car I have ever seen!” Novel design is interesting and exciting. Up to a point it is charming, even beautiful. But beyond that point it can be bothersome or repulsive to the point of ugliness because it violates existing stereotypes. So, given a choice, most people, like my editor, prefer products with designs just new enough to entertain them by reinforcing existing beliefs, values and stereotypes. Familiarity is comforting. If a product is novel enough to radically improve associated stereotypes, we applaud it as a disruptive innovation. If it is aimed at improving conditions for a product’s users and the broader culture, it would qualify as an edifying agent. As an example of disruptive innovation a car manufacturer might go all out to improve the aerodynamic efﬁciency of its products in the interest of slowing climate change and environmental degradation more quickly rather than reinforcing the status quo by merely entertaining its audience. The creators of such products could educate consumers and change their tastes through marketing and advertising programs aimed at more than sales. Engineers and designers could then proceed with less concern about offending the public and potential consumers with extremely unusual designs. In the process they would achieve ediﬁcation through disruptive innovation.

References 1. Shannon, C.E., Weaver, W.: The Mathematical Theory of Communication. University of Illinois Press, Urbana (1949) 2. Berlyne, D.E.: Aesthetics and Psychobiology. Appleton-Century-Crofts, New York (1971) 3. Osgood, C.E.: The Measurement of Meaning. University of Illinois Press, Urbana (1957) 4. Coates, F.D.: Watches Tell More than Time. McGraw-Hill, New York (2003)

Graphic Design Analysis Model Elisabete Rolo(&) CIAUD, Lisbon School of Architecture, Universidade de Lisboa, Rua Sá Nogueira, Pólo Universitário do Alto da Ajuda, 1349-063 Lisbon, Portugal [email protected]

Abstract. To understand communication design, it is essential to understand what is being done in the present and what has been done in the past. And for that, it is essential to study a large number of works and to analyse them according to coherent criteria, established by an analysis model. In this article, we reflect on a graphic objects analysis model, based on the Laurent Gervereau image analysis grid, described in the book “View, Understand, Analyse Images” (Lisbon: Edições 70, 2007. [Original publishing in French “Voir, Comprendre, Analyser les images”, 1996]), which contemplates three main ﬁelds: description, context and interpretation. Its adaptation to graphic design results in an analysis tool that we consider to be universal enough to suit all types of graphic objects, but also sufﬁciently detailed so that each analysis can be meaningful and relevant. Keywords: Graphic design history

Analysis Analysis model

1 Introduction To understand communication design it is essential to understand what is being done in the present and, especially, what has been done in the past. To get this knowledge, it is important to know what design history books have to tell us, but, above all, to observe the original design objects. Only by this way can we know all the material, shape and printing details of the object, and thus, get a better idea of its physical reality. And besides this, to study the complete work of a designer or a history period, it is essential to study a large number of works and to analyse them, according to coherent criteria established for that purpose. In this sense, it is extremely important to have an analysis model able to suit groups of different objects to analyse in a particular study. In this article, we reflect on a graphic objects analysis model, based on the Laurent Gervereau analysis grid, described in the work “View, Understand, Analyse images” (Lisbon: Edições 70, 2007 [Original edition in French “Voir, Comprendre, Analyser les images”, 1996]), which served as the basis for the graphic works analysis done in the thesis “Olhar | jogo | espírito de serviço: Sebastião Rodrigues e o design gráﬁco em Portugal (Rolo, E., Lisboa, Faculdade de Arquitectura da Universidade de Lisboa, 2015) [2].

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 433–438, 2019. https://doi.org/10.1007/978-3-030-02053-8_66

434

E. Rolo

2 The Laurent Gervereau Analysis Model The Laurent Gervereau’s analysis model is divided into three main ﬁelds: description, context and interpretation, and in each ﬁeld there are several aspects to focus and questions to answer. The description ﬁeld is intended to physically characterize the graphic object as fully and pertinently as possible. It encompasses aspects of technique, stylistics and thematic. The technique is intended to provide information on aspects related to the visual object cataloguing, such as the name of the publisher, the way of identifying the publishers, the production date, the type of medium and technique, the format and the location. Stylistics intends to take note of aspects such as the number of colours and estimation of surfaces and the predominance, the volume and intentionality of the volume, the iconic organization (what are the guidelines?). The thematic intends to observe the title and the text-image relationship, to make an inventory of the represented elements, to perceive which symbols are represented and which are the general themes and their meaning. The context ﬁeld intends to locate the object in the time and space in which it was conceived and to relate it to the problem it answers, in order to better understand its peculiarities. The analysis considers the upstream and downstream context. In the upstream context, it tries to answer the questions “from which technical, stylistic, thematic medium, does this image come?”, “Who did it and what is the relation to his personal history?” and “Who commissioned it and what relation has to the society of the moment history?”. In the downstream context, it aims to understand if “the image has known a dissemination contemporary of the time of its production or later disseminations?” and “What evidences or testimonies do we have about its mode of reception over time?”. The interpretation ﬁeld intends to ﬁnd meaning for the graphic object particularities. It is the analysis component that has the highest subjectivity degree, although we believe that any analysis always has some degree of subjective interpretation. In this ﬁeld, the analysis grid deals with the initial and posterior meanings and the balance and personal opinions. In the initial and subsequent signiﬁcations, we intend to answer the questions “did the image creator or creators suggested a different interpretation of its title, legend, or ﬁrst meaning? What analyses contemporary of its production time can we ﬁnd?” and “what further analyses can be made?”. Regarding the balance and personal evaluation, they should be enunciated after the description and context elements and after knowing any other interpretations that have been made over time. One should try to answer questions such as “How do we see this image today?” and “what subjective appreciation related to our individual taste can we give it?” (Table 1).

Graphic Design Analysis Model

435

Table 1. Laurent Gervereau analysis grid. Work identiﬁcation DESCRIPTION Technique

Stylistic

Thematic

CONTEXT STUDY Upstream

Downstream

INTERPRETATION Initial meanings, later meanings

Balance and personal evaluation

Emitter or emitters name Mode of identiﬁcation of the emitters Date of production Media type and technique Format Location Number of colours and surfaces and predominance estimation Volume and volume intentionality Iconic organization (what are the guidelines) What is the title and what is the text-image relationship Inventory of represented elements What symbols What are the general themes? (Which is the primary sense?) From what technical, stylistic and thematic means does this image come? Who made it and how does it relate to his personal history? Who commissioned it and what relation has to the society of the moment history? Has the image experienced a dissemination contemporary of the time of its production or later disseminations? What evidences or testimonies do we have about its mode of reception over time? Did the image creator or creators suggested a different interpretation of its title, legend, or ﬁrst meaning? What analyses contemporary of its production time can we ﬁnd? What further analyses can be made? Accordingly to the strong elements revealed in the description, in the study of context, in the inventory of interpretations over time, what general balance can we make? How do we see this image today? What subjective appreciation related to our individual taste – announced as that – could we give it?

3 The Analysis Model Created From Laurent Gervereau’s perspective, the presented analysis methodology is adequate to any kind of images. However, in order to transform it into a more functional working instrument adapted to the speciﬁc case of graphic design, we have introduced some changes.

436

E. Rolo

The analysis of a graphic design object will always have an interpretation subjective component, as the discipline is not a science and will always be linked to creativity and expressiveness, nevertheless, never losing the ability to communicate and fulﬁl its ultimate goal. Due to this subjectivity, each object analysed could originate almost inﬁnite speculations about its context and interpretation. This could be considered a disadvantage in a study of this nature. However, we consider that one must try to restrict these analysis ﬁelds in the way that best serves the research. As Gervereau tells us, “(…) each one must situate his analysed corpus and his questions in relation to his speciﬁc objectives. It’s preferable to focus a precise aspect than trying to embrace too much and not reach any consistent result.” [1]. The analysis model created from the Gervereau’s grid kept the same main analysis ﬁelds, but changed some topics, making them clearer and adapted to the graphic design speciﬁc case. Thus, in the description, we considered the ﬁelds of technique, thematic and design (word that we use instead of “stylistic”). The technique description ﬁeld is intended to gather a set of objective data, cataloguing the graphic object (date, format, print run, media – paper – printing technique, location, emitter and editorial structure). In the thematic description three important ﬁelds were considered: the inventory of the represented elements, the general thematic or the work primary meaning, and the relation between the theme and the image. In the design ﬁeld it is important to mention and analyse the object grid (or geometric structure), the colours, the typography, the presence (or not) of illustrations and photographs and their generic description, and ﬁnally, although briefly, the sketches and project drawings, if they exist. This ﬁeld of the analysis model is the most relevant for the study in question, since it is through it that the aspects most directly related to the discipline of graphic design are analysed. For this reason, it is important to explain in an unambiguous way what one intends to analyse in each of these ﬁeld points: grid (or geometric structure), colours, typography, illustrations and photographs and study or design drawings. In the “grid” section, we consider pertinent to analyse the structural axes of the represented graphic elements or the geometrical structure inherent to the visual ﬁeld. In the “colours” section we consider essential to list the object print colours, and the main shades formed by these print colours. The aim is to deﬁne the colour palette for each work analysed, in order to ﬁnd harmonies and contrasts. Here it is important to note that the psychological or symbolic analysis of colour shall not be considered, since this is a complex area that requires a dedicated study. In the typography ﬁeld, we consider important to analyse the typefaces used in each graphic object and the possible transformations that these typefaces may have suffered (especially in headings or letterings) in order to make them graphically more appealing and interesting. Corrections of spacing between characters should also be described. In the case of editorial design objects (the interior of books), typographic analysis should be more complex, including factors such as typeface, leading, line extension, and typographic grid of the various page layouts. Regarding illustrations and photographs, we consider essential to state whether they exist or not in the object, and if they exist, to mention their authorship (if known), and describe them briefly, in a manner to make the analysis more complete. In some graphic spoils to analyse, there are study sketches. Whenever justiﬁed, their existence must be mentioned, presented and related to the ﬁnal work.

Graphic Design Analysis Model

437

The second part of this analysis model concerns to the graphic object context, both upstream and downstream. Upstream it refers to what is prior to the work execution, and which essentially reflects data connected to the client and the designer. These aspects relate to the objectives of the design object, with the commissioner particularities it and with relevant information about the participants in the design process. The downstream context concerns the audience and the way in which the message was disseminated and received by that audience. The third part of the analysis focuses on the interpretation of the graphic message possible meanings. It is an analysis which always involves a subjective appreciation and which intends to describe the symbolic elements and their meanings and other works possible influences. In this ﬁeld, we have decided to add an item related to legibility and expressivity. We, therefore, intend to appreciate the work by trying to understand whether its nature is intended to fulﬁl a readability function (i.e. ease of communicating information) or if it is intended to fulﬁl an expressivity function (with the intention of drawing attention and to be distinguished in the visual communication scenery). We are, however, aware that these two characteristics do not exclude each other. Expressive work can also (and must) be legible and a legible work will always have its inherent expressivity (Table 2).

Table 2. Graphic design analysis grid based on Laurent Gervereau. Work identiﬁcation DESCRIPTION Technique Cataloguing elements

Thematic

Design

CONTEXT Upstream Client and designer

Date Format Print run Media (paper type) Print technique Location Emitter Editorial Structure Inventory of represented elements General thematic (Which primary sense?) Relationship theme/image? Grid/Geometric Structure Colours Typography Illustrations and photographs Study Sketches/project drawings Objectives of the design object Commissioner particularities Participants in the realization (continued)

438

E. Rolo Table 2. (continued) Work identiﬁcation Downstream How was it disseminated? Designer and audience How did the public receive it? INTERPRETATION Signiﬁcances Symbolic elements and their meanings Influences of other works Legibility and expression

Although the purpose of this analysis model is to analyse all types of graphic objects, it is essential to divide these objects into categories (such as covers, editorial design, posters, graphic marks, etc.), so that they can be compared to one another, in order to make the analysis useful. Not all of these objects imply the same type of analysis, since the amount of information they convey is different. Analysing a 200page book is different from analysing a graphic brand, since one is characterized by a large amount of information and the other is characterized by conciseness, clarity and extreme reduction of means. However, both elements have the common purpose of communicating graphically, and this justiﬁes a single analysis matrix. However, for certain graphic objects, there may be ﬁelds that do not make sense, and that should always be pointed out.

4 Final Remarks We believe that this analysis model has enough universality to ﬁt all types of graphic objects, but also enough detail to make each analysis relevant and meaningful. We believe that this kind of tools should be increasingly thought out and disseminated in order to allow the analysis of communication design objects to other people than just historians. These other people are especially the designers, who, by being inside the profession, will always have a different and deeper look at the graphic reality, which can be very enriching for the dialogue around those themes and, above all, for the construction of a richer, more diverse and less linear history. Acknowledgments. The Author gratefully acknowledges the support to this paper presentation by CIAUD, only possible by National Funds by FCT – Fundação para a Ciência e Tecnologia, Portugal.

References 1. Gervereau, L.: Ver, Compreender, Analisar as Imagens. Edições 70, Lisboa (2007). [Original publication: Voir, Comprendre, Analyser les images, 1996] 2. Rolo, E.: Olhar | jogo | espírito de serviço: Sebastião Rodrigues e o Design Gráﬁco em Portugal. Faculdade de Arquitetura da Universidade de Lisboa, Lisboa (2015) [Tese de Doutoramento em design]

Education and Training of Road Safety Auditors on the Implementation of Human Factors Principles in Safe Road Design Sophia Vardaki1(&) and Evangelos Bekiaris2 1

School of Civil Engineering, Department of Transportation Planning and Engineering, National Technical University of Athens (NTUA), Athens, Greece [email protected] 2 Center for Research and Technology Hellas (CERTH), Hellenic Institute of Transport (HIT), Thermi, Thessaloniki, Greece [email protected]

Abstract. Highway engineers are required to have the appropriate knowledge and tools to enable them to perform effectively in their work but must also be aware that road safety is everyone’s responsibility. The Greek road safety audit training program and educational materials developed for the training of engineers of highway projects who want to be certiﬁed as road safety auditors is the ﬁrst ofﬁcial systematic effort for the training and certiﬁcation of highway engineers working in the private and public sector. This paper presents the training objectives and an overview of the topic of ‘Human factors in safe road design’ of the training course and the educational material, which represents a new area in the training of highway engineers. Keywords: Human factors Road systems engineering Professional development Road safety management

1 Introduction The modern Safe System approach to road safety is the basis for road safety strategies in advanced countries [1]. This approach is centered on the starting point that it is not ethically acceptable to have deaths and serious injuries in a road trafﬁc system. Those who design the system bear the main responsibility for safety. Engineers therefore are required to have the appropriate knowledge and tools to enable them to perform effectively in their work, but must also be aware that road safety is everyone’s responsibility; it is a common duty for all decisions and choices that affect the road system infrastructure to be made with safety in mind. For the successful implementation of the road safety audit (RSA) procedure in Greece and in accordance with the ministerial decision approving the guidelines for road infrastructure safety management for the Trans-European Road Network (TEN-T), training and follow-up courses and educational material were developed for candidate road safety auditors [2, 3]. The training course and the educational material were the products of the project entitled “Development of Training Curriculum and Educational © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 439–444, 2019. https://doi.org/10.1007/978-3-030-02053-8_67

440

S. Vardaki and E. Bekiaris

Material for Candidate Road Safety Auditors”, which was assigned (October 2014) to the Hellenic Institute of Transport (HIT) of the Center for Research and Technology Hellas by the Greek Ministry of Infrastructure, Transport and Networks. For the development of the training course, the requirements in relation to professional background of candidates were taken into account, i.e. candidate road safety auditors have professional experience and knowledge in the area of road safety engineering. The structure of the course was in line with good international practices, while the content of both the course and the educational material was largely based on the road safety auditors job description [4–6]. The training course consists of two parts: theory and practice. The theoretical part includes three main sections: road safety policy and the institutional framework, road safety audit issues, and road safety engineering. The practical part involves RSA training and examination. A successful performance in the examination results in the acquisition of a certiﬁcate of competence [2, 3]. This paper presents the training objectives and an overview of the topic “Human factors in safe road design” of the training course and educational material which represents a new area in the training of highway engineers.

2 Methods and Content The topic of the human factors (namely “Human factors in safe road design”), which is currently present in most road safety audit training and accreditation internationally, contributes to the development of human factors as a signiﬁcant area of knowledge and experience of road safety engineering which is crucial in conducting Road Safety Audits. 2.1

Methodology and Process

The content of the human factors chapter of the educational material, as well the related course topic, were formulated in order to meet the following requirements: 1. They should reflect the role of Road Safety Auditors and job description, especially regarding the identiﬁcation of road user needs in relation to likely conflicts in the context of a speciﬁc project. 2. They should be adapted to candidates’ professional proﬁle regarding their knowledge background and skills, especially regarding their formal education and the way they practice their profession (e.g. guidelines). 3. They should cover modern international literature regarding the rules relating to human factors and their application in road systems. 4. They should be one component of the training curriculum along with the topics of the institutional framework of road safety management; Road Safety Audits; identiﬁcation and improvement of hazardous locations; review of existing design guidelines; signing; and vehicle restraint systems; all of which must work together as a combined unit delivered within a certain time.

Education and Training of Road Safety Auditors

441

5. In combination with the other topics, they should enhance workforce understanding and application of trafﬁc safety culture and a commitment to road safety, placing emphasis on the implementation of the principles of the safe system approach and human factors in road design. 6. They should be effectively delivered. The issues included in the material are key issues concerning human factors that should be considered in road design and trafﬁc engineering in terms of design elements and features associated with demanding driving tasks. Examples that were considered suitable for the training objectives regarding the application of human factors principles in road design and information provision were drawn from design features from the national road network (Fig. 1 shows an example).

Fig. 1. A less desirable combination of information at various levels of the driving task

Human factors principles and related concepts are not essentially presented in existing guides/standards that are available to road designers and used for safe road design. Road users adjust their driving behavior to the road environment they see and expect, and therefore their behavior depends on the design. This new way of thinking in road design entails the need for road designers to understand the new safety perspective and for road design to reflect the safe system/human factors principles [7, 8]. Human factors principles and concepts for the training of candidate auditors were based on a review of available modern guides, publications and references that reflect the current trends of an anthropocentric approach to road safety. The human factors issues in these sources cover a broad spectrum of human factors implementation in the design and operation of road systems and in various degrees of detail (e.g. [9–11]. The challenge in selecting the human factors issues for inclusion was twofold: the selection of the most relevant material to educate road designers with the qualiﬁcations of candidate auditors that would also be a part of a training curriculum addressing the training needs as a whole within a time frame of 30 h [2]. The scope of the human factors topic is largely based on the available time frame (i.e. 3 h). Figure 2 depicts the diagrammatic representation of the methodology and process for developing the topic of human factors in safe road design both for the course and for the corresponding chapter of the educational material.

442

S. Vardaki and E. Bekiaris

Fig. 2. Methodology for the development of the human factors topic for the Road Safety Audit (RSA) course

The examples regarding the implementation of these principles that were selected for presentation during the course aim to provide the candidate road safety auditors with basic human factor knowledge. They also provide the opportunity for candidates to link the basic theoretical knowledge acquired on the course with practical skills, and apply human factors principles in practice. In this regard, the emphasis in terms of human factors issues was placed on speciﬁc characteristics and safety problems of the Greek road network including roads serving multiple functions with the coexistence of different operational and design characteristics, poor speed management, the presence of work zones on an extensive part of the Trans-European Road Network. A future extension of the time frame of the course (if deemed necessary, e.g. following an evaluation of the course) may result to the expansion of the scope of the topic in order to include more road design and operational issues. 2.2

Training Objectives

References [4, 9–11] are examples of reference sources used to develop the content of “Human factors in safe road design” course topic and the chapter in the educational material, in order to meet certain training objectives. Training Objective 1. Candidates will be able to recognize the difﬁcult tasks faced by road users and incorporate needs into road design and trafﬁc engineering. The environment is sometimes too complex for the road users to cope with; road users’ capabilities and limitations affect whether they correctly comprehend road geometric design, signs and trafﬁc control devices. Road designers-candidates will be able to see the road through the eyes of its users, especially the vulnerable ones (e.g. the elderly or those unfamiliar with the road). Training Objective 2. Candidates will be able to implement human factors into road design and trafﬁc engineering. They will be able to apply positive guidance in road design and operations with the aim of reducing driver workload (i.e. facilitating the driving task), and to cater for self-explaining designs enabling road users to make good decisions.

Education and Training of Road Safety Auditors

2.3

443

Overview of Human Factor Issues

Some of the key issues included in the human factors course topic are: 1. The implications of safe system and sustainable safety principles for road design and operation. The concepts of self-explaining roads and safe speeds. 2. The characteristics of road users such as expectations, attention, workload, errors, ﬁeld of view, perception-reaction time, short-term memory time span, age-related limitations. 3. The driving task and driver performance. The way the road users seek, perceive and process information. The primacy concept, and its implications for road design and information provision. 4. Uniformity and consistency principles as applied in road design and information provision; decision sight distance, as an example of integrating time into road design and trafﬁc engineering. Positive guidance and workload management. 5. Task analysis of demanding driving tasks/maneuvers (curve driving, intersection approach and entering maneuver). Sources of driver errors and false perceptions in respect of road and trafﬁc conditions; the requirements imposed by major design elements; improvements that reduce these errors. 6. The application of human factors principles in road work zones, intersections, interchanges and transition areas from rural to urban areas (i.e. typical situations of changes in the functionality of a road and necessary speed adjustments).

3 Conclusions Current signiﬁcant trends such as increasing respect of society for human life and the associated holistic approach to road safety management imposes requirements on the qualiﬁcations of the civil and surveying engineers. Although current university courses in highway engineering reflect contemporary safety principles such as human factors and safe system principles, the RSA training course provides an opportunity for the professional development of experienced road and trafﬁc engineers educated in the context of traditional road design courses [12]. The training course and educational material for candidate road safety auditors (professionals with the prerequisites for enrollment in the training course) attempts to complement experience in road safety engineering and Road Safety Audits and to offer knowledge of the latest relevant research and practices. The human factors topic, in particular, includes basic human factors principles as applied in road systems design, focusing on road user needs, capabilities and limitations; aid and augment the judgment of experienced highway designers and trafﬁc engineers through presentation of relevant information and insights from the scientiﬁc literature; and complement existing sources of road design information. The Greek RSA training course includes examples of RSAs in the national road network so far, i.e., problems identiﬁed as well as challenges in implementing RSA recommendations [13]. We expect successful outcomes from team teaching [14], i.e., educators being members of academia as well as ﬁeld experts with hands-on experience

444

S. Vardaki and E. Bekiaris

who can contribute to training that promotes ready-to-use knowledge in the areas of RSA and human factors and, eventually, in achieving the learning outcomes [3]. Furthermore, candidates can increase their professional potential and employability, assets that are particularly important to those willing to work abroad in times of economic uncertainty [15].

References 1. OECD/ITF (Organisation for Economic Cooperation and Development/International Transport Forum): Zero Road Deaths and Serious Injuries: Leading a Paradigm Shift to Safe System (2016) 2. GGGR (Government Gazette of the Greek Republic): Approval of the certiﬁcation scheme, General and Speciﬁc Guidelines, Training program and Educational material for the Accreditation of Road Safety Auditors. Second Issue. No. Sheet 1694, 13 June 2016 3. Vardaki, S., Dragomanovits, A., Gaitanidou, E., Mavromatis, S., Bekiaris, E., Kanellaidis, G.: Development of a training course for road safety auditors in Greece. In: 95th Annual Meeting of the TRB, pp. 10–14. TRB, Washington, D.C., (2016) 4. Austroads: Guide to Road Safety Part 6: Road Safety Audit, Austroads Publication No. AGRS06/09, Sydney (2009) 5. DfT (Department for Transport): Design Manual for Roads & Bridges - Road Safety Audit (5). Assessment and preparation of road schemes Part 2. HD 19/15 Road Safety Audit. DfT – Highways Agency, London (2015) 6. Belcher, M., Proctor, S., Cook, P.: Practical Road Safety Auditing, 3rd edn. TMS Consultancy and Thomas Telford Limited (2015) 7. Larsson, P., Tingvall, C.: The safe system approach – a road safety strategy based on human factors principles. In: Harris, D. (ed.) Engineering Psychology and Cognitive Ergonomics. Applications and Services. Lecture Notes in Computer Science, vol. 8020, pp. 19–28. Springer, Heidelberg (2013) 8. Kanellaidis, G., Vardaki, S.: Highway geometric design from the perspective of recent safety developments. ASCE J. Transp. Eng. Forum Piece 137(12), 841–844 (2011) 9. Theeuwes, J., Van der Horst, R., Kuiken, M.: Designing Safe Road Systems. Ashgate, Farnham (2012) 10. Campbell, J.L., Lichty, M.G., Brown, J.L., Richard, C.M., Graving, J.S., Graham, J.L., O’Laughlin, M., Torbic, D., Harwood, D.: Human Factors Guidelines for Road Systems. Presented at the NCHRP Report 600, 2nd edn. Transportation Research Board of the National Academies, Washington, DC (2012) 11. Mackie, H.W., Charlton, S.G., Baas, P.H., Villasenor, P.C.: Road user behaviour changes following a self-explaining roads intervention. Acc. Anal. Prev. 50, 742–750 (2013) 12. Gross, F., Jovanis, P.P.: Current state of highway safety education: safety course offerings in engineering and public health. ASCE J. Prof. Issues Eng. Educ. Pract. 134(1), 49–58 (2008) 13. Vardaki, S., Papadimitriou, F., Kopelias, P.: Road safety audit on a major freeway: implementing safety improvements. Eur. Transp. Res. Rev. 6(4), 387–395 (2014) 14. Kim, C., Jackson, D., Keiller, P.: Interdisciplinary Team-Teaching Experience for a Computer and Nuclear Energy Course for Electrical and Computer Engineering Students. Am. J. Eng. Educ. 7(1), 1–8 (2016) 15. Lambropoulos, S., Pantouvakis, J.P., Marinelli, M.: Reforming civil engineering studies in recession times. Proc. Soc. Behav. Sci. J. 119, 776–785 (2014)

Ergonomic Analysis in the Welding Laboratory of the Federal Institute of Paraíba – IFPB, João Pessoa Campus Amanda Ramos de Amorim(&) and Aarão Pereira de Araújo Junior IFPB, Federal Institute of Education, Science and Technology of Paraíba, 1º de Maio Street. 720, João Pessoa 58015-435, Brazil [email protected], [email protected]

Abstract. In Brazil, the technical and technological education is ruled in the professional qualiﬁcation. The laboratory is a very important space for the students, mostly for the ones who are a part of a group of students either from the technical or technological education. The value of having a laboratory in an educational institution is extraordinary, because the students have the opportunity of doing everything that has been seen in theory. The goal of this paper was analysing the welding laboratory of IFPB according to its layout and furniture. Some visits to the lab were made, to collect physical data and photographic records as well. Ergonomic analysis of work was performed in oxy acetylene, TIG, arc and MIG/MAG welding workstations. After data analysis, it was observed that the space that has been studied needs a lot of ergonomic improvements to provide better teaching and learning quality to the users. Keywords: Welding laboratory

Interior design Ergonomics

1 Introduction The laboratories from the educational institutes that offer courses on such teaching levels are paramount for one’s successful qualiﬁcation, and it is important to understand the concept of laboratory in order to understand the subject matter of the present paper. According to Del Pino and Krüger [3], a laboratory can be considered as an environment in which speciﬁc tasks are carried out in a particular subject area. Presently, it can be noted that some welding laboratory items are not consistent with the principles of ergonomics, such as the furniture and the layout used. That’s the reason why this work aims to analyze the furniture and the layout of IFPB’s Welding Laboratory, João Pessoa campus, based on the principles of ergonomics. The methodology used consists of on-site monitoring, images, environment and furniture dimensional analysis, bibliographic research and interviews with professors that teach their classes in such an environment. The Principles of Ergonomics in the Workplace – Regularoty Norm no. 17 – were used as a reference for the analysis of the laboratory workplaces and its variables were examined with respect to the workplace.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 445–450, 2019. https://doi.org/10.1007/978-3-030-02053-8_68

446

A. R. de Amorim and A. P. de Araújo Junior

2 Ergonomics Iida [5] briefly deﬁnes Ergonomics as “the study of man’s adaptation process to his work”. With respect to the goals of the present work, the author states that “ergonomics has a wide view that embraces the planning of tasks and projects, which take place prior to the work itself, and tasks of control and assessment, which take place during and after the work itself. It is imperative that everything mentioned is done in order to achieve successful results.” The ﬁrst deﬁnition of Ergonomics was given by the Ergonomics Society, apud Iida [5], located in England, which states that “it is the study of the relationship between man and his work, equipment, workplace environment and, most importantly, the implementation of anatomical, physiological and psychological knowledge for solving the problems that emerge from this relationship.” 2.1

Principles of Ergonomics in the Workplace Environment – Regulatory Norm no. 17

This Regulatory Norm [2] aims to set guidelines that enable the adaptation of the work conditions to the workers’ psychological and physiological characteristics, in order to fully provide comfort, safety and an effective performance. The main criteria for the ergonomics project of this work that were mentioned in this norm and taken into consideration for the creation of the preliminary draft are: (a) the height and characteristics of the work surface must be consistent with the type of activity, the distance of the eyes with regard to the workplace and the seat height; (b) the work area must be within reach and appropriate for the worker’s visualization; (c) the dimensional characteristics must enable an adequate position and movement of the body; (d) if the work requires constant head inclination above 20° or 30°, the workplace must be remeasured and the seat and workbench height changed [5]. 2.2

Workplace

Iida [5] sees the workplace as the physical structure of the man-machine-environment system. It is a production unit that encompasses man and the equipment used by him to do the work, and the environment where it is done. The ergonomic emphasis of the work place is almost entirely based on the biomechanical analysis of the posture, as well as the interactions between man, system and environment. Among the great amount of criteria used to examine the adequacy of the workplace, taking into consideration the ergonomic point of view, the posture and the efforts required from the workers are highlighted for it determines the main body parts that reduce muscle contraction [5].

Ergonomic Analysis in the Welding Laboratory

2.3

447

Anthropometry

Panero and Zelnik [6] deﬁne anthropometry as the science that particularly studies the measure of the human body with the purpose of distinguishing individuals and groups. It is necessary to be aware of the values that should be taken into consideration between the minimum and maximum measures of the population under analysis. Iida [5] states that the fact that almost all male anthropometric measures are higher than females’, the maximum measure is represented by the percentile 95% for males and the minimum percentile 5% for females. The maximum ones correspond to the access and the minimum measure within reach. The following nine measures were taken for the workplaces mentioned in the present study: (a) height; (b) head height while seated; (c) eye height while seated; (d) shoulder height while seated; (e) elbow height while seated; (f) thigh height; (g) seat height (popliteal fossa); (h) forearm length; (i) arm length. These measures were analyzed through statistical method. According to Iida [5], if the average is 170 cm and the standard deviation is 10 cm, it is possible to calculate the interval from 5 to 95% in which k = 1,65 is shown as follows: to 5% = 170 − 1,65 10 = 153,5 cm; to 95% = 170 + 1,65 10 = 186,5 cm. Thus, 90% of the population are between 153,5 cm and 186,5 cm. 5% is below 153,5 cm whereas the other 5% is above 186,5 cm. The use of the human mathematical model simpliﬁes the measurement steps of the project. And the remaining measures could be deduced through mathematical formulae and calculations consistent with Roozbazar’s model (1977) apud Iida [5], in which the constant mathematics is the height.

3 Results 3.1

Description of the Environment Under Analysis and Characterization of the Laboratory Users

The environment under analysis is the Welding Laboratory, located at the Federal Institute of Education, Science and Technology of Paraíba - IFPB, in João Pessoa, in the Mechanics and Industrial Automation building. The group that uses the welding laboratory consists of 92 (ninety-two) students from the age of 17 onwards, 2 (two) professors and 1 (one) laboratory worker. 3.2

Anthropometric Proﬁles of the Students that Use the Laboratory

A sample with 80 students – 40 males and 40 females – from a variety of courses at IFPB was analyzed, in order to describe the students’ anthropometric proﬁles. The students’ height and weight were measured. Their height was measured through a tape measure that was placed on a wall and they were asked to take off their shoes so that there wouldn‘t be any inaccuracies. Subsequently, their weight was measured through a digital portable weighing scale (Table 1).

448

A. R. de Amorim and A. P. de Araújo Junior

Table 1. Minimum and maximum anthropometric measures of the population under analysis Static anthropometric measures

Criteria Min. Max. a. Stature • b. Head height while seated • c. Eye height while seated • d. Shoulder height while • seated e. Elbow height while • seated f. Thigh height • g. Seat height • h. Forearm length • i. Arm length • Source: Personal Archive (2018)

3.3

Females 5% 50% 153 163 80 85,2 69,5 74 51,7 55,1

95% 173 90,5 78,5 58,5

20,6 22

23,3 22,4 23,6 24,8 24,8

13,1 38,1 38,9 70,7

14,9 43,1 44 80

14 40,6 41,4 75,3

Males 5% 50% 166 175 86,8 91,5 75,4 79,4 56,1 59,1

14,3 41,3 42,1 76,7

15 43,6 44,4 80,1

95% 184 96,2 83,5 62,2

15,8 45,8 46,7 85

Measures taken 184,0 96,2 69,5 51,7

15,8 45,8 38,9 70,7

Data Analysis

For the data analysis of this work, the following two workplaces were presented: the coated electrode welding workplace, which was executed in enclosed booths, and the oxy acetylene welding one, which was executed in an open environment. These analyses will briefly discuss the workplaces in accordance with images of students executing these tasks and determine whether they are consistent with the recommendations of the Regulatory Norm no. 17 (Fig. 1).

Fig. 1. Students executing the coated electrode welding. Source: Personal Archive

A few problems were identiﬁed in this workplace. The workbench was very low which led to the inclination of the spine that went to extremes. This bad posture could cause a short-term pain. There was a lack of space to ﬁt the thighs appropriately in the workbench and to keep commonly used supplies. The chair was unsuitable for the work. The seat area was made out of a hard material which causes discomfort and pain

Ergonomic Analysis in the Welding Laboratory

449

in the popliteal fossa. Enclosed booths can cause students to inhale gases that are produced throughout the process. For manual work while seated or standing, the workbenches, desks, writing desks and boards must provide workers with good work conditions for their posture, visualization and execution, as well as be consistent with the following requirements: (a) the height and characteristics of the work surface must be consistent with the type of activity, the distance of the eyes with regard to the workplace and the seat height: NOT CONSISTENT (b) the work area must be within reach and appropriate for the worker’s visualization: PARTIALLY CONSISTENT (c) the dimensional characteristics must enable an adequate position and movement of the body: NOT CONSISTENT (d) if the work requires constant head inclination above 20° or 30°, the workplace must be remeasured and the seat and workbench height changed: NOT CONSISTENT (Fig. 2)

Fig. 2. Students executing the oxy acetylene welding. Source: Personal Archive (2018)

A few problems were identiﬁed in this workplace. The workbench was very low which led to the inclination of the spine that went to extremes. This bad posture can cause a short-term pain. There was a lack of space to ﬁt the thighs appropriately in the workbench and to keep commonly used supplies. The chair was attached to the workbench which prevented the student from adjusting it and approaching to the workbench. There was a lack of a backrest which can cause low back pain. The seat area was made out of a hard material, causing discomfort. The workbench was not adequately measured which led the student to be partially out of the workplace. With respect to the recommendations of the Regulatory Norm no. 17 and Iida [5], for manual work while seated or standing, the workbenches, desks, writing desks and boards must provide workers with good work conditions for their posture, visualization and execution, as well as be consistent with the following requirements:

450

A. R. de Amorim and A. P. de Araújo Junior

(a) the height and characteristics of the work surface must be consistent with the type of activity, the distance of the eyes with regard to the workplace and the seat height: NOT CONSISTENT (b) the work area must be within reach and appropriate for the worker’s visualization: NOT CONSISTENT (c) the dimensional characteristics must enable an adequate position and movement of the body: NOT CONSISTENT (d) in case the work requires constant head inclination above 20° or 30°, the workplace must be remeasured and the seat and workbench height changed: NOT CONSISTENT

4 Conclusion This very work provided an opportunity to implement some tools presented in Ergonomics. The ergonomic analysis of work, as well as the use of minimum and maximum measures as a recommendation for the creation of a workplace for both men and women were emphasized. This work enables an intervention in the area under analysis and propose an interior design preliminary draft that can embrace other project-related variables, such as the layout rearrangement, the proposal of new furniture and work environments, accessibility, etc.

References 1. ABERGO. O que é Ergonomia (2000). http://www.abergo.org.br/internas.php?pg=o_que_e_ ergonomia. Accessed 16 May 2017 2. Brasil, Ministério do Trabalho e do Emprego. NR 15 - Atividades e operações insalubres. http://www.guiatrabalhista.com.br/legislacao/nr/nr15.htm. Accessed 26 July 2017 3. Del Pino, J.C., Krüger, V.: Segurança no laboratório. Universidade Federal do Rio Grande do Sul, Rio Grande do Sul (1997) 4. IFPB. Mecânica. https://estudante.ifpb.edu.br/cursos/163. Accessed 20 Feb 2017 5. Iida, I.: Ergonomia: Projeto e produção, 2nd edn. Bluncher, São Paulo (2005) 6. Panero, J., Zelnik, M.: Dimensionamento humano para espaços interiores. Gustavo Gili, Barcelona (2015)

Overview of Empathetic Approaches to Design Inclusive Products Maria Giovanna Trotta Munno(&), Luis Alberto Rosa Sierra, and Fabiola Cortes Chavez Universidad Panamericana. Facultad de Ingeniería. Prolongación Calzada Circunvalación Poniente 49, 45010 Zapopan, Jalisco, Mexico {gtrotta,lurosa,fcortes}@up.edu.mx

Abstract. When we talk about inclusive design we are trigger a life style aspiration of uniﬁcation and shared responsibility. This can foster the concept of facilitating people to operate in extraordinary environments and special people to work in ordinary ones. The design areas, which deﬁne the structure of this work, are Inclusive Design, Universal Design, Design for all and Design for Accessibility. Every time designers start with the creative process of a product development, they infer about other people’s needs, thoughts, emotions, dreams and possibilities. Several times design concepts are failing in producing good results, because of the misunderstanding about the factors that lead the behaviors of the users. The literature speciﬁes that the relevant criteria for assessing information are the accuracy, the legitimacy and completeness of the data gathered. The study of the different users can be staged or naturalistic, following the different goals of the analysis, and the different point of views (the designer or the user). The involvement and the mutuality of the interaction between designers and users are influencing the quality of the communication, of the information gathered, and of the resultant proposals. In addition, the time, available for the development of a project, is influent for all the time-related variables, like the motivation necessary to understand or to trigger social behaviors and emotions. The time or the lack of time is also determining the actors that should be involved into the project, like mediators as facilitators for the synchronic understanding of the user’s problems. Enhance the emphatic communication, the social sensitiveness, and the awareness on users, involves abilities that require a certain level of knowledge. This is related to cognitive biases, mental models, abilities, perspectives, experiences, emotional states, the differences correlated to the genders, cultures and so on. This paper offers an overview of approaches which have the ambition of reducing the distance between designers and users by using empathetic dynamics, to deﬁne and then change emotional states and moods and allow a certain awareness to make signiﬁcant decisions about design solutions. Keywords: User understanding

Mood regulation Emotions map

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 451–456, 2019. https://doi.org/10.1007/978-3-030-02053-8_69

452

M. G. T. Munno et al.

1 Introduction Moods are pervasive, they can change completely the perception about our and the other people conditions. They represent the state of being, and in best case of the wellbeing. People in a good mood are more inclined to transfer the wellness to the others, thus, creating an indirect welfare transfer network. Design for experience is designed to make possible the transition from a negative state to a desired one. The good mood influences the perception, the motivation and behaviors of people to incentive them to repeat the experience, and make it memorable. Several researches proved how aesthetics influences people’s emotions in certain context and environments, like airplane, hospitals, waiting rooms, ofﬁces, and urban spaces. Donald Norman in his book Emotional Design describes the emotional responses of people to the objects [13]. Most of people’s evaluations and choices are driven by emotions, this is at the center of all the persuasive design consumer studies. The study of emotions has a key role in marketing and advertising. Enjoyment, fun and pleasure are emotional states that we deﬁne as very good, which stimulates the user’s curiosity, fantasy, interaction, and challenge [2]. John Carroll deﬁne fun products as things, which are surprising us, don’t feel like they look, don’t sound like they feel, need an interpretation but they are transparent in the way they have to be used [1]. The main difference between mood and emotions relay in the dependence from the environment that not always worth for the ﬁrst. Mood help us to maintain the balance between internal resources and external interactions. Positive moods stimulate to invest the resources [3]. The role of Inclusive Design, Universal Design”, Design for all, Design for Accessibility, when we talk about inclusive problems, is to trigger a life style aspiration of uniﬁed quality and shared responsibility for all people. This paper pretends to offer an overview of design approaches that can guide designers during the concept deﬁnition, which can reduce the distance between designers and users by using empathetic activities. Empathy allows a better understanding of the user to make signiﬁcant decisions about design solutions.

2 Method The emotions and related aesthetics become the structure of the experience, the key for the users, to functionally approach the world and its understanding. After an accurate analysis of the state of art, in this paper are presented two approaches, which have been partially modiﬁed to allow a mutual close-up between users and designer, on the emotional states of short or long-lasting duration.

3 Existing Approaches Over the last years the experiential products have proposed highly interactive solutions. The behavioral study, the reactions to certain stimuli have been related to intelligent algorithm able to learn about the user behavior, and to adapt the stimuli or the functioning appropriately, with a sort of affective loop.

Overview of Empathetic Approaches to Design Inclusive Products

453

Physical interaction, trough gestures, time of reaction, location, direction, manipulation, type of responses, and expressions, guide the generation of a step by step guided experience in which or the intelligent product reacts and adapts his responses to the user to support or to train him [4]. A method which expresses an important concept is Kansei Engineering. It tries to optimize products based on the user’s feelings by measuring certain features (as the face expressions, the eye movements) [10]. As P. Desmet deﬁnes, the mood-interaction during the experience can be monitored in different ways, by using: (1) wearables (having sensors for heart rate, skin conductance and temperature, and respiratory rate); (2) by using contact sensors in intelligent product which are able to measure the body movement weight (impact), time (lack or sense of urgency) the posture, and gestures; (3) without contact sensors, by using video recording, facial expression analysis; (4) auto-deﬁnition of the own feelings. Emotions are part of human complex dynamic experience. They can be perceived, expressed differently due to the previews experiences, mental or physical state social or environmental factors. Affective interaction is the term used to describe an emotional state of shorter (emotion) or long-lasting duration (mood) [5], as shown next in Fig. 1.

Fig. 1. Timing and responsiveness of affective interactions

In user centered design, the understanding, of the user’s needs, has a central role for exploring the important factors, to design a potential transition from the present to the future state, that the user wants for his life. Emotions are the reflex reactions to the another’s emotional state. Empathy, feeling as you are the other, the user, and not only getting emotional but also knowing, having the perspective of the other is crucial to create enjoyable products. Time spent with the users is really indispensable to code the coincidences among motivations, thoughts, mental models, priorities, needs, attitudes, inner conflicts and dreams. Establish the sufﬁcient engagement to establish the connection with the user, is just as important as accepting the differences between the designer and the user. The observations, and studies done in the user’s environment help to discover and deﬁne the variables of use. The communication elucidates more about his story and state, to establish a better understanding of the potential effective languages for the future solution. The roleplay, the simulation of the user’s abilities or limitations allows a signiﬁcant immersion, meaningful to the further reduction of the distance between user and designer. In this direction, the creation of experience prototypes it’s

454

M. G. T. Munno et al.

required [6]. Nowadays with virtual reality (VR), the user stories are presented in an immersive narrative, character development, scenario and storytelling [9]. Exist several methods which experts use to map the user interactions such as: the customer journey map, the empathy map, the mind map, the mood board, the service blueprint, the empathy board, the story board each providing different information, and has distinctive characteristics (complexity, time, provider) [8]. For the reason that emotions result in customer attention, persuasion and memory retention, once, the designer understands the potential effective language for the solution, he can plan what kind of emotions to stimulate trough the physical characteristics of the design solution. Through Table 1, the user and the designer can together deﬁne and visualize emotions: the previous (P), the current (C) and the goal (G), the desired state [7]. Once the designer has founded the distance between the reality and the future state, he goes deeply into the deﬁnition of the ”emotion oriented features”, physical or methodological elements which will be implemented for producing the change of the emotional state. Table 1. eMap for the deﬁnition of the previous (P), the current (C) and the goal/desired state (G), Peter Boatwrite and Jonathan Cagan, 2010 Negative emotions −3 −2 −1 0 1 2 3 Positive emotions Emotion oriented features Reliant Independent Vulnerable Secure Uncertain Conﬁdent Incapable Powerful Apathetic Passionate Neglectful Compassionate Envious Content Depressed Optimistic Sad Joyful Embarrassed Proud Undesirable Sensuous Routine Adventurous Inconsiderate Honorable Sparse Luxurious Lonely Connected Common Distinct

For instance, if the prior emotional state was “independent”, and it refers to a n existing product, the best of the state of art, the designer writes “P” in the correspondent cell of evaluation, given by the recognizes values of the prior solution on the market or the user interpretation (from −3 to 3). The suggestion given in this paper is: let the user/customer/client/enterprise deﬁne the prior and the current state to then propose or design a goal together. Once, the table has been fulﬁlled designer can trace the graph vertical lines, connecting the same state, using a distinctive color for each, to distinguish rapidly the differences between prior, current and goal states. The Table 1 help to visualize immediately the emotional mapping during the time. The emotion oriented

Overview of Empathetic Approaches to Design Inclusive Products

455

features can vary depending on the idea of beauty, on the emotion, the semantic of shapes related to it, and the senses that will be stimulates [11]. If the concept of aesthetics describes the correspondence between the subject, his intentions and the encoded characteristics, then the design solution, with engage the user emotionally intentionally evokes aesthetics, as reflective form of the user’s purpose. The features become the way in which the object reflects, represents and express the knowledge study behind it [12]. When emotions persist over time, they become moods. Regarding to the mood regulation strategies several authors offer different solutions. Considering the table with the 20 activities, deﬁned by Desmet, we will take in account, only, the ﬁrst two columns as guide to full ﬁll the third column of strategies following the previous client/designer mapping scheme. The idea is that the past/previous mood activities can be deﬁned with the user to be then implemented by the changes suggested by the designer. Desmet deﬁned 3 focuses to change a mood: (a) seek RELIEF (reducing unkind feelings), (b) restore BALANCE (reach a balance between resources and investments) or (c) build RESILIENCE (modify a negative mood in something constructive) in mood. Next, the activities have been hidden to allow the designer ﬁll the space with new ones. The columns on the right side, with activities have been divided in two parts, one for the user and the other one for the designer. This, for a further dialog and understanding between user and designer, and to easier visualize the possible changes in the design development process (Table 2). Table 2. Twenty strategy-based mood regulation to design the new activities. Focus Seek RELIEF

Strategy

Seek distractions Self-reward Vent Repress Think positively Trivialize Seek relaxation Seek refreshment Restore BALANCE Reduce demand Avoid new demand Withdraw Eliminate energy drainers Rest Energize Seek social support Build Rationalize RESILIENCE Analyze Transform creatively Embrace Detach

Existing activities

Researched new activities

456

M. G. T. Munno et al.

4 Conclusion The approaches described in this paper, have the ambition of reducing the distance between designers and users facilitating certain transitions between emotional states. Users are a complete complex world to discover. Think about emotions in a design guided process turns the concept of users as integral part of the process, not only considered to solve functionally their needs. The emotions and related aesthetics become the structure of the experience the key for the user to functionally approach the world and its understanding.

References 1. Desmet, P.M.A.: Design for mood: twenty activity-based opportunities to design for mood regulation. Int. J. Des. 9(2), 1–19 (2015) 2. Carroll, J.: Beyond Fun, October 2004. http://interactions.acm.org/archive/view/ september-october-2004/beyond-fun1. [Consultato il giorno 25 Junio 2018] 3. Monk, A., Hassenzahl, M., Blythe, M., Reed, D.: Funology: designing enjoyment 4. Bruns Alonso, M., Hummels, C.C.M., Keyson, D.V., Hekkert, P.P.: Measuring and adapting behavior during product interaction to influence affect. Pers. Ubiquit. Comput. (2013) 5. Krøger, E.: Logging and Visualizing Affect to Facilitate Communication in a Therapeutic Context. Department of Information Science and Media Studies, University of Bergen (2015) 6. Kouprie, M., Visser, F.S.: A framework for empathy in design: stepping into and out of the user’s life. J. Eng. Des. 20(5), 437–448 (2009) 7. Boatwrite, P., Cagan, J.: Built to Love, Creating Products that Captivate Customers. The Science of Product Emotion (2010) 8. Kanella, C.: Development of a methodology for service interaction design (2017) 9. Shin, D.: Empathy and embodied experience in virtual environment: to what extent can virtual reality stimulate empathy and embodied experience? Comput. Hum. Behav. 78, 64–73 (2018) 10. Wodehouse, A., Vasantha, G., Corney, J., Jagadeesan, A., MacLachlan, R.: Realising the affective potential of patents: a new model of database interpretation for user-centred design. J. Eng. Des., 08 March 2018 11. Urquhart, L., Wodehouse, A.: The line model of form and emotion: perspectives on western design. Hum. Technol. 14, 27–66 (2018) 12. Folkmann, M.N.: Exploring aesthetics in design: implications for human–computer interaction. Hum. Technol. 14, 6–26 (2018) 13. Norman, D.A.: Emotional Design. Basic Books, New York (2004)

Assistive Devices for Lower Limbs Under Mechanism of Neuromodulation and Blood Circulation Hui-I Yin(&) and Fong-Gong Wu Department of Industrial Design, National Cheng Kung University, Tainan, Taiwan [email protected]

Abstract. Due to the changing work patterns and diet, ofﬁce workers suffer from Knee osteoarthritis (OA) earlier as results of overweight and lack of exercise. Good assistive effectively improve nerve and vascular in regulating peripheral, so the body can move more efﬁciently. The current studies on improving osteoarthritis and designs mostly focus on minimally invasive surgery or arthroplasty, so there is no physiological mechanism and correlation in considering overall degradation factor. As from the aspect of collaboration mechanism in nerves and peripheral vessels as well as the activation of motor cortex at lateral brain, this research analyzes the key factors that influence middle-aged and elder’s body temperature and skelasthenia in order to integrate the principle and parameters of medical prevention. It aims to develop the improvement devices for joint activity through external and timely treatments. These devices will not only stimulate the cycle metabolism, but also enhance middle-aged and elder’s exercise outcome. Keywords: Knee osteoarthritis

Assistive device Design

1 Introduction 1.1

Background

Due to the changing work patterns and diet, ofﬁce workers suffer from knee osteoarthritis (OA) earlier as results of overweight and lack of exercise. Since the birth rate decreases and stress on daily spending increases, elders’ physical status becomes a key. After 35 years old, the rate of hyperplasia of cartilage is greatly reduced, which makes the cartilage wear up linearly and let many people over 50 years old have knee discomfort, which is related to dementia, stroke and coronary heart disease. It is listed as the four major diseases that cause disability in the elderly (Wanless et al. 2006). According to the statistics of the National Health Insurance for 95 years, at least two million people in Taiwan have suffered from joint disease, and 92,581 patients (3.38%) have knee joint disorders. In addition, the total health insurance costs program in 2000 can be found. About 10,000 people have undergone artiﬁcial knee replacement surgery. The annual health insurance expenditure is 2.63 billion NT dollars, which is the second place in the hospitalization expenses. It is very impressive. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 457–462, 2019. https://doi.org/10.1007/978-3-030-02053-8_70

458

H.-I. Yin and F.-G. Wu

Most of the current treatment methods are minimally invasive or invasive. On the one hand, it does not completely reduce the deterioration of patients. On the other hand, it is expensive and will cause the patient to have a psychological burden and hesitation in the treatment. Especially, the elderly over 65 years old are afraid of the replacement of large joints by artiﬁcial joints. 1.2

Purpose

Due to the blood circulation and nerve conduction stimulation of knee osteoarthritis, its medical and prevention applications have not been explored. To improve the mechanism of cartilage degradation in knee joints, it is necessary to explore and summarize the design parameters by experiment. Therefore, the purpose of this study is as follows: (1) To explore whether blood circulation can effectively improve the rehabilitation effect of elderly people, and the relationship between body temperature regulation and knee joint activity angle. (2) To explore whether current stimulation can effectively improve the rehabilitation of elderly people, and how to effectively stimulate bone potential generation to facilitate bone growth and bursa synovialis’s secretion of cartilage fluid (3) To explore how blood circulation and current stimulation can be used to relief knee pain and propose innovative medical product design or rehabilitation treatment.

2 Literature Review 2.1

Blood Circulation and Knee Osteoarthritis

The blood circulation has two important functions. One is to provide the nutrients needed by the organization, and to carry away the metabolic products. The second is the conduction and convection of heat, which transfers heat from the inside of the body to the skin. Most of the blood flow in the skin’s blood circulation is to regulate body temperature, prevent body tissue from overheating, protect the internal organs of the body and maintain the homeostasis of physiological functions, so people can protect themselves from being too cold. In terms of bone quality, many studies have shown that blood plays a very important role in the regulation of bone metabolism, including the maintenance of posterior differentiation of articular cartilage and the secretion of synovial fluid from synovial tissue (Table 1).

Assistive Devices for Lower Limbs

459

Table 1. The summary of researches about blood circulation in knee osteoarthritis Plasma and blood cell

Synovial fluid

Blood flow increase

2.2

Research content The bone osteogenesis factor

Adrenocortical hormone and Norepinephrine White blood cell and Macrophage Osteoblasts and osteoclasts must be combined with antibodies from immune T cells at the beginning of their formation. The secretion of synovial fluid affects the differentiation of mesenchymal stem cells (MSCs) and the activation of macrophages, which are the key to the differentiation of osteoblasts. Core body temperature contributes to release nerve growth factor (NGF) which is widely recognised as a mediator of chronic pain. Increased the blood flow can improve the metabolism of the femur. Reduce muscle tension and neurological excitability Reduce muscle tension and neurological excitability Increased the blood flow can improve shearing of synovial fluid and activate TGF-b

Conclusion Bone osteogenesis The production of synovial fluid Bone osteogenesis Bone metabolism The production of synovial fluid Bone osteogenesis

Author Zhang et al. (2016)

Bone osteogenesis

Fahy et al. (2014)

Reduce the pain

Sagar and Nwosu (2016)

Bone metabolism

Kloet et al. (2013) Weerapong et al. (2005) Albro et al. (2012) Albro et al. (2012)

Reduce the pain Reduce the pain The production of synovial fluid

Grässel (2014) Orellana et al. (2016) Moradi et al. (2014)

Neurological Mechanisms and Knee Osteoarthritis

Over one’s lifetime, bone remodelling is tightly regulated through a complex network of hormones, cytokines and direct cellular interactions. At the local level, bone remodelling is guided by tightly regulated crosstalk between osteoblasts and osteoclasts.As bone formation and bone resorption cannot occur simultaneously at the same skeletal site, mechanisms of mutual inhibition of osteoblasts and osteoclasts have been hypothesized, to enable either bone formation or bone resorption at one speciﬁc site (Ohlsson 2013). Additionally, Semaphorin-3A is an osteoblast-derived secreted protein that increases bone mass, both via reduced osteoclast differentiation and increased bone formation (Table 2).

460

H.-I. Yin and F.-G. Wu

Table 2. The summary of researches about neurological mechanisms in knee osteoarthritis Research content SEMA3A improved implant osseointegration and ﬁxation in the proximal tibiae of ovariectomized rats The treatment of estrogens can effectively inhibit bone loss and cartilage thinning in ovariectomized rats When the synovial fluid is insufﬁcient, the inter-cartilage friction causes a rapid increase in nerve growth factor (NGF) inflammation, which leads to nerve proliferation and increased pain Two main OA determinants, mechanical stress and proinflammatory factors, including IL-1b and extracellular visfatin/NAMPT, stimulated the expression and the release of NGF by articular chondrocytes

Conclusion Semaphorin-3A reduced osteoclast differentiation and increased bone formation In osteoclasts, estrogens function to inhibit bone resorption activity and vitality rather than differentiation Sema3A can inhibit NGF and reduce the pain

Author Li et al. (2016)

Sema3A can inhibit NGF and reduce the pain

Pecchi et al. (2014)

Imai et al. (2009) Ashraf et al. (2012)

3 Method Knee osteoarthritis is known for its broad impact on both physical and mental levels. It is necessary to improve knee biomechanics to delay disease progression and improve function. However, traditional aids may bring negative impacts or discomfort duo to existing assistive method.Thus, the purpose of this research is to develop an innovative assisting device, which can relief painful symptom of KOA. Meanwhile, it should not restrict knee movement to prevent negative impact and increase comfort. 3.1

Study Process

This research is divided into three parts: (1) Study 1: Development of Design Criterions In this study, the current stimulation experiment will be ﬁrst carried out. The design parameters will be summarized from the experimental results. Furthmore, The design parameters、the blood circulation and the neurological mechanisms that learned from literatures will be discussed in expert focus group to understand more about their effects and applications. Also, literature reviews of existing assisting devices will be considered part of the design criterions. (2) Study 2: Design of Assisting Device The mechanism of knee osteoarthritis and design criterions acquired from study 1 will be used in co-design activity. Participants from design ﬁeld will discuss and generate innovative assisting device concepts. After that, concepts will converged and reﬁned into ﬁnal design and prototype.

Assistive Devices for Lower Limbs

461

(3) Study 3: Evaluation of Prototype In order to evaluate the efﬁciency of ﬁnal design, objective and subjective methods are used in experiments to assess whether pain is relieved and whether it is due to the blood circulation and the current stimulation.

4 Expected Results The expected results of this study is as follows: (1) Experimental results: Through the experiments, the relationship between body temperature regulation and knee joint activity angle and the effects on the nerve endings of the legs will be obtained. (2) Design parameters and guidelines: Provide design parameters and guidelines of the system model through the focus group with experts. (3) Innovative assistive device: According to design guidelines to develop the concepts and prototyping, the output in this phase of study including detailed sketches, rough model to ﬁne model. After repeated discussion and adjustment, the innovative assistive device will be proposed. (4) Design assessment: Considering human factors, safety regulations, and market analysis, design assessments and recommendations are proposed. Provides the research and development for the design of medical device systems.

References Albro, M.B., Cigan, A.D., Nims, R.J., Yeroushalmi, K.J., Oungoulian, S.R., Hung, C.T., Ateshian, G.A.: Shearing of synovial fluid activates latent TGF-b. Osteoarthr. Cartil. 20(11), 1374–1382 (2012) Ashraf, S., Mapp, P.I., Burston, J., Millns, P., Chapman, V., Walsh, D.A.: Increased sensitivity to nerve growth factor in the monosodium-iodoacetate model of osteoarthritis. Osteoarthr. Cartil. 20, S63 (2012) Fahy, N., de Vries-van Melle, M.L., Lehmann, J., Wei, W., Grotenhuis, N., Farrell, E., van der Kraan, P.M., Murphy, J.M., et al.: Human osteoarthritic synovium impacts chondrogenic differentiationof mesenchymal stem cells via macrophage polarisation state. Osteoarthr. Cartil. 22(8), 1167–1175 (2014) Imai, Y., Youn, M.Y., Kondoh, S., Nakamura, T., Kouzmenko, A., Mat-sumoto, T., et al.: Estrogens maintain bone mass by regulating expression of genes controlling function and life span in mature osteoclasts. Ann. N.Y. Acad. Sci. 1173(Suppl 1), E31–E39 (2009) Li, Y., He, D., Liu, B., Hu, J.: SEMA3A suspended in matrigel improves titanium implant ﬁxation in ovariectomized rats. J. Biomed. Mater. Res. Part B Appl. Biomater. (2016)

462

H.-I. Yin and F.-G. Wu

Moradi, B., Schnatzer, P., Hagmann, S., Rosshirt, N., Gotterbarm, T., Lorenz, H.-M., Tretter, T., Zeifang, F.: Accumulation of CD4+CD25+/highCD127low/- regulatory T cells in osteoarthritis joints –analysis of frequency and phenotype in synovial membrane, synovial fluid and peripheral blood. Osteoarthr. Cartil. 22, S449 (2014) Ohlsson, C.: Bone metabolism in 2012: novel osteoporosis targets. Nat. Rev. Endocrinol. 9(2), 72–74 (2013) Orellana, C., Calvet, J., Navarro, N., García-Manrique, M., Gratacós, J., Larrosa, M.: Higher synovial fluid white blood cell count in patients with knee osteoarthritis and metabolic syndrome. Osteoarthr. Cartil. 24, S84–S85 (2016) Pecchi, E., Priam, S., Gosset, M., Pigenet, A., Sudre, L., Laiguillon, M.C., Houard, X.: Induction of nerve growth factor expression and release by mechanical and inflammatory stimuli in chondrocytes: possible involvement in osteoarthritis pain. Arthritis Res. Ther. 16(1), 1 (2014) Kloet, R., Teule, G.J., Heyligers, I.C.: In vivo measurements of blood flow and bone metabolism in osteoarthritis. Rheumatol. Int. 33(4), 959–963 (2013). https://doi.org/10.1007/s00296-0122478-1 Sagar, D.R., Nwosu, L.: Dissecting the contribution of knee joint NGF to spinal nociceptive sensitization in a model of OA pain in the rat. In: Schaible, H.G. (ed.) Joint Neurophysiology and Pathophysiology: Nerves, Receptive Fields, Sensitization (2016) Grässel, S.: The role of peripheral nerve ﬁbers and their neurotransmitters in cartilage and bone physiology and pathophysiology. Arthritis Res. Ther. 16, 485 (2014). https://doi.org/10.1186/ s13075-014-0485-1 Wanless, D., Forder, J., Fernández, J.L., Poole, T., Beesley, L., Henwood, M., Moscone, F.: Wanless social care review: securing good care for older people, taking a long-term view. King’s Fund (2006) Weerapong, P., Hume, P.A., Kolt, G.S.: The mechanisms of massage and effects on performance, muscle recovery and injury prevention. Sports Med. 35(3), 235–256 (2005) Zhang, W., Likhodii, S., Aref-Eshgh, E.: Relationship between blood plasma and synovial fluid metabolite concentrations in patients with osteoarthritis. J. Rheumatol. (2016)

Packaging Design to Support Small Business Enterprises in the Republic of El Salvador Alberto Rossa-Sierra1(&), Maria Giovanna Trotta1, Fabiola Cortes-Chavez1, and Francisco González-Madariaga2 1

2

Universidad Panamericana. Facultad de Ingeniería. Prolongación Calzada Circunvalación Poniente 49, 45010 Zapopan, Jalisco, Mexico [email protected] Centro Universitario de Arte, Arquitectura y Diseño. Universidad de Guadalajara, Guadalajara, Mexico

Abstract. As part of the “Program for Economy Promotion and Employment in Central America”, sponsored by German Agency for Technical Cooperation (GIZ), a new mechanism has been created to contribute to the goal of strengthening the competitiveness of micro, small and medium enterprises (SME) in Central America. The ﬁrst phase of the program (2013–2015) was carried out in Guatemala, El Salvador, Honduras, Nicaragua and Costa Rica, and was used to ﬁnance technical assistance on innovative projects in the area of economic development and employment, with an emphasis on youth and women. One of the projects supported by the program was titled “Industrial design and digital manufacturing for better employment of young people in El Salvador and Costa Rica”, one of the goals of the project was to carry out a workhop with young students, developing interventions in the design and development of packaging for small local food production companies in El Salvador. The work developed by the design consultants, consisted in elaborating the didactic material, learning objectives for a course in virtual mode and a presential course where different applications of packaging were developed with SME located in San Salvador. This paper presents the results obtained during the development of the project, the visual and structural analysis, and the new design proposals of the previous packaging used by the SME, as well as a description of the didactic material developed and the evaluation of the intervention. Our research, in accordance with our work experience in the area, presents our economic proposal for the module “Fundamentals of packaging technology”, being approved by the organizing entities. The proposal presented involves the following activities: (a) Development of learning material (b) Selection and coaching of the required trainers (c) Development of the virtual diploma program, in the corresponding module (d) Development of packaging examples for SMEs of the region. It should be noted that the coaching of the people involved, required a stay at Don Bosco University for supervision and application of the virtual diploma module for 3 weeks. Keywords: Packaging design

Small enterprises Central America

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 463–468, 2019. https://doi.org/10.1007/978-3-030-02053-8_71

464

A. Rossa-Sierra et al.

1 Introduction In Central America, an emerging niche has been found in the market for industrial design and digital manufacturing. The introduction of these new training services, aimed mainly at young people, opens possibilities to improve manufacturing processes at low cost for small and medium companies in the region; where there is also a great opportunity to innovate and develop new useful products. The “Facilidad” Program is executed by the GIZ (German Agency for Technical Cooperation) in cooperation with the General Secretariat of the Central American Integration System (SG-SICA) and Promotion Center of Micro and Small Enterprises in Central America (CENPROMYPE). Under the modality of fund competitions for technical assistance; The FACILIDAD Program supports innovative projects involving academic, public and private sectors, in order to improve the competitiveness of micro, small and medium enterprises in the region and offer better employment opportunities, with emphasis on young people between 15–24 years. The project was presented for evaluation and granted to “Don Bosco University” (UDB) in El Salvador in conjunction with “Don Bosco Salesian Education Center” (CEDES) in Costa Rica, since, to be accepted this kind of projects, they must be supported by at least two Central American entities located in different countries. 1.1

Project Objective

The appearance of this new ﬁeld with great potential results in this kind of project, who develops a new training program in industrial design and digital manufacturing. The training program aims to generate skills in young people between 15 and 24 years of age, through 7 training modules in the areas of: Creativity and innovation, Concepts for product design, Design of new products, CAD-CAM-CAE applications, Simulation of production processes, Basics of packaging and packaging and Development of entrepreneurial characteristics It is intended that future trained young people have the opportunity to enter the labor market or create their own companies. 1.2

Methodological Approach

The main elements of the methodological approach are: • • • • • •

Preparation and Implementation at Don Bosco University Development of the required Teaching Materials Selection and training of the required trainers Call and selection of young participants Development of the virtual diploma program Product development for SMEs

Preparation and Implementation at CEDES (Costa Rica) • Experts of Don Bosco University at El Salvador replicate the training process at CEDES in Costa Rica

Packaging Design to Support Small Business Enterprises

465

Project Sustainability • Formalization of alliances with private sector • Development of a strategy to manage support through current projects executed under Public Private Partnership (PPP) modality • Continuity of the virtual diploma Project Conclusion • Insertion in industrial design labor market and products through the employment exchanges of the executing entities and participants • Midterm and ﬁnal project evaluation • Documentation of successful cases of young people trained and inserted in labor market 1.3

Expected Results

(a) The diploma is available in virtual format to be taught that modality (b) 16 instructors trained in industrial design and product development themes (8 in UDB and 8 in CEDES) (c) 80 young people trained in industrial design and product development themes (40 in UDB and 40 in CEDES) (d) Alliances with private sector and the public sector necessary to achieve the results of the project have been formalized (e) 48 new jobs generated in the area of industrial design and product development, from which 60% of young graduates of Program have been inserted into the labor market or self-employed with the support of CEDES and UDB (f) A total of 10 Industrial and Product Design projects have been presented, with their respective business plans in Costa Rica and El Salvador (10 projects: 5 in UDB and 5 in CEDES) with the potential to generate a business opportunity and self-employment. It is within this great framework of the “Facilidad” Program that Don Bosco University as coordinating entity turned to the task of search in Latin America specialists in the seven modules in which diploma was divided, and that was: 1. 2. 3. 4. 5. 6. 7.

Creativity and innovation Concepts for product design Design of new products CAD-CAM-CAE applications Simulation of production processes Fundamentals of packaging Development of entrepreneurial characteristics.

466

A. Rossa-Sierra et al.

Our intervention, in accordance with our work experience in the area, we present our economic and work proposal for the module “Fundamentals of packaging technology”, being approved by the organizing entities. The proposal presented involved the following activities: (a) (b) (c) (d)

Development of learning material Selection and training of the required trainers Development of the virtual diploma program, in the corresponding module Development of packaging examples for SMEs in the region.

It should be noted that the training of the people involved, requires a stay at Don Bosco University for supervision and application of the virtual diploma module for 3 weeks.

2 Development of the Training and Support Project for SMEs One of the project premises was support for SMEs companies in the region, so we search for local producers who want to improve or update their packaging, in order to improve the presentation of their products, a task that was achieved by recruiting three small food-producing companies in El Salvador: A flour producer, a honey-product cooperative and a coffee-growing company. In all three cases, brieﬁngs were prepared, which are deﬁned as design programs, which are written documents that gather all necessary information so that a design project can be correctly developed. All-time devoted to writing, in the most detailed way possible, the brieﬁng should be considered well spent since it will save time in the development of the project and encourage a well-oriented creativity (AIDO [1]). These brieﬁngs contained the information related to the company and where they wanted to guide the communication strategy through the packaging. Examples of the brieﬁngs are shown in Fig. 1. Once the projects were established, the participants, following the methodology proposed by the trainers, started developing the initial ideas (Fig. 2). The ﬁnal proposals were exposed to micro entrepreneurs for their approval, some of the proposals made during the training are shown below (Figs. 3 and 4).

Fig. 1. Examples of developed brieﬁngs for the packages to be redesigned.

Packaging Design to Support Small Business Enterprises

467

Fig. 2. Methodological process for packaging design and development [2].

Fig. 3. Proposal of packaging re-design for gourmet line flour. Design by Gabriela García

Fig. 4. Proposal design for “Coscafé”® brand in cardboard box, design by Eneida Cruz.

As you can see in the preceding images, we tried to cover different markets, companies, and types of packaging. In addition to the design exercises applied to local companies, it was part of the work team’s task to prepare the contents for a platform that allows taking the diploma courses at distance learning platform, for which multimedia content, readings and activities were created. Actually, they are already hosted on Don Bosco University servers.

468

A. Rossa-Sierra et al.

Additionally, an accompanying text was developed for the teaching of diploma courses, which consists of 140 pages divided into 10 chapters, this instrument has been highly valued by the students, and actually, a textbook is currently in progress and preparation for a professional edition [3].

3 Conclusions As you can see in the presented proposals, the intervention of design professionals ostensibly improves the presentation and attractiveness of the product, which, in turn, should translate into a sales increase and positioning of regional companies in the Central American market. This type of small improvements that do not economically impact the product, it is possible to perform them periodically with the intention of maintaining the brand image of the SMEs. This type of action should be carried out without fear, just as the large transnational food companies do. The experience would be very useful to transfer it to our national context, grouping small entrepreneurs demonstrating the beneﬁts of this type of actions, both for design profession, and the survival and growth of small production companies. Finally, it must be recognized that all this work has been carried out within the framework of the support of German Federal Government, with an effort of international cooperation, promotes this type of action in the less developed countries, and the better, with the participation as facilitators of regional specialists, which strengthens the cooperation among the participating institutions. As an example of this, we can remark that, derived from the participation in this project, complementary and additional actions were derived, such as the creation of an Ibero-American network in industrial development area for innovation on digital manufacturing for SMEs, which has recently been supported by the Ibero-American Program of Science and Technology for Development (CYTED) with funds to work during the next three years (2017– 2020), and which we hope will soon become available of it.

References 1. AIDO: Guía de buenas prácticas para diseñadores de productos industriales impresos. Instituto Tecnológico de Óptica, Color e Imagen, Valencia (2009) 2. Rosa, A.: Marca, envase y embalaje para PyMES. Editorial Universitaria, Universidad de Guadalajara, Guadalajara (2009) 3. Rosa, A., González, F.: Fundamentos de Envase y Embalaje. Editorial del Sindicato de Trabajadores Académicos de la Universidad de Guadalajara, Guadalajara (2016)

The Importance of Industrial Design in Medical Devices in the 21st Century Fanny Guadalupe Valdivia-Márquez(&), Pilar Hernandez-Grageda, Gabriela Durán-Aguilar, and Alberto Rossa-Sierra Universidad Panamericana, Facultad de Ingeniería, Prolongación Calzada Circunvalación Poniente 49, 45010 Zapopan, Jalisco, Mexico [email protected]

Abstract. The medical device business is strategic both for its high international competitiveness and its potential impact. Mexico has participated in this business mainly from the manufacturing sector, but if it seeks as a country to achieve a greater participation at the design trends of medical devices, it will be of great importance to consider the role of Industrial Design as a factor to improve the user’s quality life through innovative products. This will be the result of a multidisciplinary work among professionals of Biomedical Engineering and Industrial Design, being conscious that the better the designer knows the user that will interact with the product -with support from Ergonomics and Psychology-, better decisions will be made in the process of design and as a result, globally competitive results will be obtained, with technologically innovative products based on proposals that meet the user’s needs, triggering favorable economic results for the industry and improving the global positioning of the country. Keywords: Medical design Industrial Design Ergonomics in design Aesthetics function

Psychology in design

1 Introduction The Secretariat of Economy, the Boston Consulting Group and the federal agency ProMexico, at their study Diagnosis of advantages and limitations for attracting Foreign Direct Investment to México, identiﬁed the productive sectors and sub-sectors on which México has competitive advantages, compared to other countries, for the attraction of investments. The combination of the health services sector and the electronic sector shows that the medical devices business is strategic because of its current high competitiveness and the potential impact it represents. In the last decade, the medical devices industry in México has presented a dynamic and sustained growth, from the development of manufacturing capabilities by the main companies in the ﬁeld, who have found in the country a key partner for their strategies of investment and business. México has become the main supplier of medical products for the most important market in the world: United States. Currently, México is the ﬁfth largest exporter of medical devices in the world, the second largest market in Latin America [1].

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 469–474, 2019. https://doi.org/10.1007/978-3-030-02053-8_72

470

F. G. Valdivia-Márquez et al.

2 Development of the Topic There are enough resources to maintain Mexico as an innovative and expert country in the medical industry. But what does the industry in Mexico need to become a global reference, not only in the assembling and manufacturing sector? Which position do Mexican designers and engineers keep in relation to this growing market? As a country aspiring to be an international benchmark both as a producer and a developer in the sector, could Mexico develop the best possible infrastructure? To answer these questions, we will look at the importance of design in the medical industry. Industrial Design links innovation, technology, research, business, and customers to provide new value and competitive advantage across economic, social, and environmental sphere [2]. Industrial Design is a trans-disciplinary profession with strategic problem-solving processes that drive innovation, build business success, and lead to a better quality of life for the user through innovative products, systems, services, and experiences. As a result, the user should obtain a product that is not just useful for a speciﬁc task, but that has taken into account the user’s needs in its everyday life. When we talk about medical devices design, several users are detected. There is the medical staff, who are the ones who usually interact actively with the medical product; there are also the patients, on whom the function of the product falls. The medical staff are the ones who will decide the acquisition of a new product because of their technical knowledge, but the patient/user is the one who will decide if the new product is the one that covers most of its needs. As Peter J. Ogrodnik mentions on his book Case studies in Medical Devices Design: “the work of a good designer with his product is not to force an ideal, but to generate an idea that was there waiting to come out. If he succeeds, the person will grow up to become his faithful user. If it were a meal, I would describe the taste, the color, the temperature, the ethnic origins, etc. However, he has not declared the ingredients or how they can arm themselves, that’s the menu, his job is to ﬁnd the menu that satisﬁes the hungry customer” [3]. Analogically, at the case we are studying, the designer sets the values, while the chef is the expert on the menu. For medical devices design, the main ﬁgures are the biomedical engineer and the industrial designer. The ﬁrst provides the technical part, while the second integrates the ergonomic and psychological part to ensure a positive impact on the user, without neglecting the function or aesthetics of the medical device. The task of a biomedical engineer is to improve patient care by solving problems in medicine and biology. This ﬁeld of engineering requires professionals to use advanced technology and combine knowledge from many disciplines and sources. A biomedical engineer often works with technicians, doctors and nurses as part of a team [4]. 2.1

The Impact of the Ergonomics Considerations on Medical Devices Design

There has been an important effort on boosting Mexico as a main actor on the medical devices and advanced technology sector, and the starting point of this effort is a huge choice for a science, technology and materials based innovation, which seeks to earn much better practices on product manufacturing. Nevertheless, it is essential to know that no matter how much the innovation processes evolve, it will never be able to forget

The Importance of Industrial Design in Medical Devices

471

about the user of the devices as the main - and most important- reason of any other effort. In other words, a medical device only exists since it will be a solution for a speciﬁc need of users with certain characteristics. When designers want to get to an optimal knowledge of the ﬁnal user of their products, they can rely on Ergonomics - or Human Factors -, the discipline that establishes the best practices, considerations and methods to make analysis and evaluations of users and environment; the conclusions obtained by these analysis and evaluations can give the designers the best recommendations that assure that every designed product will optimize the user’s wellbeing and that it will meet the task for which it was designed. Using a systemic approach, this discipline assures that the designers take into account some important elements for the product they are developing: the user that interacts with the product, the task that the product must play and the environment where the task will be played. Regarding the user, it is important to analyze its physical, social, physiological and psychological characteristics, in order to get as much information possible about the ﬁnal user. As for the task, it is important to study the complexity that the user will face when interacting with the product pretending to accomplish a job, identifying the actions the user has to make, the decisions implicated on these actions and the attention they require from the user to be accomplished. Finally, the environment includes the physical (things), organizational (activities) and social (culture) characteristics that, together, will make the user interact in a speciﬁc way with the product while trying to perform a particular task. This way, we can understand the clear difference in the result of a medical device design process if the aforementioned elements are taken into account. Actually, there may be different circumstances of use for the same device; for example, an insulin pump manufacturer who wants to get its product to adult patients in Germany that have advice in a private clinic, as well as to adolescent patients in Latin America who are looking for an active routine, will understand that it cannot offer a product with identical characteristics for both situations, and that the characteristics that it implements in its medical devices to meet the particular needs of users will be a consequence of the Ergonomics methodologies and tools considered. The fact that a medical device overcomes laboratory tests and in a controlled environment has proven to be safe and effective, does not automatically make it an equally safe and effective product for the real world [5]. Ergonomics has the potential to act through design, ensuring that users do not get hurt by using certain products or receiving a service; that products or services are proﬁtable for companies and for society, and that they are accessible to people with different capabilities and aspirations [6]. In addition, Ergonomics can contribute to the innovation of a product from the analysis of its usability and user experience, from the observation of the way in which people use it and the detection of existing problems and needs. Speaking of medical devices, the variables that can be evaluated ergonomically include how effective they are, ease of use, comfort and level of acceptance [7]. At this point we can make a clear idea that the ergonomics considerations have a very important weight in the early stages of the design process for medical devices, which is necessary to understand thoroughly the system in which the new product is immersed - people, tasks and environment - to determine the orientation of the design

472

F. G. Valdivia-Márquez et al.

and the decisions that are generated to achieve the best result. However, ergonomic evaluations do not only participate in the early stages of the design process, but they add great value to the ﬁnal result when they become to a control medium as it moves towards the ﬁnal product, generating a better experience for the user, less familiarization time with the medical device, better adaptation for its use, fewer errors during use and, therefore, greater efﬁciency of the medical device in the real world. The consequences of not taking seriously the ergonomic considerations in the user-product interaction include error in the use, little hooking with the product and even complete abandonment of it. The aforementioned is particularly important for medical devices with which the user interacts directly [5]. 2.2

The Impact of the Psychological Considerations on Medical Devices Design

The Strategy 3.3 of the Sector Development Program 2013–2018 talks about “guaranteeing the quality, safety and efﬁcacy of medicines, biologicals and health supplies” [8]. It is on this last aspect where it becomes essential to reflect on medical design, ergonomics and also, undoubtedly, the impact that the Psychology of the user has on achieving the effectiveness of the device. It should also be noted that the issue of effectiveness and risk is directly associated with a basic component of Psychology: user perception, as mentioned by the United States Government on its report “it is worth reiterating that when researching approaches to altering health behavior, it is not the current risk and actual efﬁcacy, but the perceived risk and perceived efﬁcacy, that drive an individual’s actions. In many cases an individual’s perception of risk and efﬁcacy are very different than his or her current risk and efﬁcacy [9]”. Privitera [10] indicates “in psychosocial modeling the phrase perceived risk is used to refer to an individual’s belief on how detrimental (or beneﬁcial) to given course of action is to his or her health. The phrase perceived effectiveness refers to an individual’s belief in how much control he or she has over given situation [10]”. From this dynamic between the real and the perceived, a natural gap is derived, in such a way that considering an interdisciplinary approach in the design of medical devices between Psychology, Ergonomics and Design in itself, will shorten that distance. In the ﬁeld of health care, two of Psychology’s objectives are [11]: 1. Promotion of health: it deals with all those actions aimed at promoting a healthy lifestyle, understanding health not only the absence of disease but the presence of a state of physical and mental well-being in the individual. 2. Evaluation and treatment of speciﬁc diseases: such as cancer, AIDS, etc. In these cases, psychology works together with other health specialists combining the usual medical treatments along with the psychological aspects. Engel [12] proposed an alternative model to the biomedical, named biopsychosocial model. It hypothesizes the importance of considering psychological, social and cultural factors along with biological factors as determinants of the disease and its treatment: “The medical model must take into account the patient, the social context in which he lives, and the complementary system used by society to deal with the detrimental effects of the disease, that is, the role of the doctor and the health care

The Importance of Industrial Design in Medical Devices

473

system”. As the knowledge of the strong relationship of the user’s psychological part advances in the medical design ﬁeld, speciﬁcally on its affective and cognitive characteristics, the consideration of aspects of the emotional design in the devices arises, since they influence the increase of the patient’s adherence by generating greater efﬁcacy in the treatment; hence the great interest in this link that is essential to develop devices that really improve the quality of life of users, an optimal objective for a designer. Another relevant psychological factors for the designers to take into account are motivation and desires. From the emotional, cognitive and personality point of view, a device with a self-expressive advantage, that is, something that allows the user to identify with it and, through it, expresses who he wants to be, or at least does not inhibit self-expression, will make the user more willing to use it. Users differ from each other in terms of personality as well as of physical, emotional, cognitive and socio-cultural qualities; these qualities are considered the determining factor when choosing and using medical devices. When interacting with a device, the ﬁrst thing the user does is visual contact, from which various emotions are raised, mainly of acceptance or rejection, depending on what they perceive and think about it (Visceral design). The user makes connections and associations at that moment, even when information is missing, this includes how he interprets and understands the product, how easy or difﬁcult it is to use; It is the moment of empathy and that emotionally will impact on the habit of use (Conductive design). An information ﬁlter is also carried out where the user’s preferences and cognitive abilities influence. Experiments have shown that we can only retain 3 or 4 pieces of information in working memory (Reflective Design) [13]. Personalizing the medical device, understanding the impact of the sensations, perceptions, emotions, thoughts, and ideals of the users in the use of the devices, promotes its efﬁciency and adds value to the experience of use.

3 Conclusion It is ideal to work with multidisciplinary teams all along the innovation, design and development process of medical devices in order to reach a ﬁnal product that is globally competitive, technologically innovative and, most of all, formed around the user’s capabilities and aspirations. There are several cases that demonstrate how the design of medical devices is determinant in the way that they will be used. The ideal objective for companies in Mexico and anywhere in the world that carry out research and development of new medical devices, is that their products obtain the best possible acceptance by users, which will trigger a series of favorable economic and positioning results for the business. The design of medical devices has a lot to do with the incorporation of features that make them easy to understand and use. To do this effectively, designers must understand the perception, the emotions, the users’ cognition and their socio-cultural aspects. By knowing how the user perceives and interprets information in the world around them, medical product designers can provide signals that allow devices to operate intuitively, reliably and safely.

474

F. G. Valdivia-Márquez et al.

References 1. Sandoval Ríos, M., Carreón Sánchez, M., Ortíz Porcayo, D., Moreno Blat, M.: Diseñado en México Mapa de ruta de dispositivos médicos. PROMEXICO (2011) 2. IDSA: World Design Organization. IDSA (2015). http://www.idsa.org/news/dblog/what-id 3. Ogrodnik, P.: Class 1 Devices Case Studies in Medical Devices Design. Elsevier, London (2015) 4. Farnen, K.: ¿Qué hace un ingeniero biomédico?. La voz de Houston (2011) 5. Lang, A., Martin, J., Sharples, S., Crowe, J.: The effect of design on the usability and real world effectiveness of medical devices: a case study with adolescent users. Appl. Ergon. 44, 799–810 (2013) 6. A strategy for human factors/ergonomics: developing the discipline and profession. Ergonomics 377–395 (2012) 7. Jacobs, K.: Ergonomics for Therapists. Mosby Elsevier, St. Louis (2008) 8. Organización Mundial de la Salud: Organización Mundial de la Salud, Noviembre 2013. http://www.who.int/mediacentre/factsheets/fs384/es/ 9. U.S. Goverment: Modelling in healthcare/Complex Systems Modelling Group. American Mathematical Society, United States of America (2010) 10. Privitera, M.: Biomarker strategies SnapPath® case study: Design research program to integrate a breakthrough diagnostic system in to the clinical environment. Contextual Inquiry for Medical Device Design, nº 211–227 (2015) 11. Oblitas, L.: Enciclopedia de psicología de la salud. PSICOM, Bogotá (2007) 12. Engel, G.: The need for a new medical model: a challenge for biomedicine. American Association for the Advancement of Science 13. Donald, N.: El diseño emocional: por qué nos gustan (o no) los objetos cotidianos. PAIDOS, Barcelona (2005)

IHSED 2: Innovative Materials for Product Development

Prevention of Work: Related Musculoskeletal Disorders Using Smart Workwear – The Smart Workwear Consortium Carl Mikael Lind1,2(&), Leif Sandsjö3,4, Naﬁse Mahdavian5, Dan Högberg5, Lars Hanson5,6, Jose Antonio Diaz Olivares2, Liyun Yang1,2, and Mikael Forsman1,2 1

Unit of Occupational Medicine, Karolinska Institutet, Stockholm, Sweden [email protected] 2 Division of Ergonomics, KTH Royal Institute of Technology, Huddinge, Sweden 3 Faculty of Caring Science, Work Life and Social Welfare, University of Borås, Borås, Sweden 4 Design & Human Factors/Department of Industrial and Materials Science, Chalmers University of Technology, Gothenburg, Sweden 5 School of Engineering Science, University of Skövde, Skövde, Sweden 6 Scania CV, Södertälje, Sweden

Abstract. Adverse work-related physical exposures such as repetitive movements and awkward postures have negative health effects and lead to large ﬁnancial costs. To address these problems, a multi-disciplinary consortium was formed with the aim of developing an ambulatory system for recording and analyzing risks for musculoskeletal disorders utilizing textile integrated sensors as part of the regular workwear. This paper presents the consortium, the Smart Workwear System, and a case study illustrating its potential to decrease adverse biomechanical exposure by promoting improved work technique. Keywords: Ergonomics Human factors Human-Systems integration Work technique Smart textiles Musculoskeletal disorders Prevention

1 Introduction Work-related musculoskeletal disorders (WMSDs) are common in the working population globally. In the European Union, about 20 million workers report work-related health problems annually, of which 2/3 of the problems concern the musculoskeletal system [1]. Major physical risk factors for WMSDs include e.g. manual handling, frequently adopting non-neutral postures such as bending and twisting of the trunk or upper arm elevation, and repetitive movements of the trunk and upper extremities [2, 3]. Apart from directly affecting the individual worker’s health and economy, WMSDs lead to large costs for the health care system and also premature exits from the labour market. The ﬁnancial cost for musculoskeletal disorders has in the Nordic Countries been estimated to be approximately 3–5% of the gross national product [4]. Additionally, several studies support that poorly designed workplaces and work tasks increase the rate © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 477–483, 2019. https://doi.org/10.1007/978-3-030-02053-8_73

478

C. M. Lind et al.

of quality deﬁciencies, resulting in increased costs for employers [5, 6]. Hence, to reduce the associated costs and reduce the number of premature exits from the labour market, the working conditions need to be improved in physically demanding jobs. To reduce WMSDs, multi-component interventions are usually advocated [7]. Studies from different labour market sectors support that interventions that include risk assessment can have positive effects on reducing adverse health effects such as WMSDs [8, 9]. Today, occupational health services personnel, consultants and ergonomists assess risks for musculoskeletal disorders by using direct or in-direct observation based methods. Assessments based on observation methods have, however, been shown to lack in validity [10], render high costs if repeated measurements are required, and require competent and experienced assessors. Therefore, new, valid, reliable, and user-friendly methods that can be used to improve assessment of WMSDs risk factors and effectively communicate the results from the assessment need to be developed. In this paper, a multi-disciplinary consortium with the aim of developing an ambulatory system that can be used to assess biomechanical exposures and communicate the results is presented. Further, the Smart Workwear System developed by the consortium is presented. Finally, the capacity of the developed Smart Workwear System to influence operators’ work technique is demonstrated in a group of manufacturing operators.

2 The Smart Workwear Consortium In this project, funded by Sweden’s innovation agency VINNOVA (Vinnova/UDI 201603782), a multi-disciplinary consortium has joined forces to develop a system for recording and analyzing risks for musculoskeletal disorders, based on smart textiles as part of the regular workwear. The lead partner of the project is Hultafors Group. In their role as a leading manufacturer of workwear they play an important part in the design for usability and producibility of the resulting smart workwear solutions, but also in the manufacturing of the prototypes. The industrial partners (Scania CV, Volvo Trucks and Volvo Cars) are the problem owners as employers are responsible for workers’ safety and health, and, thus, represent future users of the project results. The occupational health service providers (Feelgood and Avonova) provide knowledge and experience about WMSD issues in current industrial production and how to intervene when problems appear and, thus, represent the future users of the developed solution. The academic partners and research institutes (Fraunhofer-Chalmers Centre, Karolinska Institutet, KTH Royal Institute of Technology, Swerea IVF, University of Borås, University of Gävle and University of Skövde) cover different roles in developing, testing, validating and documenting the solutions and potential outcome. The academic partners and research institutes jointly build up a team that fundamentally consists of three established research areas: smart textiles, ergonomics, and system development. University of Borås and Swerea IVF have a strong tradition within textile and ﬁbre research. KTH, University of Gävle and University of Skövde are established groups within ergonomics, physiological measurements and processing of physiological signals. Karolinska Institutet and Fraunhofer-Chalmers Centre have a history of developing effective and userfriendly mobile device applications and software for ergonomics assessments.

Prevention of Work: Related Musculoskeletal Disorders

479

3 The Smart Workwear System The proposed system is based on sensors and electronics integrated in textile garments in order to gain information on heart rate, breathing, postures, movements and forces exerted. The recorded signals are analyzed and the risks assessed in real time using research-based assessment criteria (e.g. [3, 11]) or speciﬁc assessment methods developed by the participating companies. The real-time analyses and risk assessments allow results to be communicated directly to the worker, e.g. via an audio device from a set of automated verbal instructions, here called the “virtual coach” scenario or to an ergonomist or instructor for coaching the worker, here called the “human coach” scenario (Fig. 1). This way the system allows assessment of individual worker’s movement patterns in order to provide feedback on appropriate work techniques and is believed to be very useful both to new employees, learning the job, and more experienced workers. In addition to physically strenuous work activities, also sedentary or other types of low intensity work activities that induce negative health effects can be assessed using this technology. On top of being used for real-time analysis and feedback, the recorded

Fig. 1. System overview of the Smart Workwear System illustrating data flow in terms of “sensor” and “ergonomics” information and the two main use scenarios: (i) the “human coach” where an instructor (the coach) interacts with the operator (i.e. the worker) to give feedback on the performed work based on on-line analysis of sensor data presented to the instructor from the BSN server by means of a tablet or laptop, and (ii) the “virtual coach” where audio feedback is automatically provided to the operator based on automated analysis of sensor data communicated by means of a mobile device.

480

C. M. Lind et al.

sensor signals are transferred to a server for long-term storage to enable ergonomics evaluation on aggregated data on both individual and group level. This way, risks from work behaviour and workplaces can be assessed over time, and the need for improvements of work and workplaces can be visualized. The ergonomics analysis methods used in the system can be similar to those used within digital human modelling (DHM) tools. The approach of utilizing the same methods for ergonomics assessments bridges the gap between the design phase and the operative use phase of workplaces, and facilitates two-directional knowledge transfer between design and operations. The consortium’s long-term aim is to develop a commercially available workwear product and risk assessment system for the global market.

4 Case Study A prototype of the Smart Workwear System was tested for its capacity to decrease exposure to adverse postures and movements of the upper arms and trunk. The prototype system comprised a workwear T-shirt with integrated inertial measurement units (IMUs) to measure upper arm elevation angles and movements, trunk flexion and extension. The test was carried out in six separate workdays in a manufacturing plant in Sweden on newly employed manufacturing operators attending a three-day training program [12]. Twenty-four operators participated in the test on their last day of the training program, in which they worked a full workday on one of four assembly line workstations. The participants were randomized into three scenario groups: (a) control scenario (control group), (b) human coach scenario and (c) virtual coach scenario, each comprising eight subjects. Each participant was monitored for about a half workday starting after their regular lunch break. Nine of the participants were women and ﬁfteen were men. The mean (SD) age, stature, and body mass were: 28 (10) years, 171 (10) cm, and 70 (13) kg respectively. The ﬁrst two work cycles after the lunch break served as baseline for each subject. To test the system’s capacity to guide the operators to decrease biomechanical exposure, the median number of seconds in exposure equivalent to, or exceeding the exposure criteria for the ﬁrst two work cycles (baseline), was compared with the median number of seconds for the last two work cycles. The improvements for the human coach scenario group and virtual coach scenario group were compared with the improvements for the control scenario. For the control scenario, the participants received normal feedback from an instructor without any aid of the Smart Workwear System. For the human coach scenario the participants received feedback from the instructor who used the feedback from the Smart Workwear System to instruct the participants. In the virtual coach scenario, in addition to receiving normal feedback from the instructor (who did not receive aid of the Smart Workwear System), the participants received audio feedback from the Smart Workwear System via wireless earphones. The audio feedback instruction was given after each participant had ﬁnished the work cycle. The audio feedback comprised a voice message which informed the

Prevention of Work: Related Musculoskeletal Disorders

481

worker of the “highest risk”, e.g. “Right arm. Please aim at lowering your right elbow”. The study was approved by the Regional Ethical Review Board in Stockholm (2017/1586-31/4) and all participants provided written informed consent prior to participating. Descriptive statistics were calculated using SPSS (version 19). As shown in Fig. 2, descriptive data from the initial analysis of the test indicates a reduction in the median number of seconds of exposure for both the left and right arm, equivalent to, or exceeding the exposure criteria when using the system’s virtual coach mode. For both the right and left upper arm, the largest reduction was observed for the virtual coach scenario. For trunk flexion, the test indicates the largest reduction in the number of seconds equivalent to, or exceeding the exposure criteria for the human coach scenario. For trunk extension, a tendency for increase in exposure was observed only for the control scenario. Median seconds difference in exposure 40 20 0 -20 -40

Right Right Left upper Left upper Trunk -60 upper arm upper arm arm ≥30° arm ≥60° flexion ≥30° ≥60° ≥20° -80

Trunk flexion ≥45°

Trunk extension ≥10°

-100 Control scenario

Virtual coach scenario

Human coach scenario

Fig. 2. Median seconds difference in time equivalent to, or exceeding the exposure criteria for the three scenarios. Negative scores denote improvements (reduction in exposure) and a detected change in work technique.

5 Discussion and Final Remarks The Smart Workwear Consortium represents a unique assembly of stakeholders, where each of the partners can contribute their key expertise to the multi-disciplinary actions needed to address WMSDs in industrial production. The proposed Smart Workwear System provide a platform for ergonomics assessment and intervention both on individual and group level with large potential both to employers and the research community to address WMSDs issues in industrial production. The performed case study points to a reduction in adverse biomechanical exposure associated with the use of the smart work wear system for the upper arms. Due to the relatively small study population, more studies are needed to conﬁrm if this trend can be attributed to the system or other factors such as individual learning effects [13]. Further studies are also needed to test if the feedback can be improved by other factors, such as, how frequently the feedback is given as well as other types of feedback modalities, e.g. vibrotactile feedback.

482

C. M. Lind et al.

The Smart Workwear System opens up a new market for advanced services. Risk assessments can be largely automated which may improve both the quality of the assessments and lower the costs. The potential savings of costs for the society, the companies and the individuals is very high, and so is the potential to improve workrelated health. Acknowledgments. The Smart Workwear Consortium is funded by Vinnova, the Swedish Innovation Agency, under the call Challenge Driven Innovation (Vinnova/UDI 2016-03782), and by the participating organizations. The Smart Workwear Consortium partners are Hultafors Group; Avonova; Feelgood; Fraunhofer-Chalmers Centre; Karolinska Institutet; KTH Royal Institute of Technology; Scania CV; Swerea IVF; University of Borås; University of Gävle; University of Skövde; Volvo Trucks and Volvo Cars.

References 1. European Commission: Statistics in Focus. Ofﬁce for Ofﬁcial Publications of the European Communities, Luxembourg (LU) (2009) (Eurostat publication ISSN 1977-0316) 2. National Research Council: Musculoskeletal Disorders and the Workplace: Low Back and Upper Extremities. National Academy Press, Washington (DC) (2001) 3. Lind, C.M.: Assessment and Design of Industrial Manual Mandling to Reduce Physical Ergonomics Hazards – Use and Development of Assessment Tools [dissertation]. KTH Royal Institute of Technology, Stockholm (SE) (2017) 4. Winkel, J., Westgaard, R.H.: Editorial: a model for solving work related musculoskeletal problems in a proﬁtable way. Appl. Ergon. 27(2), 71–77 (1996) 5. Falck, A.C., Örtengren, R., Högberg, D.: The impact of poor assembly ergonomics on product quality: a cost-beneﬁt analysis in car manufacturing. Hum. Factors Ergon. Manuf. 20 (1), 24–41 (2010) 6. Eklund, J., Yeow, P.: Integrating ergonomics and quality concepts. In: Wilson, J.R., Sharples, S. (eds.) Evaluation of Human Work, 4th edn, pp. 931–956. CRC Press, Boca Raton (2015) 7. Silverstein, B., Clark, R.: Interventions to reduce work-related musculoskeletal disorders. J. Electromyogr. Kinesiol. 14(1), 135–152 (2004) 8. Cantley, L.F., Taiwo, O.A., Galusha, D., Barbour, R., Slade, M.D., Tessier-Sherman, B., Cullen, M.R.: Effect of systematic ergonomic hazard identiﬁcation and control implementation on musculoskeletal disorder and injury risk. Scand. J. Work Environ. Health 40(1), 57–65 (2014) 9. Carrivick, P.J.W., Lee, A.H., Yau, K.K.W., Stevenson, M.R.: Evaluating the effectiveness of a participatory ergonomics approach in reducing the risk and severity of injuries from manual handling. Ergonomics 48(8), 907–914 (2005) 10. Takala, E.P., Pehkonen, I., Forsman, M., Hansson, G.A., Mathiassen, S.E., Neumann, W.P., Sjogaard, G., Veiersted, K.B., Westgaard, R.H., Winkel, J.: Systematic evaluation of observational methods assessing biomechanical exposures at work. Scand. J. Work Environ. Health 36(1), 3–24 (2010)

Prevention of Work: Related Musculoskeletal Disorders

483

11. Arvidsson, I., Dahlqvist, C., Enquist, H., Nordander, C.: Action Levels for Prevention of Work Related Musculoskeletal Disorders. Report nr 18/2017, Arbets- och miljömedicin Syd, Lund (SE) (2017). (in Swedish) 12. Mahdavian, N., Lind, C.M., Antonio Diaz Olivares, J., Iriondo Pascual, A., Högberg, D., Brolin, E., Yang, L., Forsman, M., Hanson, L.: Effect of giving feedback on postural working techniques. In: 16th International Conference on Manufacturing Research, Skövde (SE) (2018) 13. McCreery, J.K., Krajewski, L.J.: Improving performance using workforce flexibility in an assembly environment with learning and forgetting effects. Int. J. Prod. Res. 37(9), 2031– 2058 (1999)

Smart Textiles and Their Role in Monitoring the Body’s Fitness and Medical Conditions Mohamed Eldessouki1,2(&), Redha Taiar3, Tareq Ahram4, and Stanislav Petrik1 1

3

Institute for Nanomaterials, Advanced Technology and Innovation, Technical University of Liberec, Liberec, Czech Republic [email protected] 2 Faculty of Engineering, Mansoura University, Mansoura, Egypt GRESPI, Research Group in Engineering Sciences, University of Reims Champagne-Ardenne, Reims Cedex, France 4 IASE, University of Central Florida, Orlando, FL, USA

Abstract. Textiles are considered the “second skin” for humans, as they have direct contact with most parts of the body all the time. This allows “smart textiles” to monitor the body and maintain a record of its vital signs and physiological activities. In this paper we present advanced materials integrated into textiles to serve speciﬁc sensing and/or actuation purposes. The paper also survey the utilization of textile wearable devices to monitor breathing as one of their medical applications. Finally, the paper reports on the development for ﬁbrous structures with chemosensing elements that can be integrated in cloths to act as a sensor for hazardous gases in the environment. This work shows the promising role of smart textiles in monitoring body’s ﬁtness, diagnosing medical conditions, and alarming against possible issues. Keywords: Smart textiles Body’s vital signs

Thermochromic materials Chemosensor

1 Introduction Doctors diagnose diseases based on clinical examination of the patients and - most importantly - on their physiological signs. Patients, however, tend to give inaccurate reports on their vital activities and the severity of their conditions, and this misleading information might end up with a wrong diagnosis of their disease. Likewise, therapists can not predict the degree of patients compliance to rehabilitation instructions without a continuous record of their individual movements. Smart textiles with “sensory functions” provide solutions for these challenges as they allow: soft, flexible, deformable, breathable, comfortable, durable, washable, light weight, unobtrusive, and cost efﬁcient means for long-term monitoring of the body’s physiological parameters. Smart textiles were successfully used in measuring many physiological parameters such as the breathing patterns, heart and muscle activities, blood pressure, oxygen saturation, body movement and posture, electrodermal activity, as well as the

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 484–490, 2019. https://doi.org/10.1007/978-3-030-02053-8_74

Monitoring the Body’s Fitness and Medical Conditions

485

composition of body fluids. To achieve that, advanced sensing materials are integrated in the textile structures to monitor body’s activities. Therefore, in this work we will present the principle for these sensing materials, overview one of their applications, and an example of our developed sensitive membranes.

2 Advanced Sensitive Materials Advanced materials used in sensors are characterized by a change in their “state” after the application of external stimulants, as demonstrated schematically in Fig. 1. The external stimuli might be: mechanical, chemical, thermal, optical…etc., while the modulation of the material’s state can be observed as a change in: color, shape, orientation of structure, polarity…etc. Traditionally, naming these advanced materials consists of two (usually Greek) parts that reflect the type of stimulant and the material’s reaction to that stimulant. For example, thermochromic materials react to external alteration in thermal energy with a change in their color (chromic properties, hence named thermo-chromic). Similarly, piezoelectric materials react to external change of mechanical forces (piezo) by generating electrical current.

Fig. 1. Stimulants that result in changes in the material’s state of sensors (Numbers 6 and 9 at left and right of the ﬁgure do have special meaning, but they are used to demonstrate the change of material's state)

Among these sensitive materials, chemosensors will be presented here in more details because they will be used in this work as a demonstrator for potential applications. The colorimetric (or chromogenic) chemosensor is a material characterized with a change in its color or fluorescence upon the occurrence of a chemical binding of a guest stimulant. This provides an informative sign that can be implemented in guestsensing systems. Colorimetric chemosensors generally have two subunits: a receptor and an indicator (signaling) and they work according to two main principles [1]:

486

M. Eldessouki et al.

• Binding mechanism: Where the indicator subunit changes its properties once a binding occurs at the receptor part; as demonstrated at the top mechanism of Fig. 2. In this mechanisms the two subunits are linked through a covalent bond and the binding of the gust molecule results in a perturbation of the electronic structure, which results in a new color properties of the formed molecule.

Fig. 2. Mechanisms of chromogenic chemosensor operation; the mechanism at the top is commonly utilized while the replacement mechanism at the bottom of the ﬁgure is relatively limited in application

• Replacement mechanism: Where the receptor is weakly attached to an indicator with a speciﬁc color (or a fluorescence). The color of the indicator changes after a replacement interaction with the targeted guest; as shown at the bottom mechanism in Fig. 2. This mechanism depends on the release of the indicator with its characteristic change in the optical properties after the coordination at certain sites of the molecule. These colorimetric chemosensors became of special interest in detecting the presence of anions/cations and the pH modulation as well as monitoring of the toxic species [2–4], through color changes that can be observed by the naked eyes without any complicated instruments. Although other techniques are used for detecting the pH changes such as: microelectrodes [5], acid-base indicator titration, and potentiometric titration [6], these techniques might be restricted due to their processing complications, and/or economical constraints. On the other hand, colorimetric sensors are relatively cheap and have high sensitivity levels [7, 8] that allowed them to be used in many applications [9, 10].

3 Medical Applications of Smart Textiles Smart textiles utilizing mechanical, electrical, chemical or physical strains exerted by the body organs to generate responsive signals. For example, breathing is associated with a change in the chest volume and results in an expansion and contraction of the

Monitoring the Body’s Fitness and Medical Conditions

487

ribcage. These mechanical strains can be measured using strain gauges or utilized to stimulate speciﬁc reactions of the sensing materials. This principle can be demonstrated with piezoresistive elements that are embedded in textile structures and change their electrical resistance due to the modulations of the applied mechanical strains [11, 12]. Similarly, changes in properties of lightwaves traveling through optical ﬁbers that are integrated in the textile structure will be observed during alterations of these mechanical strains [13]. Modulation of the material’s capacitance under mechanical strains are also used as a sensing principle for breathing monitoring [14–16]. The type and structure of the textile product can affect its sensitivity, and elastic materials are commonly used in knitted [11, 12] and woven [16] structures. Knitted structures with different patterns [17] are commonly used because they provide more elasticity and deformation during breathing. Placing the sensor in the garment structure also affects its performance. For breathing monitoring, sensors are commonly placed on the chest and on the abdomen [16, 17], although they might be used in a single position and utilized in the form of a belt [15]. To test the performance of these sensitive textile products, multiple physical and breathing scenarios are commonly used, where subjects are tested in: upright, sitting, and supine positions. Breathing rate is another factor that should be considered and it is necessary to test the performance at several breathing speeds: normal, rapid, slow as well as periodical holding of the breath [11]. Moreover, smart textiles and wearable devices were used in other medical and health monitoring applications. Multiple review articles presented and discussed these applications from different perspectives as can be found in [18–22].

4 Experimental Results and Discussions Functional nanoﬁbrous membrane samples were produced based on the integration of pyrrolinone ester hydrazone dye into the polyacrylonitrile (PAN) solutions, as described in our previous work [23]. The dye at different concentrations (0.5, 1, 1.5 and 2 wt%) was used as a colorant for the PAN polymeric solutions. To test their response to changes in pH values of their medium, the prepared colored nano-membranes were exposed to ammonia. Small samples of the functionalized membranes were immersed in a series of ammonia solutions with different concentrations (2.5, 5, 10, 15, 20, 25%) and their color change was visually observed and photographed. The ﬁber diameter distribution for the produced samples was calculated and found to range between 150–250 nm. The smooth surfaces of the ﬁbers as well as the even distribution of the colors and their uniform appearance imply a compatibility and a good mixing between the dye and the polymer within the solution. The dye also did not show any leakage from the membrane during the washing and the wet handling of the samples. The sensitivity of the functional membranes to their medium’s pH was tested using different concentrations of ammonia solutions. The control sample (i.e. pure PAN) did not show any change in color after exposure to the ammonia, while the functional membrane samples demonstrated a clear shift in color from yellowish orange to dark brown color, as shown in Fig. 3. This change can be attributed to the presence of the

488

M. Eldessouki et al.

hydrazone dye, which is characterized by the formation of anion azo form in alkaline medium. The change in color was also observed to be reversible; where samples that are exposed to ammonia then titrated in an acidic medium (or washed in a neutral medium) returned back to their original color. It is necessary to point out that, the process kinetics and the speed of change in color was slower in the solid state case (i.e. in the functional membrane) as compared to the change when the same dye was tested with ammonia in liquid state (i.e. in solution).

Fig. 3. The effect of ammonia solution on the change of membrane color; The control sample (without ammonia) is shown to the far left and the other pictures present the samples’ color after immersion in ammonia solutions with concentrations indicated below each picture.

5 Conclusions Advances in materials are associated with novel applications and generated new opportunities for integration within the flexible textile structures. In this work, the principle for sensitive materials was introduced with an example for the colorimetric chemosensors. A functional dye was used to introduce sensing characteristics to nanoﬁbrous membranes of PAN, and the produced membranes have shown a potential application in visual detection of hazardous gases in industrial environments through their integration in labor’s garments or air ﬁlters. The functional dyes used in this work have shown also some thermochromic behavior, which can be implemented in other applications such as a visual indicator for the body’s temperature. In general, there are enormous biomedical applications for smart textiles and they become among the recent hot research topics with promising advances. Acknowledgments. The research reported in this paper was ﬁnancially supported by the Ministry of Education, Youth and Sports (MŠMT) of the Czech Republic in the framework of the targeted support of the “National Programme for Sustainability I” (LO 1201). The forth author also acknowledges the support of the MŠMT and the European Union - European Structural and Investment Funds in the frames of Operational Programme Research, Development and Education - project Hybrid Materials for Hierarchical Structures (HyHi - Reg. No.: CZ.02.1.01/0.0/0.0/16_019/0000843).

Monitoring the Body’s Fitness and Medical Conditions

489

References 1. Kumar, V.: Synthesis and characterization of multidentate schiff base podands and their use as chemosensors and catalysts. Ph.D. thesis, Guru Nanak Dev University (2011) 2. Zhou, Y., Zhang, J.F., Yoon, J.: Fluorescence and colorimetric chemosensors for fluorideion detection. Chem. Rev. 114(10), 5511–5571 (2014) 3. Zhou, X., Lee, S., Xu, Z., Yoon, J.: Recent progress on the development of chemosensors for gases. Chem. Rev. 115(15), 7944–8000 (2015) 4. Sun, W., Guo, S., Hu, C., Fan, J., Peng, X.: Recent development of chemosensors based on cyanine platforms. Chem. Rev. 116(14), 7768–7817 (2016) 5. Gameiro, P., Reis, S., Lima, J., de Castro, B.: Calibration of pH glass electrodes by direct strong acid/strong base titrations under dilute conditions. Anal. Chim. Acta 405(1), 167–172 (2000) 6. Balázs, N., Sipos, P.: Limitations of pH-potentiometric titration for the determination of the degree of deacetylation of chitosan. Carbohyd. Res. 342(1), 124–130 (2007) 7. Qian, Y., Cao, L., Jia, C., Boamah, P.O., Yang, Q., Liu, C., Huang, Y., Zhang, Q.: A highly selective chemosensor for naked-eye sensing of nanomolar Cu (II) in an aqueous medium. RSC Adv. 5(95), 77965–77972 (2015) 8. Saleem, M., Lee, K.H.: Optical sensor: a promising strategy for environmental and biomedical monitoring of ionic species. RSC Adv. 5(88), 72150–72287 (2015) 9. Yu, R.-Q., Zhang, Z.-R., Shen, G.-L.: Potentiometric sensors: aspects of the recent development. Sens. Actuators B Chem. 65(1), 150–153 (2000) 10. Scheller, F.W., Wollenberger, U., Warsinke, A., Lisdat, F.: Research and development in biosensors. Curr. Opin. Biotechnol. 12(1), 35–40 (2001) 11. Guo, L., Berglin, L., Wiklund, U., Mattila, H.: Design of a garment-based sensing system for breathing monitoring. Text. Res. J. 83(5), 499–509 (2012) 12. Loriga, G., Taccini, N., De Rossi, D., Paradiso, R.: Textile sensing interfaces for cardiopulmonary signs monitoring. In: Proceedings of the Conference on IEEE Engineering Medicine and Biology Society, vol. 7, pp. 7349–7352 (2005) 13. Grillet, A., Kinet, D., Witt, J., Schukar, M., Krebber, K., Pirotte, F., Depre, A.: Optical ﬁber sensors embedded into medical textiles for healthcare monitoring. IEEE Sens. J. 8(7), 1215– 1222 (2008) 14. Yang, C.M., Yang, T.L., Wu, C.C., Hung, S.H., Liao, M.H., Su, M.J., Hsieh, H.C.: Textilebased capacitive sensor for a wireless wearable breath monitoring system. In: Digest of Technical Papers - IEEE International Conference on Consumer Electronics, pp. 232–233 (2014) 15. Merritt, C.R., Nagle, H.T., Grant, E.: Textile-based capacitive sensors for respiration monitoring. IEEE Sens. J. 9(1), 71–78 (2009) 16. Guo, L., Berglin, L., Li, Y.J., Mattila, H., Mehrjerdi, A.K., Skrifvars, M.: Disappearing Sensor’-textile based sensor for monitoring breathing. In: 2011 International Conference on Control, Automation and Systems Engineering (CASE), pp. 1–4 (2011) 17. Guo, L., Peterson, J., Qureshi, W., Kalantar Mehrjerdi, A., Skrifvars, M., Berglin, L.: Knitted wearable stretch sensor for breathing monitoring application (2011) 18. Tay, F.E.H., Guo, D.G., Xu, L., Nyan, M.N., Yap, K.L.: Yap: MEMSWear-biomonitoring system for remote vital signs monitoring. J. Franklin Inst. 346, 531–542 (2009) 19. Chana, M., Estèvea, D., Fourniolsa, J.-Y., Escribaa, C., Campo, E.: Smart wearable systems: current status and future challenges. Artif. Intell. Med. 56, 137–156 (2012)

490

M. Eldessouki et al.

20. Zeng, W., Shu, L., Li, Q., Chen, S., Wang, F., Tao, X.-M.: Fiber-based wearable electronics: a review of materials, fabrication, devices, and applications. Adv. Mater. 26, 5310–5336 (2014) 21. Khan, Y., Ostfeld, A.E., Lochner, C.M., Pierre, A., Arias, A.C.: Monitoring of vital signs with flexible and wearable medical devices. Adv. Mater. 28, 4373–4395 (2016) 22. Majumder, S., Mondal, T., Deen, M.J.: Wearable sensors for remote health monitoring. Sensors 17, 130 (2017) 23. Eldessouki, M., Aysha, T., Ratičáková, M., Šašková, J., Padil, V.V.T., Ibrahim, M., Černík, M.: Structural parameters of functional membranes for integration in smart wearable materials. Fibres Text. East. Eur. 25(5), 73–78 (2017)

Quantifying Sense of Depth Towards Visual Texture Using Optics Simulation Masaki Shimomura1(&), Ruriko Kamesaka1, Mugi Nishihara1, Kei Matsuoka1, Takamasa Yoshimura2, Takeo Kato1, and Yoshiyuki Matsuoka1 1

Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama-shi, Kanagawa 223-8522, Japan [email protected], [email protected], [email protected], [email protected], {kato,matsuoka}@mech.keio.ac.jp 2 OPTIS Japan K.K., Paciﬁc Century Place Marunouchi 16F, 1-11-1 Marunouchi, Chiyoda-ku, Tokyo 100-6216, Japan [email protected]

Abstract. This study aims to quantify the sense of depth using optics simulation by the following two steps. First, the relationship between an artifact and the sense of depth was represented using a ﬁshbone diagram, which can clarify for clarifying mutual relationship among the physical properties of the factors related to the sense of depth (i.e. the lightness and the saturation). Second, to determine which factor is affecting the sense of depth and in what proportions, a sensory evaluation experiment was conducted based on Scheffe’s paired comparison and using the evaluation samples made by optical simulation. The results of this experiment indicate the lightness represents an important parameter in the feeling of the sense of depth. Keywords: Sense of depth

Texture Quantifying Optical simulation

1 Introduction Since the 1970’s, the need for functionality of products is mostly fulﬁlled by the rapid development of science and technology, but recently, people prefer focusing on their appearance in order to decide the purchase of them. A similar tendency can be seen in the automotive interior industry, and the study of the design and the sense of quality is conducted actively to cope with the demand. Some previous studies on the textures of the car’s interiors indicated that the sense of depth is one of in the main factors of the sense of quality and relates to several physical properties [1]. Besides, in a study on the relationship between textures and the physical properties, it appears that optics simulations are very effective for speciﬁed quantiﬁcation [2]. Therefore, to establish a design guideline to enhance the sense of quality, this study aims to quantify the relationship between the sense of depth and the considered physical properties using optics simulation.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 491–497, 2019. https://doi.org/10.1007/978-3-030-02053-8_75

492

M. Shimomura et al.

In order to do so, this paper follows the two following steps. First, the relationship between an artifact and the sense of depth was represented using a ﬁshbone diagram, which can clarify for clarifying mutual relationship among the physical properties of the factors related to the sense of depth (i.e. the lightness and the saturation). Second, to determine which factor is affecting the sense of depth and in what proportions, a sensory evaluation experiment was conducted based on Scheffe’s paired comparison and using the evaluation samples made by optical simulation.

2 The Cause and Effect Diagrams of the Sense of Depth Figure 1 represents a ﬁshbone diagram of causation created on the basis of the conventional research papers and expert interviews. It depicts the mutual relationship among the physical properties of the factors related to the sense of depth. Research papers are more than 70 articles dealing about color and texture evaluation (including documents from Society of Automotive Engineers of Japan, Architectural Institute of Japan and Japanese Society for the Science of Design). In this study, the “Lightness”, “Saturation” and “Eye point” were chosen as key parameters to evaluate the sense of depth in texture.

Fig. 1. Fishbone diagram of artifacts and texture

Quantifying Sense of Depth Towards Visual Texture

493

3 Sensory Evaluation Experiment with Simulation 3.1

Optical Simulation Environment and Setting

The evaluation sample was made within SPEOS, an optical simulation software dealing with photometric models and based on the Monte-Carlo ray-tracing method. This software was chosen because it can treat the following physical parameters: lighting spectrum, intensity, material property, texture, BRDF and BSDF. A D65 (white light) lighting spectrum was used for this experiment, and the illuminance sensor’s orientation was deﬁned at 30, 60, 90, 120 and 150° (Fig. 2).

Fig. 2. Overview of the simulation environment

The texture evaluation sample represents an automotive interior decorative panel. It is composed of two layers: the base is the colored part and the top is a clear PMMA. In this paper, the sample’s hue is red, only the lightness and saturation vary (Table 1). Table 1. Evaluation sample’s physical parameter

494

M. Shimomura et al.

3.2

Outline of Experiment

To determine which factor is affecting the sense of depth and in what proportions, a sensory evaluation experiment was conducted using the evaluation samples made by optical simulation. The experiments include static view and dynamic view tests, with the lightness and saturation as parameters. Table 2 indicates the experimental conditions. Table 2. Experimental conditions Research subject The number of evaluation sample Evaluation method Display Display resolution Display luminance

20’s university student, 9 people (Male: 7, Female: 2) 22 Scheffe’s paired comparison, 5 grades evaluation (+2 to −2) Eizo color management monitor 2560 1440 pixel 200 cd/m²

In the static view experiment, only sample images from a 90° observation angle were considered. Two sample images were placed side by side, and the subjects were asked if the sense of depth was more or less perceived in the right one than in the left one (the reference). Since compared images were presented at the same time, the Nakaya’s variation method (which does not consider the order effect) was used. In the dynamic view experiment, the sample images changed at a constant speed (every 2 s) each 30°, to comprehensively compare the total of 5 viewpoints. (Fig. 3). In dynamic experiments, it was necessary to consider the order effect, thus the Ura’s variation method was used.

Fig. 3. Color comparison with simulation results

Quantifying Sense of Depth Towards Visual Texture

3.3

495

Experiment Results

The results for each subject were merged into an overall cross table, so a ANOVA could be conducted (Table 3). In the ANOVA table, each variance S value was obtained ﬁrst, then the degree of freedom u and the unbiased variance V and ﬁnally F0. Each calculation formulas are described below in Eqs. 1–9. Here, n is the number of samples and N is the number of subjects. Table 3. ANOVA table of color change experiment S V F0 Signiﬁcance probability 82.8 27.6 51.5 ** 37.0 0.5 96.3 32.1 56.2 ** 39.4 0.6

Factor Main effect Error Dynamic Main effect Error Static

Sa ¼ SaðkÞ ¼ Sb ¼

2 1 X X:j: Xi:: : 2nN

2 1 Xn Xn X:jk Xi:k Sa : i¼1 j¼1 2n

ð1Þ ð2Þ

2 1 Xn Xn Xij: Xji: Sa : i¼1 j¼1 2N

ð3Þ

1 X2 N nð n 1Þ

ð4Þ

Sr ¼

Xn 1 X 2 Sr k1 ::k nð n 1Þ

ð5Þ

Se ¼ Ss SaðkÞ Sb Sr SrðkÞ

ð6Þ

SrðkÞ ¼

Ss ¼

Xn Xn Xn i¼1

j¼1

k¼1

2 Xijk

ð7Þ

V¼

S : U

ð8Þ

F0 ¼

V Ve

ð9Þ

The F0 values were compared with the F distribution table and the signiﬁcant differences were tested. When a signiﬁcant difference was observed at a signiﬁcant level of 5%, * was written in the test result column, and ** was written in the test result column when a signiﬁcant difference was observed at the signiﬁcant level of 1%.

496

M. Shimomura et al.

From the results of the ANOVA, signiﬁcant differences were found in the factors of “stimulation” in both static and dynamic experiments using lightness and saturation as indicators. The scale value is expressed on the yard stick (Fig. 4).

Fig. 4. Experiment result with yard stick

In both static and dynamic experiments, samples B and C have negative scale values, resulting in a feeling of depth not being felt. On the other hand, samples A and D have a positive scale value, so a feeling of depth was actually felt. Furthermore, in order to clarify the relationship between samples, a difference test was carried out. As a result, in both static and dynamic experiments, it was recognized that there was a signiﬁcant difference between samples B and C, C and A at the signiﬁcant level of 1%. Since there was no signiﬁcant difference between samples A and D, it was considered that the feeling of depth was the same. Based on the above results, the samples B and C which did not generate the feeling of the sense of depth compared to the samples A and D, which did actually generate the feeling of the sense of depth, appear to have a high value of the lightness L*. Therefore, a tendency to feel more the sense of depth in the sample with low lightness than in the samples with high lightness, regardless of the saturation value, was observed. As a result, it was considered that the effect of lightness is more important than that of saturation. In addition, when comparing the samples B and C, in which both signiﬁcant differences were observed, a tendency of a sample having higher saturation tends to generate a greater feeling of the sense of depth was also observed.

4 Conclusion In this study, the effectiveness of the three selected parameters “Lightness”, “Saturation” and “Eye point” was investigated by sensory evaluation experiment using 4 optical simulated samples and the ANOVA. As for future works, since the current number of samples in this experiment is small (only 4), it is necessary to increase the number of samples. It is also planned to reduce the orientation sampling of the sample to make the dynamic view smoother, and to consider the BRDF as a parameter of the feeling of the sense of depth.

Quantifying Sense of Depth Towards Visual Texture

497

References 1. Matsuoka, K., Maedomari, H., Mori, Y., Kishimoto, Y.: Visual evaluation factors for interior material textures as a design guideline. In: Proceedings of the 62nd Annual Conference of JSSD, vol. 62, p. 63 (2015). (in Japanese) 2. Nagata, N., Dobashi, T., Manabe, Y., Usami, T., Inokuchi, S.: Modeling and visualization for a pearl-quality evaluation simulator. IEEE Trans. Vis. Comput. Graph. 3(4), 307–315 (1997)

Factor Analysis of Synesthetic Perceptual Dimensions Using Aluminum Alloy Material Textures Surface in Industrial Products Jialun Huang(&), Xiaozhou Zhou, Chengqi Xue, Lei Zhou, and Yafeng Niu Department of Industrial Design, School of Mechanical Engineering, Southeast University, Nanjing 211189, China {220163579,zxz,ipd_xcq,zhoulei,nyf}@seu.edu.cn

Abstract. In the context of the rapid development of industrial design, selection and combination of materials have become a signiﬁcant topic of design research. Material texture can be perceived by the sense of vision and touch, but previous academic studies have only focused on single channel perception of materials. Here, we explore the correlation between the factors of material synesthetic perception dimensions and the evaluation dimension of aluminum alloy material surface. Analysis reveals four major dimensions: “Color,” “Glory,” “Roughness,” and “Regularity.” These dimensions include three factors that previous studies have regarded as fundamental in different studies, as well as one new factors: “Regularity”. This new factor put forward new ideas from the perspective of visual-tactile perception. Additionally, we indicate that when the factors “color” and “Glory” effect separately, its signiﬁcance on material perception is higher than their simultaneous effect. Keywords: Synesthesia perceptual dimensions Texture surfaces Industrial products

Aluminum alloy material

1 Introduction In the current industrial design practice, selection and combination of materials have become a signiﬁcant topic of design research, as Choi [1] indicated that materials play a decisive role for an innovative product design because consumers interact with products through materials to experience the unique texture of products. Generally, material texture can be perceived by the sense of vision and touch, but previous academic studies have only focused on single channel perception of materials. Tactile perceptual dimensions of materials could be explored by using sensory vocabulary and the research pointed out six major dimensions [2]. The sensory and emotional aspects of touch are related when evaluating wooden surfaces using ﬁngertips [3]. Similarly, sensory perception of material texture could be described using the four dimensions [4, 5]. In visual perspective, the visual perception of materials could be divided into ﬁve factors: brightness and color, translucency, transparency,

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 498–505, 2019. https://doi.org/10.1007/978-3-030-02053-8_76

Factor Analysis of Synesthetic Perceptual Dimensions

499

gloss and shape [6]. Some studies also focus on materials and their properties from the perspective of visual perception [7, 8]. However, when applied singularly, both vision and touch have their limitation in texture perception because vision and touch tend to interact with each other. In order to obtain more complete dimension space of material perception, we propose the synesthetic perceptual dimensions based on material texture surface. Our study focuses on the aluminum alloy material texture surface because aluminum alloy is one of the most widely used nonferrous metal structure materials in home appliance products. The aluminum alloy material is classiﬁed according to the number, which can be divided into eight categories [9]. It should be pointed out that the 5000 series is a commonly used alloy aluminum plate series and widely used in aviation and electronic products, thus we use this series as experimental samples. The aim of this study is to explore the correlation between the factors of material synesthetic perception dimensions and the evaluation dimension of aluminum alloy material surface. To investigate these empirically, we used forty-two aluminum alloy material texture samples from four different processing craft and surface treatment. The ﬁndings are further discussed in relation to possible applications in the design of aluminum alloy products to maximize the positive aspects of product experiences.

2 Materials and Methods 2.1

Pre-experimental Designs

In psychological research, perceptual evaluation requires pre-screening and determination of the bipolar vocabulary of material perception in order to reduce the unnecessary tasks of formal experiments and improve the efﬁciency. Participants. Four healthy participants aged 20–30 years (mean age = 25.5 years; 2 males and 2 females) took part in the pre-experiment. Each participant had normality of visual or tactile sensory systems and took 90°

*Audition

Sound level

Haptics Size/Reach

Force in Z axis Stature

Posture Movement Speed

Posture style Robot speed

*Control Bimanual tasks *Locomotion Walking speed *Strength *Force Control Motivation Autonomy Competence Relatedness * User-speciﬁc factors

Heavy parts Angles between the upper arms and the elbows Providing choice Control over robot Support by robot

Choosing robot speed Hand-guiding Holding parts

Comfort Design in Human Robot Cooperative Tasks

525

5 Conclusions HRC workcells have been gaining more attention recently by industries due to small lot size production and high demand for adaptability and customization. Workplace design usually considers the ergonomic factors. Standard anthropometric data presents information about static or fundamental features in usual postures. However, these data cannot provide practical capabilities, such as body ranges and moves for speciﬁcally a human-robot cooperation job. Furthermore, the considerations to cognitive aspects in HRC have not been considered in details previously. This paper reports experimenting on a human-centric design of HRC task and evaluated particular mental and physical impacts on subjects. The study deﬁned the baseline as an HRC job which robot is transferring large parts of a truck engine to human. The second experiment had main features of user-personalized space and robot hand-guiding mode. The updated HRC space design tries to overcome the common issues created for users and provide physical and mental comfort to users for better productivity and higher efﬁciency. The novelty of this concept is that the design is considering human mental and physical viewpoints at the same time. User studies revealed that user’s ergonomic postures were improved. Utilization of hand-guiding on the robot made the users think they have control over the system and the robot is a helper rather than a machine passing a part. For future work considerations, adjusting the robot speed and performing the task for older users for example above 60 could be considered to study the effects on users. Acknowledgments. Authors would like to thank Jane and Aatos Erkko Foundation and Technology Industries of Finland Centennial Foundation for the support of UNITY (2016-2019) project and the Academy of Finland for the project ‘Competitive funding to strengthen university research proﬁles’, decision number 310325.

References 1. Krüger, J., Lien, T.K., Verl, A.: Cooperation of human and machines in assembly lines. CIRP Ann. Manuf. Technol. 58, 628–646 (2009) 2. Robla-Gomez, S., Becerra, V.M., Llata, J.R., Gonzalez-Sarabia, E., Torre-Ferrero, C., PerezOria, J.: Working together: a review on safe human-robot collaboration in industrial environments. IEEE Access 5, 26754–26773 (2017) 3. Caputo, F., Di Gironimo, G., Marzano, A.: Ergonomic optimization of a manufacturing system work cell in a virtual environment. Acta Polytech. 46, 21–27 (2006) 4. Wykowska, A., Chaminade, T., Cheng, G.: Embodied artiﬁcial agents for understanding human social cognition. Phil. Trans. R. Soc. B. 371, 20150375 (2016) 5. Macfarlane, G.J., Thomas, E., Papageorgiou, A.C., Croft, P.R., Jayson, M.I.V., Silman, A.J.: Employment and physical work activities as predictors of future low back pain. Spine (Phila. Pa. 1976) 22, 1143–1149 (1997) 6. Zander, J.E., King, P.M., Ezenwa, B.N.: Influence of flooring conditions on lower leg volume following prolonged standing. Int. J. Ind. Ergon. 34, 279–288 (2004)

526

A. Changizi et al.

7. Tissot, F., Messing, K., Stock, S.: Studying the relationship between low back pain and working postures among those who stand and those who sit most of the working day. Ergonomics 52, 1402–1418 (2009) 8. Changizi, A., Lanz, M.: The comfort zone concept in a human-robot cooperative task. In: Ratchev, S.M. (ed.) Proceedings of the Precision Assembly Technologies and Systems, Chamonix, France (2018) 9. Ryan, R.M., Deci, E.L.: Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am. Psychol. 55, 68 (2000) 10. Sheldon, K.M., Elliot, A.J., Kim, Y., Kasser, T.: What is satisfying about satisfying events? Testing 10 candidate psychological needs. J. Pers. Soc. Psychol. 80, 325 (2001) 11. Motmans, R.: DINBelg 2005 - anthropometry table. http://dinbelg.be/anthropometry.htm 12. Molenbroek, J.: DINED- Anthropometry in design. https://dined.io.tudelft.nl/en 13. Sanders, M.S., McCormick, E.J.: Human Factors in Engineering and Design. McGraw-Hill, New York (1987) 14. Smith, M.W., Sharit, J., Czaja, S.J.: Aging, motor control, and the performance of computer mouse tasks. Hum. Factors 41, 389–396 (1999) 15. Li, K.Z.H., Lindenberger, U., Freund, A.M., Baltes, P.B.: Walking while memorizing: agerelated differences in compensatory behavior. Psychol. Sci. 12, 230–237 (2001) 16. Mattsson, S., Fast-Berglund, Å., Åkerman, M.: Assessing operator wellbeing through physiological measurements in real-time—towards industrial application. Technologies 5, 61 (2017) 17. Li, D., Landström, A., Mattsson, S., Karlsson, M.: How changes in cognitive automation can affect operator performance and productivity. In: The Sixth Swedish Production Symposium (2014) 18. Mather, M., Carstensen, L.L.: Aging and motivated cognition: the positivity effect in attention and memory. Trends Cogn. Sci. 9, 496–502 (2005) 19. Grolnick, W.S., Ryan, R.M.: Autonomy in children’s learning: an experimental and individual difference investigation. J. Pers. Soc. Psychol. 52, 890 (1987) 20. Groves, R.M., Fowler Jr., F.J., Couper, M.P., Lepkowski, J.M., Singer, E., Tourangeau, R.: Survey Methodology. John Wiley & Sons, New York (2009) 21. Harvey, C., Stanton, N.A.D., Young, M.S.: Guide to Methodology in Ergonomics: Designing for Human Use. CRC Press, London (2014) 22. Fozard, J.L., Gordon-Salant, S.: Sensory and perceptual changes with aging. In: Handbook of the Psychology of Aging, vol. 5 (2001)

PlayCube: Designing a Tangible Playware Module for Human-Robot Interaction Vinicius Silva1(&), Filomena Soares1,2, João Sena Esteves1,2, and Ana Paula Pereira3 1

Algoritmi Research Centre, University of Minho, 4800-058 Guimarães, Portugal [email protected], {fsoares,sena}@dei.uminho.pt 2 Department of Industrial Electronics, University of Minho, 4800-058 Guimarães, Portugal 3 Research Centre on Education, Institute of Education, University of Minho, 4710-057 Braga, Portugal [email protected]

Abstract. In general, humans can express their intents effortlessly. On the contrary, individuals with Autism Spectrum Disorder (ASD) present impairments in this area. Researchers are employing different technological strategies in order to improve the emotion recognition skills of individuals with ASD. Among those technological solutions, the use of Objects based on Playware Technology (OPT) in context of serious games is getting increasing attention. Following this trend, the present work proposes the development of an OPT module to be used as an add-on to the human-robot interaction with children with ASD in emotion recognition activities. To evaluate the proposed approach, usability tests with typically developing children in a school environment were conducted. Overall, the different evaluations allow estimating how the children interacted with the OPT. Keywords: Human-computer interaction

Playware Tangible interfaces

1 Introduction Understanding the emotional state of the others is essential for a successful interaction. In general, humans are naturally expressive. However, the understanding of social signs can be a very difﬁcult task for children with Autism Spectrum Disorders (ASD) [1]. Currently, researchers are proposing new interactive tools in order to mitigate the emotion recognition impairments that individuals with ASD usually present. Among those solutions, it is worth to point out the use of Objects based on Playware Technology (OPT) and assistive robots [2, 3]. Initially proposed by Henrik Lund, the term “playware” deﬁnes a combination of intelligent hardware and software with a goal of producing new innovative play and playful experiences among users [2]. This technology emphasizes the role of interplay between morphology and control using processing, input, and output. Few research projects have been using objects with playware technology as an intervention tool with © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 527–533, 2019. https://doi.org/10.1007/978-3-030-02053-8_80

528

V. Silva et al.

children with ASD. Henrik Lund [4] proposed and developed a modular robotic playware with the goal of allowing easy creation of games by being flexible in both setup and activity building for the end user. The approach consisted in the development of interactive quadratic shaped tiles with a local source of energy and wireless communication, with different games. A proposed game consisted in mixing the tiles in order to produce new colours. The authors conducted a set of experiments with a group of seven children with ASD. The results allowed the authors to conclude that OPT can be playful tools for cognitive challenged children. Concerning social robots, systems that can synthesize affects – for example virtual characters (avatars) – and robotic platforms can be applied in autism intervention [5]. Research has demonstrated that robots promote new social behaviours in children with ASD [6]. They can be a social support, helping to identify, measure, and promote social behaviours, as they can be repeatable and objective. Considering the different designs employed in the assistive robotics research, more recently, researchers are using robots with a humanoid design, since they allow a pronounced potential for generalisation, particularly in tasks of imitation and emotion recognition [7, 8]. Following this trend, the present work consists in the development of an OPT module to be used as an add-on to the human-robot interaction with children with ASD in emotion recognition activities. A motivation for this work is that none of the works in the related literature – at least to the authors’ knowledge – presents an approach of using assistive robots together with OPT. The module consists of an RGB camera and a cube shaped OPT called the PlayCube. The cube is a tangible interface between the user and the robot: it gives representations of the robot and enables controlling the robot. This new device has an Inertial Measurement Unit (to detect the cube orientation and to allow for emotions randomly displayed), a microcontroller and two actuators (a motor and a group of LEDs) to give haptic and visual feedback to the user. The immediate feedback provided by the cube allows stimulating children engagement in the task by giving positive reinforcement according to the children performance. In order to modify the robot behaviour, a machine learning model was trained in order to detect and track the OPT during the interaction with the children. Usability tests with typically developing children were performed in a school environment in order to evaluate the system constraints as well as the tangible interaction, i.e., the gestures that can be performed with the PlayCube and the different feedbacks (visual and haptic). The present article is divided in ﬁve sections. The developed module is presented in Sect. 2. Section 3 describes the general methodology used to implement the object detection and tracking algorithm as well as the general procedure used for evaluating the developed module. Section 4 shows the results. Finally, Sect. 5 addresses the ﬁnal remarks and future work.

2 Developed Module - PlayCube The developed module, depicted in Fig. 1, is composed of a camera, a computer, and the new OPT called PlayCube.

PlayCube: Designing a Tangible Playware Module

529

a)

c)

b)

Fig. 1. Experimental setup (a) camera, (b) PlayCube, and (c) computer.

The developed OPT has the shape of a cube (dimensions: 7 cm 7 cm 7 cm). At its top face it has a RGB OLED display, an Inertial Measurement Unit (IMU) and a touch sensitive surface placed over the OLED screen, Fig. 2. The designed OPT was developed with the following guideline in mind: to offer a tangible and adaptive experience, that is intuitive and easy to manipulate, with different sources of immediate feedback. Following this idea, the PlayCube was developed to adapt and be used in different activities and contexts (e.g. play, learning, among others) by physically (re-) programming it. The users expect to see and feel results from their actions immediately, so the feedback is an important part to have in mind when designing a new interface. This is

Fig. 2. General perspective of the developed OPT (PlayCube).

530

V. Silva et al.

more important if the target group is children with ASD since learning via reinforcement can be one of the most effective approaches in their case [10]. In order to ensure the immediate feedback, the developed OPT has a built-in multicolor ring with a total of sixteen LEDs, as well as a haptic driver with a Linear Resonant Actuator (LRA), providing both visual and haptic feedback to the user. In order to interact with the cube, a 6 Degree of Freedom (DoF) IMU and touch sensitive surface is used. Therefore, interacting with the PlayCube just means touching the physical object and manipulating it via natural gestures (e.g. rotation, shake, tilt, among others). The camera used is a full high deﬁnition RGB HP web camera. This camera was used in order to track the OPT during an intervention session by using computer vision algorithms.

3 Procedure This section describes the methods used for detecting and tracking the OPT and the general procedure employed during an exploratory study with the goal of evaluating the OPT and detecting the system constraints. 3.1

OPT Detection and Tracking

In order to detect the PlayCube a Histogram-of-Oriented-Gradient (HOG) based object detector from the Dlib library [9] was trained in python. This detector was trained with 120 images of the cube in different positions. The correlation tracker from the Dlib was also used, allowing the detection and tracking of the PlayCube in real-time during an intervention session. 3.2

Evaluation Methodology

An exploratory study with eight typically developing children was conducted in a school environment with the goal of testing and observing the children’s reaction to the OPT. Since the study involves working with children, the following issues were ensured to meet the ethical concerns: the school which participated in the studies established a protocol with the research group and informed consents were signed by the parents/tutors of the children that participated in the studies. Initially, a questionnaire was presented to the children in order to evaluate the “emojis” used in the cube. In this questionnaire, each child was asked to label the emotion that the “emoji” was expressing. Then, in the next activity, the child was asked to scroll through the cube and ﬁnd an emotion. This test had the main goal to evaluate two different groups of gestures: tilt up/down and tilt left/right. Besides the quantitative evaluation, a qualitative analysis was performed. This analysis had one source – observation of a child’s interaction behaviour during the task. With this in mind, the main purpose of this evaluation was to identify particular events such as the reaction of the child to the OPT, the different gestures, and the child’s reaction to the different feedback sources of the cube.

PlayCube: Designing a Tangible Playware Module

531

4 Results and Discussion Figure 3 shows the results regarding the object detection and tracking module. It is possible to conclude that the module can successfully detect and track the PlayCube.

Fig. 3. Tracking the PlayCube in real time.

Table 1 presents the results of the questionnaire conducted during the exploratory study with eight typically developing children. The children correctly labelled, with a success rate of 100%, the happiness, sadness, anger, and surprised emotions. However, they had more difﬁculties in labelling the last two emotions – afraid (50%) and neutral (38%). Table 1. Children’s answers to the questionnaire. Emotion Happiness Sadness Anger Surprised Afraid Neutral

Response time 100% 100% 100% 100% 50% 38%

As qualitative remarks, two groups of different tilt gestures – up/down and left/ right – allowing the children to scroll through the different faces displayed by the cube were evaluated. In general, it was observed that the up/down tilt gestures were more natural and easier for the children then the left/right tilt gestures when manipulating the cube. Additionally, the different feedback sources (visual and haptic) contributed to the way the children interacted with the cube since they understood when they were selecting a correct answer.

532

V. Silva et al.

5 Final Remarks and Future Work Understanding human emotions is paramount to any social communication. However, there are individuals that present impairments in this area, in particular children with ASD. With this in mind, the presented work proposes the PlayCube, a new OPT that will be used as an add-on to the human-robot interaction with children with ASD. In order to evaluate the proposed approach and to ﬁnd the system constraints, usability tests were conducted. Typically developing children tested the PlayCube by handling it in a school environment. By analysing the results of the tracking module, it is possible to perceive that this module can successfully detect and track the OPT. Concerning the results of the exploratory study, it is possible to conclude that the children interacted and understood the mechanics of the OPT. In general, it was observed that children preferred the up/down tilt gestures to manipulate the cube. The visual and haptic feedbacks are important functions as they allow the children to know when selecting a correct answer. Additionally, the data extracted from the cube (e.g. IMU data) and the object tracking module can be used to study how the child interacts with the cube. As future work this module will be added to the framework, and a larger study will be conducted with children with ASD in order to evaluate how the OPT can be used as a valuable tool to promote emotion skills (imitation and recognition) of children with ASD. Acknowledgements. The authors would like to express their acknowledgments to COMPETE: POCI-01-0145-FEDER-007043 and FCT – Fundação para a Ciência e Tecnologia within the Project Scope: UID/CEC/00319/2013. Vinicius Silva also thanks FCT for the PhD scholarship SFRH/BD/ SFRH/BD/133314/2017. The authors thank the teachers and students of the Elementary School of Gualtar (EB1/JI Gualtar) in Braga for the participation.

References 1. Christensen, D.L., et al.: Prevalence and characteristics of autism spectrum disorder among children aged 8 years — autism and developmental disabilities monitoring network, 11 sites, United States, 2012. MMWR Surveill. Summ. 65(3), 1–23 (2016) 2. Lund, H.H., Dam Pedersen, M., Beck, R.: Modular robotic tiles: experiments for children with autism. Artif. Life Robot. 13(2), 394–400 (2009) 3. Pennisi, P., et al.: Autism and social robotics: a systematic review. Autism Res. 9(2), 165– 183 (2016) 4. Lund, H.H., Klitbo, T., Jessen, C.: Playware technology for physically activating play. Artif. Life Robot. 9(4), 165–174 (2005) 5. Tapus, A., Member, S., Scassellati, B.: The grand challenges in socially assistive robotics. IEEE Robot. Autom. Mag. 14, 1–7 (2007) 6. Kim, E., Paul, R., Shic, F., Scassellati, B.: Bridging the research gap: making HRI useful to individuals with autism. J. Hum. Robot Interact. 1(1), 26–54 (2012) 7. Costa, S.: Affective robotics for socio-emotional skills development in children with autism spectrum disorders. University of Minho (2014)

PlayCube: Designing a Tangible Playware Module

533

8. Begum, M., et al.: Measuring the efﬁcacy of robots in autism therapy. In: Proceedings of the Tenth Annual ACM/IEEE International Conference on Human-Robot Interaction - HRI 2015, pp. 335–342 (2015) 9. King, D.E.: Dlib-ml: a machine learning toolkit. J. Mach. Learn. Res. 10, 1755–1758 (2009) 10. Schuetze, M., Rohr, C.S., Dewey, D., McCrimmon, A., Bray, S.: Reinforcement learning in autism spectrum disorder. Front. Psychol. 8, 2035 (2017)

Design for Seniors: A Case Study Based on Human Centric Lighting Aldo Deli1, Massimo Di Nicolantonio1(&), and Emilio Rossi1,2 1

2

Department of Architecture, University of Chieti-Pescara, Viale Pindaro, 42, 65127 Pescara, Pescara, Italy [email protected], [email protected], [email protected] Emilio Rossi Design Consulting, Via Venezia 4, 66026 Ortona, Chieti, Italy

Abstract. In last years, the world population’s average age has constantly increased. Thus, the ‘seniors’ class will be, in the near future, one of the most important groups of customers able to signiﬁcantly condition the products’ demand. In the lighting industry, the ‘age factor’ will be one of the latest aspects to take into account in order to design new generations of enabling projects. The HCL (Human Centric Lighting) is an innovative approach able to give value to both visual and non-visual light effects – i.e. emotions – that are still not considered for the current mass production. This paper introduces the HCL approach in lighting industry and, then, the processes of action-research used to design a new domestic lighting system. The HCL allowed to design a radically new product, with a set of focused technical-aesthetic solutions that make easier the light control by seniors. Keywords: Seniors Human Centric Lighting New product development Lighting system

Design approach

1 Seniors and Human Centric Lighting From the anthropological point of view, the ability to perceive spaces and objects is a vital characteristic for humans, being essentially entrusted to the quality of sight. This aspect describes humans as a vision-centric species. While the need to ‘see the world’ is still a basic factor, the human ability to perceive, properly, all phenomena varies throughout years, being affected by both physiological and cognitive factors (i.e. reduced sensitivity of contrasts, poor night vision, lack of color perception, etc.). In last years, the concept of ‘old age’ has been radically questioned. Recent estimations state that in the near future the number of people over 65 will constantly and signiﬁcantly increase [1]. This aspect represents a priority both from the design point of view – i.e. new generations of user-friendly solutions are needed – and from the commercial one – the increasing number of seniors will produce a modiﬁcation in the market demand. Until a few years ago the collective idea of old person was linked to an archetype describing, in principle, someone with reduced physical skills, while its cognitive and/or perceptual abilities were still less investigated even from the design point of © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 534–539, 2019. https://doi.org/10.1007/978-3-030-02053-8_81

Design for Seniors: A Case Study Based on Human Centric Lighting

535

view. As a result, products used by this emerging class were still conceived using traditional UCD approach. While UCD has been extensively used to conceive and produce everyday object, recent advances in Design Research have demonstrated its weakness in facing complex socio-physical issues affecting, for example, seniors [2]. A new design approach able to manage and lead the development of new enabling products is needed. In lighting industry, the HCL (Human Centric Lighting) Approach is one of the most recent and innovative design approaches usable to conceive lighting systems. It considers light not only as a performance parameter, but it also ponders the non-visual effects – i.e. biology influence, emotions, etc. – that are still not taken into account for the current productions [3]. HCL allows the user to have an absolute control of light factors through its various intrinsic characteristics such as: direction, intensity, color gradation and exposure time. In addition to the evaluation and management of light’s psycho-physiological effects, HCL influences two of the main emerging technologies: Home Automation and IoT (Internet of Things). The ﬁrst big potential of HCL concerns the product’s personalization: considering its scientiﬁc and functional character, HCL allows the development of user-friendly lighting solutions, namely, adjustable and/or updatable solutions and components that meets end-users’ needs over time. Accordingly, the main trends that can be mostly influenced by smart lighting solutions made using HCL are [3]: • Smart applications: thanks to technologies such as Bluetooth and Wi-Fi, lighting systems can be controlled via smartphone, moving closer to future home automation systems. • Energy efﬁciency and green building: considering the incessant growth of global electricity consumption, the HCL and smart systems are able to save energy and money. While the ultimate HCL’s goal is the improvement of the physiological and emotional conditions of people, in terms of applications, HCL acts as a multi domain enabler. It can enhance the quality of many spaces, such as: • Hospitals (enhancement of the overall patients’ well-being) improving the quality of sleep preventing the rise of chronic diseases – hospitals and home care are the most promising scenarios for short terms applications. • Domestic, public and professional spaces used by seniors like homes, ofﬁces, public spaces, gardens, museums, etc. Stakeholders are slowly realizing how lighting can affect productivity, vigilance, absenteeism and fatigue, as well as reducing stress. • In schools light plays an important pedagogical function, increasing concentration and therefore reducing errors to achieve better performance.

2 Aims This paper shows to demonstrate the applicability of HCL in the development of a domestic lighting system usable by all people and, in particular, by seniors having natural physical and cognitive lacks. The goals of this work are:

536

A. Deli et al.

• Demonstrate the validity and the utility of HCL Approach in the development of new domestic lighting systems and solutions usable by all, even by seniors with reduced abilities. • Show the technical and design improvements resulting from the adoption of HCL Approach in the design of new ‘seniors-friendly’ domestic lighting systems.

3 Concept Design: The Domestic Lighting Systems The testing ground used to verify the applicability of HCL concerns the domestic space [4]. The experimentation, here synthetically summarized, allowed to design a radically new product, with a set of focused technical-aesthetic solutions that make easier the light control by seniors. Speciﬁcally, the product here presented has been conceived starting from the analysis of end-users’ needs and, in particular, those of senior compared to the need to interact with domestic solutions [4]. The design intent was to conceive a coherent set of elements that has been later proposed through simplicity and formal clarity. As shown in Fig. 1, the product is then by ﬁve cardinal elements: a base, a reflector, a diffuser, a lamp on the rod and the controls.

Fig. 1. Overview of the domestic lighting system, where all the parts are conceived to achieve the formal clarity.

The process of design development led by HCL allowed to conceive an integrated and coherent product, which is composed by the simplest components designed for their speciﬁc purposes. The overall of the light systems height – 180 cm – has been set with the precise intent to create a user-friendly shape and, then, to mitigate the feeling

Design for Seniors: A Case Study Based on Human Centric Lighting

537

of ‘visual intrusiveness’ in the home environment and contain the ‘fear of use’ resulting from its natural physicality – i.e. aesthetics. The use of HCL made the product both dependent for user’s control and integrated with the environment in which it can be inserted. In fact, the light system is able to detect the external parameters, such as the amount of light already available per day and hour, the external dynamic inputs (i.e. movements), etc. and independently adjusts quality and the type of light, its intensity and color temperature. The main functions introduced by the lighting system are related to the different type of activities that seniors can exploit throughout the day. First of all, an integrated light sensor detects the amount of light in the environment and autonomously decides to inform end-users about its presence; this is performed by activation of the lights placed on controls and on the proﬁle. When the lighting system starts working, the controls remain lit (Fig. 2, left). Streetlight provides night lighting if/when end-user needs to move from the lamp’s surrounding area to other rooms or space areas. The position of source lights allows to illuminate the floor and, partially, the close furniture, suggesting how to reach the nearest switch and reducing the risk of falls (Fig. 2, right).

Fig. 2. Domestic lighting system: the automatic ignition (left) and the streetlight for the night (right).

Figure 3 shows the normal condition of the lighting system. To be more precise, during its normal use, the lighting system generates indirect light effects, which can illuminate the environment through regulation of the intensity – it ranges from 800 to 1.600 lm. Furthermore, for relaxing activities (i.e. reading), the light source on the rod provides direct illumination through regulation of intensity – it ranges from 400 to 600 lm. The color temperature changes automatically but end-users also have the chance to change it through the digital control (Fig. 4).

538

A. Deli et al.

Fig. 3. Domestic lighting system: normal use (left) and use for relaxing activities (right) and 4000 K (left) and 2.700 K (right).

Fig. 4. Domestic lighting system: examples of color temperatures at 4.000 K (right) and 2.700 K (left).

Finally, the design experimentation pointed out that the HCL approach is inconsistent if not combined with digital controls. Aging issues introduce many new challenges, including the investigation of the relationship between senior and technology. Technology is for them a tool to be active, remaining socially and intellectually connected with the world. Therefore, for this product, an App has been developed as main

Design for Seniors: A Case Study Based on Human Centric Lighting

539

interface to get new control functions: end-user can freely change all parameters. The decision to locate the control within an App is due to several factors: studies have shown that seniors have less interactions’ difﬁculties if they use touchscreen devices [5], because it requires less cognitive efforts and coordination; in addition, touchscreen devices are today widely used in homes to improve and make easier the daily life of seniors.

4 Conclusions Compared to the existing socio-technical issues affecting seniors, HCL can be considered as one of the most promising design and research approaches usable to conceive future generations of lighting solutions. As it has been shown, HCL allowed to design a new domestic lighting systems having new easy-to-use solutions that make easier the end-users’ interaction with products. Through the use of new functions, dynamic regulations and digital interconnections, it has been possible to design a new senior-friendly product having an high commercial appeal, which is also able to meet the fast-growing idea of ‘intelligent device’ able to exploit and maximize the quality of light applicable to any scale and in every context. Because in the next years the need to have ‘seniors-friendly’ solutions will increase, HCL will allow to design efﬁcient and effective lighting systems. Therefore, through HCL, the Design discipline can become an enabler for the diffusion of technology and a very strong lever for Social Inclusion. Acknowledgments. The paper showed the preliminary design results of the M.Sc. thesis in Architecture of the ﬁrst author. However, this paper is the result of a collective writing process of all authors. The writing of ‘Concept Design: The Domestic Lighting Systems’ can be attributed to the ﬁrst author; ‘Abstract’ and ‘Seniors and Human Centric Lighting’ to the second author; ﬁnally, ‘Aims’ and ‘Conclusions’ to the third author.

References 1. Kohlbacher, F., Herstatt, C. (eds.): The Silver Market Phenomenon: Business Opportunities in an Era of Demographic Change. Springer, New York (2008) 2. Bandini Buti, L.: Ergonomia Olistica: Il Progetto per la Variabilità Umana. Milan: Franco Angeli 3. Human Centric Lighting Society. http://humancentriclighting.org/ 4. Deli, A.: Human Centric Lighting approach: concept design di un sistema di illuminazione per il settore elderly. M.Sc. thesis in Architecture, University of Chieti-Pescara, Italy (2018) 5. Sixsmith, A., Gutman, G. (eds.): Technologies for Active Aging. Springer, New York (2013)

Human Diversity and Organizational Culture Barbara Mazur(&) Faculty of Management, Technological University of Lublin, Lublin, Poland [email protected]

Abstract. The aim of this paper is to examine and explore the relationship between, religion, culture and organizational culture at the level of the basic underlying assumptions. Culture and religion are believed to be influential sources of basic cultural assumptions. They are reflected in organizational culture. The hypothesis of this research assumes that religion has a meaningful impact on the basic assumptions of organizational culture. Religion in this research is understood as a cultural factor with the following dimensions: individualism/collectivism, uncertainty avoidance, power distance, and masculinity/femininity), The main aim of the paper is to base oneself on the level of basic cultural assumptions to construct a model of organizational culture. This model will later be tested by a two-tiered veriﬁcation process., which engages companies based and operating daily in varied religious surroundings. The scope this research has adopted includes organizational cultures of two enterprises. Both of them are based in the Podlasie region in Poland. This research is based on the method of a comparative case study developed on the basis of quantitative techniques in research. Keywords: Organizational culture Culture

Basic assumptions Religion

1 Introduction In research studies conducted so far it has been believed and shown that one of the major and most signiﬁcant factors critically influencing organizational culture is religion. Strong and numerous cases in research can be found to prove that religious beliefs play a major role in one’s behavior, including such aspects as economic performance. Strongly convincing evidence coming from past research conﬁrms the existence of a tie between religion and varied attitudes displayed in the work environment. This is especially important when talking about the notions of job satisfaction, motivation and (most importantly) organizational commitment [11]. In this study, a model of basic assumptions in organizational culture is presented. Those basic assumptions, as understood in this study, are externally derived in most part from the surrounding culture and religion. There exists a cognitive component in organizational culture. It is made of beliefs, values, assumptions, attitudes and norms which are mutually shared between the organization’s members. This cognitive component tends to shape the organization’s members’ mental schemes [2] and plays a vital role in the process of assigning meaning as well as preparing reactions to particular situations taking place within the organization. Therefore, organizational culture tends © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 540–546, 2019. https://doi.org/10.1007/978-3-030-02053-8_82

Human Diversity and Organizational Culture

541

to create a unique set of values and assumptions which serves as a point of reference to its users. This reference frame can direct the actions, interpretations, and even perceptions of the members of the organization [15]. Hence, it can exercise an immediate impact on any given process within the organizations. Therefore, it can affect its overall performance. Culture within an organization is rooted deeply and is built upon assumptions commonly shared between its members. Organizational culture is in fact dependent on the culture-producing forces which are external to the organization itself. What is more, there are similarities or variations within organizational cultures which can be explained by correlated similarities or variations within culture on the societal level [1]. Consequently, perceiving an organization as a “cultural island,” isolated from its surroundings, an approach more grounded in the real-life seems to be one considering the organization and its culture as a nexus which draws from all of the societal dimensions of culture [10].

2 Basic Assumptions of Organizational Culture and Their Dimensions In order to investigate organizational culture, the structural model of culture put forward by Schein [14] was adopted as framework. The advantage of opting for this model in the research is that it received little criticism and that it has been operationalized before already. In this research, the understanding of organizational culture is one of it being a speciﬁc constellation of commonly shared basic assumptions. Those assumptions came to light during the lifespan of a group who, during the course of facing external difﬁculties and working towards internal integration, deemed them useful, true and valid, and worth of teaching to the other (new) members within the group. This process builds group attitudes, reactions and feelings towards particular phenomena [14]. Numerous efforts to assess cross-cultural dimensions could have been observed already. The most widely used framework for this has been the one developed by G. Hofstede. Its popularity and usefulness is due to the fact that it became the ﬁrst large-scaled and widely-praised study of culture. Its reviewers valued its research design, the way the data was collected, and the comprehensive and wholesome theory explaining national variations between cultures. Hofstede has identiﬁed and empirically tested four dimensions which are used to create distinctions between particular cultures. Additionally, four values have been often times used in order to understand and explain variations in styles of leadership [12] and managerial skills [16]. Hofstede’s dimensions have been already used in order to observe, examine and explain differences in basic underlying assumptions of different organizational cultures [6, 8, 9]. The Hofstede Model of Cultural Dimensions has been adopted in numerous research to analyze country’s culture. The model assumes that each dimension is based upon basic assumptions [4, 7]. Those cultural dimensions are presumed to be the following: – Individualism: it considers one’s identity either as based only on the individual or based on the individual who is a part of a group.

542

B. Mazur

– Power distance: it is a way of dividing the society into groups and tiers because of status. Those with higher status are assumed to possess more power by those with lower status. – Uncertainty avoidance: it reflects the group’s aversion to unknown or risky situations or actions. – Masculinity (assertiveness): it shows the group’s preference towards masculine values of competition and achieving aims and outcomes in contrast to feminine values of dedication to process and co-operation. Based in the assumption that societal culture is influenced by religion, and as such directly impacts the organizational culture, the model of basic underlying assumptions of organizational culture was constructed. This model is presented in Fig. 1 below.

Fig. 1. Model of basic assumptions organizational culture; Source: [9, p. 70].

This thorough model embraces the assumptions of organizational culture and can thus be applied in describing organizations.

3 Examining of Basic Assumptions in Organizational Cultures This research, which involves religion perceived to be an independent factor influencing organizational culture [5, 17], has been conducted in the Podlasie region in Poland. Podlasie lies in the North-Eastern part of the country and it is the living habitat to Catholic and Orthodox Churches believers. Those two ethnic groups make up for the largest in the entire region, with Catholics standing for 77% and Orthodox standing for 13,5% of the inhabitants. Within the Podlasie region, there are districts where the vast majority of the inhabitants represents the Orthodox Church, like Hajnowka with 80% of population declared Orthodox believers [13]. The research sample in this research was chosen purposefully. The comparative study conducted included the scope of two companies. One of the companies is based

Human Diversity and Organizational Culture

543

in the Catholic environment while the other one – in Orthodox. One is located in a part of the Podlasie region where 99,7% of the inhabitants declares Catholicism, while the other one – in a region where Orthodox believers constitute 60,6% of the population, Catholics 25,3% accordingly, and 14,1% declare other religions or atheism. One of the companies employs 300 workers while the other – 51. 99 employees in total took part in the research. 64 of them were Catholic while the remaining 35 represented the Orthodox confession. In this study, case research method was adopted. It is believed to be of particular use especially in the business environment [3, 18]. Data for this research was collected with the use of a questionnaire. The questionnaire included various statements which reflected the four cultural dimensions. Each dimension in the questionnaire was represented by two extreme statements. In between those two extreme statements, there was a scale which the respondent was to use in order to indicate which one of the statements resonates with his/her personal views more (where his/her personal preferences lie). The data gathered through the questionnaire were later analyzed statistically with the application of the U Mann-Whitney test. The test allowed the differences between the gathered answers to be outlined. The main objective of the statistical analysis was to discover whether there occurred statistically valid differences between the answers given by the two groups of respondents representing the two Christian confessions – Catholicism and Orthodox. U Mann-Whitney test was also used when comparing the medians within the analyzed groups. The entire research was based on a zero hypothesis which assumed that both research groups came from a wider population with the very same median. The alternative hypothesis assumed that analyzed groups come from different populations with differ medians. The zero hypothesis would therefore suggest that the Catholic and Orthodox environment do not influence the researched employees in the two companies and that their answers do not vary signiﬁcantly in regards to the four cultural dimensions. The alternative hypothesis suggests the contrary – that the employees of the companies embedded in the Catholic and Orthodox surroundings do vary signiﬁcantly statistically when it comes to the answers they have given in regards to the four cultural dimensions. The results of the statistical analysis suggested the rejection oh hypothesis zero in favor of the alternative. As the results or the research, ﬁgures portraying basic assumptions of organizational cultures have been created for both: the Catholic and Orthodox variations included in this study. The dimensions are depicted in Fig. 2 below. What resulted of this study is a conclusion with regards to the organizational cultures influenced by the local culture and confession. Namely, while comparing the two separate organizational cultures, in the companies whose organizational culture is derived from the Catholic culture environment, the characteristics of this organizational culture are: higher levels of masculinity and individualism as well as lower levels of power distance and uncertainty avoidance. Consequently, in the organizational cultures derived by the Orthodox-dominated surroundings, the main characteristics tend to be: higher level of collectivism and femininity and lower levels of uncertainty acceptance. Veriﬁcation of the Research Findings. To further test the constructed model of organizational culture, one more study within the two companies took place. The goal of the second research was to analyze the basic assumptions of the organizational

544

B. Mazur

Key: Strong influence Weak influence

Fig. 2. Models of basic assumptions of organizational cultures influenced by confessions. Source: [9, p. 125].

cultures of the two companies. This research implemented a survey which was composed of 8 statement that were in opposition. Each cultural dimension of the four had two statements aligned to it. The survey was completed by the same employees as the ﬁrst research. The respondents were asked to rank on a scale (in the same manner as in the previous research) between the two extremes. However, this time, the respondents were asked to indicate how strongly they feel the presence of one dimension in their companies. Individualism/Collectivism. The results of the research again conﬁrmed the usefulness and validity of the models of organizational culture surrounded by particular confessions. Additionally, the results helped to further prove the existence of statistically valid and important differences within the organizational cultures with regards to the cultural dimensions. Overall, the usefulness of the constructed models of organizational cultures surrounded by particular confessions was proven. Despite the examined differences between the basic assumptions in the organizational cultures, there exist also numerous similarities. It is therefore essential to bear in mind that while the differences might stand out more, the similarities constitute the organizational culture just as well.

4 Conclusion The culture of an organization is an exceptionally interesting ﬁeld for research and it has been an ever more popular one. The concept of organizational culture is vital to analyze and worth of investigating more mostly due to the fact of its persistent, exceptional and original impact on the behavior and thus performance not only of the organization’s members, but the organization as a whole.

Human Diversity and Organizational Culture

545

Its shape is determined in a large part by the basic assumptions of social culture. Social culture is influenced by the dominant religion in a given area. The role of religion as a culture-creating factor differentiating the dimensions of culture has been proven in the study presented in the article. It is of utmost importance to understand how societal culture can impact organizational culture. In cases where the national and organizational cultures do not coincide, there might arise problem areas and internal tensions. Those tend to lead to difﬁculties of the functioning of the organization as well as its performance. Future research ideas and recommendations include segmenting the research sample in order to be able to investigate such factors as age, profession or one’s identity. This further division would allow determining where culture and its subcultures have their boundaries while having the societal culture in mind. Additionally, it could be useful to distribute the same questionnaires again at a later point in time in order to establish whether national cultural identity changes as the time goes by. This repeated research could also serve as basis for research of the notions and effects that globalization and developing ICT has on the population, especially ton those employees who grew up surrounded by it.

References 1. Alvesson, M.: The play of Metaphors. In: Hassard, J., Parker, M. (eds.) Postmodernism and Organizations. Sage, London (1993) 2. Alvesson, M.: Understanding Organizational Culture. Sage Publications, London (2002) 3. Buchanan, M., Iyer, R., Karl, C.A.: The case study in business research.iyer “Global Research Business” (1999). http://www.globalresearchbusiness.com 4. Gahan, P., Abeysekera, L.: What shapes an individual’s work values? An integrated model of the relationship between work Values, national culture and self-construal. Int. J. Hum. Res. Manage. 20(1), 126–147 (2009) 5. Hofstede, G., Hofstede, G.J.: Cultures and Organizations: Software of the Mind. McGrawHill, New York (2005) 6. Karczewski, L.: Etyka biznesu: kulturowe uwarunkowania. Oﬁcyna Wydawnicza Politechniki Opolskiej, Opole (2008) 7. Kostera, M.: Zarządzanie międzykulturowe. In: Koźmiński, A.K., Piotrowski, W. (eds.), Zarządzanie. Teoria i praktyka, pp. 596–599. Wydawnictwo Naukowe PWN, Warszawa (2010) 8. Mazur, B.: Cultural diversity in organizational theory and practice. J. Intercultural Manag. 2 (2), 5–15 (2010) 9. Mazur, B.: Kultura organizacyjna w zróżnicowanym kulturowo otoczeniu. Oﬁcyna Wydawnicza Politechniki Białostockiej, Białystok (2012) 10. Meyerson, D., Martin, J.: Cultural change: an integration of three different views. J. Manag. Stud. 24(6), 623–647 (1987) 11. Parboteeah, P.K., Paik, Y., Cullen, J.B.: Religious groups and work values: a focus on Buddhism, Christianity, Hinduism, and Islam. Int. J. Cross Cult. Manage. 9(1), 51–67 (2010) 12. Pavett, C., Morris, T.: Management styles within a multinational corporation: a ﬁve country comparative study. Hum. Relat. 48(10), 1171–1191 (1995)

546

B. Mazur

13. Sadowski, A.: Zróżnicowania kulturowe mieszkańców województwa podlaskiego. In: Krzysztofek, K., Sadowski, A. (eds.) Pogranicza i multikulturalizm w warunkach Unii Europejskiej: implikacje da wschodniego pogranicza Polski, t.1. Wydawnictwo Uniwersytetu w Białymstoku, Białystok (2004) 14. Schein, E.H.: Organizational Culture and Leadership. Jossey-Bass Publisher, San FranciscoWashington-London (1992) 15. Schein, E.H.: Organizational Culture and Leadership. Wiley, New York (2004) 16. Shipper, F., Kincaid, J., Rotondo, D.M., Hoffman, R.C.: A cross-cultural exploratory study of the linkage between emotional intelligence and managerial effectiveness. Int. J. Organ. Anal. 11, 171–191 (2003) 17. Smircich, L.: Concepts of Culture and Organizational Analysis. Administrative Science Quarterly No. 28/3 (1983) 18. Yin, R.K.: Case Study Research Design and Method. Sage Publications, London (2002)

Predictors of Preference for the Activity-based Flexible Ofﬁce Linda Rolfö1(&), Helena Jahncke2, Lisbeth Slunga Järvholm3, Maria Öhrn3, and Maral Babapour4 1

2

Department of Ergonomics, KTH Royal Institute of Technology, Huddinge, Sweden [email protected] Centre for Musculoskeletal Research, Department of Occupational and Public Health Sciences, University of Gävle, Gävle, Sweden 3 Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden 4 Department of Industrial and Materials Science, Chalmers University of Technology, Gothenburg, Sweden

Abstract. Activity-based Flexible Ofﬁces (A-FOs) are implemented with varying degree of success. Employees relocate from cell or open-plan ofﬁces, from different organizational backgrounds, varying design and implementation processes, and have different types of work tasks. This study aims at investigating whether preference for the A-FO correlate with these preconditions. The results from Chi-square tests and Spearman’s non-parametric correlation of postrelocation questionnaires distributed to 11 A-FO sites, showed that a high preference for the A-FO correlated strongest with an A-FO preference prior to relocation, being a former open-plan ofﬁce occupier and with frequent performance of innovation. Low preference for the A-FO correlated with frequent performance of concentration demanding tasks. Working with tasks with high conﬁdentiality did not predict the preference ratings. Keywords: ABW

Work activities Planning process

1 Introduction The Activity-based Flexible Ofﬁce (A-FO) concept is implemented by organizations worldwide [1]. A-FOs are non-territorial ofﬁces that provide various ofﬁce settings and workstations on a ﬁrst-come-ﬁrst-served basis [2]. Suggested preconditions for appraisal of the A-FO concept are ofﬁce type prior to relocation, organizational context and workspace design process, as well as type of work activities [3–6]. Ofﬁce type prior to relocation to A-FOs, such as the cellular ofﬁce or the open-plan ofﬁce, has been suggested to influence appraisal [3], such as preference for the A-FO. It is suggested that former open-plan ofﬁce occupiers are likely more positive towards the A-FO concept than those relocating from cellular ofﬁces as they are more accustomed to the open character of A-FOs [3].

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 547–553, 2019. https://doi.org/10.1007/978-3-030-02053-8_83

548

L. Rolfö et al.

The internal organizational context [4, 7] and workspace design process [1, 5, 8] are other suggested influencers of employees’ ofﬁce preference. For example, a bureaucratic culture with hierarchical power conditions may be less suited to implement AFOs compared to innovative or supportive cultures [4]. Furthermore, during the workspace design process organizations have opportunity to design ofﬁces and determine a usage that support employees’ work and match employees’ preferences [8, 9]. A design and implementation process that is perceived meaningful and manageable increases acceptance of change [10], and may, therefore, change employee’s negative preconceptions of A-FOs to positive. An indicator of the suitability of the internal organizational context or the outcomes of the workspace design process is employee attitudes towards the A-FO concept prior to relocation to A-FOs. Memory-demanding tasks, decision-making, learning and searching for information, and gaining an overview of multiple things are typical tasks conducted by ofﬁce workers. Type of work task and cognitive demands as predictors for preference for AFOs have not been investigated in previous research on A-FOs. In summary, cross-case comparisons of A-FO implementations are sparse [exceptions are 2, 5]. Therefore, the extent of preference for the A-FO between cases remains unclear. Suggested preconditions for preference for the A-FO are (1) ofﬁce type prior to change, (2) design process and internal context-related factors, and (3) task-related factors. However, these preconditions have not been further studied. The aim of this study is to investigate whether preference for the A-FO can be predicted by some of these preconditions. More speciﬁcally the research questions (RQ) are: RQ 1: Does employee preference for the A-FO vary between case sites? RQ 2: Does preference for the A-FO correlate with (a) ofﬁce type prior to relocation (b) ofﬁce type preference prior to relocation, and/or (c) work tasks?

2 Method A post-relocation questionnaire was sent to 11 A-FO case sites in Sweden with a total of 780 employee responses (response rate 70%), 47% women, 53% men, between 2–18 months after relocation (Table 1). The cases were a mix of non-proﬁtable, private, public and municipality organizations. Case sites 8–11 belonged to the same public organization. Case site sizes varied between 40 and 228 employees. Table 1. Case organizations’ demographics Case

C1

Org. type

Private Private Nonprof.

C2

Private Private Municipality Municipality Public Public Public Public

Quest. distribution (months)

3

6

9

C3

2

C4

C5

6

C6

12

C7

18

C8

12

C9

12

C10

12

C11

12

Site size

100

50

40

160

65

158

228

98

144

46

226

Questionnaire invitations

79

50

40

33

65

158

179

98

144

46

226

Respondents Females/Males

66 32/34

46 6/40

36 28/8

24 17/7

40 23/17

91 73/18

152 50/102

57 23/34

85 25/60

35 24/11

148 67/81

Response rate

84%

92%

90%

73%

58%

58%

85%

60%

64%

76%

66%

Predictors of Preference for the Activity-based Flexible Ofﬁce

549

Questions regarded ofﬁce type (11 cases) and preference (9 cases) prior to relocation, ofﬁce type (11 cases) and preference (11 cases) after relocation, and task related demands: memory, innovation, decision making, learn/search for information, generate texts, count, overview multiple things, speak on the phone (10 cases), secrecy (9 cases) and concentration (6 cases). The seven response categories for questions on ofﬁce type preference and ofﬁce type prior to relocation were merged into two categories; “A-FO” or “other ofﬁce type”, and “Cell” and “Open-plan ofﬁce” (rooms with two occupiers or more) respectively. The response categories on questions for task related demands were merged into “always/often” and “sometimes/seldom/never”. To investigate case variation for preference (RQ 1), analysis of variance of the proportions with a 0.05 level of signiﬁcance was performed in SPSS. To investigate correlations for the whole population (RQ 2), bivariate comparisons were performed using Chi-square tests and Spearman’s non-parametric correlation. Gender and age correlations with the outcome variable were checked.

3 Results 3.1

Case Comparison

The Chi-square test showed that 35% of the employees across all cases preferred the A-FO before other ofﬁce types after relocation. However, the analysis of variance of the proportions showed that the percentage of employees preferring the A-FO after relocation varied signiﬁcantly between cases (Table 2). In four cases less than 20% of the employees preferred the A-FO before other ofﬁce types, while in four cases more than 60% of the employees preferred the A-FO. Table 2. Percentage of employees in the different cases, preferring the A-FO before other ofﬁce concepts after relocation.

3.2

Ofﬁce Type Prior to Relocation

Bivariate comparisons using Chi-square test and Spearman non-parametric correlation showed that ofﬁce type prior to relocation correlated signiﬁcantly with preference for the A-FO after relocation (R = 0.19). The proportion of employees preferring the A-FO was signiﬁcantly larger (p < 0.001) amongst the employees who had relocated from open-plan ofﬁces (44.0%) than those who had relocated from cell ofﬁces (27.5%).

550

3.3

L. Rolfö et al.

Ofﬁce Type Preference Prior to Relocation

The Chi-square and Spearman correlation test also showed that preference for the AFO prior to relocation correlated signiﬁcantly with preference for the A-FO after relocation (R = 0.585). The proportion of employees preferring the A-FO was signiﬁcantly larger (p < 0.001) amongst the employees who had preferred the A-FO prior to the relocation (54.5%) than those preferring other ofﬁce types prior to relocation (45.5%). Of the employees who had preferred other ofﬁce types prior to relocation, 21% started preferring the A-FO after relocation. In the opposite direction, of the employees who had preferred the A-FO prior to relocation, 12% started preferring other ofﬁce types. 3.4

Work Tasks

The Chi-square and Spearman correlation test showed that preference for the A-FO after relocation correlated signiﬁcantly with often performing tasks involving innovation. Out of the employees preferring the A-FO, 80% reported to often or always have innovation demanding tasks. Out of the employees preferring other ofﬁce types only 67% reported to have innovative tasks. The proportions varied signiﬁcantly (R = 0.140, p < 0.001). The analysis also showed that preference for the A-FO after relocation correlated signiﬁcantly but inversely with often performing concentration demanding tasks, text generating tasks and often speaking on the phone. The proportion of employees that often or always performed concentration demanding tasks was larger amongst the employees preferring other ofﬁce types (86.2%) than those (72.6%) preferring the A-FO (R = 0.156, p = 0.001). Next, the proportion of employees that often or always worked with writing or text generation was signiﬁcantly larger amongst the employees preferring other ofﬁce types (64.2%), than those (55.8%) preferring the A-FO (R = 0.081, p = 0.03). Furthermore, the proportion of employees that often or always spoke on the phone was larger amongst the employees preferring other ofﬁce types (71.1%) than those (59.9%) preferring the A-FO (R = 0.11, p = 0.008). There was no correlation between preference for the A-FO and tasks involving memory demands (n = 715, p = 0.22), decision making (n = 568, p = 0.33), counting (n = 714, p = 0.030), overviewing multiple things (n = 568, p = 0.94), conﬁdential information (n = 656, p = 0.19) and learn and search for information (n = 713, p = 0.058). There was no signiﬁcant correlation between preference and the individual factors age (p = 0.08) or gender (p = 0.69). In summary, the factor with highest correlation coefﬁcient for preference with the A-FO was preference for the A-FO prior to relocation. The factors with signiﬁcant correlations are shown in Table 3.

Predictors of Preference for the Activity-based Flexible Ofﬁce

551

Table 3. Factors signiﬁcantly correlating with preference for the A-FO after relocation. The preferred/advantageous ofﬁce type for the factor is bolded. Factor Preference for the A-FO prior to relocation Ofﬁce type prior to relocation Frequent task: innovation Frequent task: concentration Frequent task: speaking on the phone Frequent task: write/generate texts

Response categories A-FO/Other ofﬁce types Cell ofﬁce/Open-plan ofﬁce A-FO/Other ofﬁce types A-FO/Other ofﬁce types A-FO/Other ofﬁce types A-FO/Other ofﬁce types

n 505 679 714 473 568 714

R 0.59 0.19 0.14 0.16 0.11 0.08

4 Discussion This study investigated differences in preference for the A-FO between 11 case sites and correlations of this preference with ofﬁce type prior to relocation, ofﬁce type preference prior to relocation and type of work tasks frequently performed. The large difference in preference per case site implies that there is a dominant case effect contributing to A-FO preference, rather than differences on an individual level. A higher percentage of employees relocating from open-plan ofﬁces, rather than cell ofﬁces, preferred the A-FO. Earlier research comparing ofﬁce types has shown that open-plan ofﬁce occupiers have lower employee satisfaction and performance in terms of concentration, memory and learning due to reduced auditory, visual and informative privacy [11, 12]. Therefore, employees from open-plan ofﬁces may perceive more environmental support for their work by the provision of break-out spaces and quiet zones in A-FOs [cf. 13], and for cell ofﬁce occupiers it is perhaps a more drastic change coming from an environment with few distractions. The correlation between pre- and post-relocation preferences indicates that the employees who accepted the change and perceived it as meaningful [cf. 10] prior to the change took place, also preferred the A-FO after relocation. In other words, the implementation of the A-FO seemed to fulﬁll employees’ expectations, which is why they did not change their preferences to other ofﬁces after relocation. This is supported by Nielsen and Randall [14], implying that the design and implementation process influences the outcomes of an intervention. The correlation of pre- and post-relocation preferences may, however, also be due to organizational preconditions, such as a supporting culture [cf. 4], working procedures congruent with the activity-based working and power relations [8]. The results indicate that A-FO do not to support focus-demanding tasks. It would have been interesting to also investigate whether the A-FO supports communicative tasks. Further, results showed that performing tasks with conﬁdential information did not predict the preference ratings. However, conﬁdentiality of information may vary as well as access to environments supporting secrecy. Other factors may also be considered for evaluating preference outcomes. In a recent study it was shown that the A-FO did not support certain work activities due to e.g. high employee-to-workstation ratio [6]. The ﬁt or misﬁt between the work environment and work activities may be a strong contributor to preference rather than the

552

L. Rolfö et al.

work activity itself [15]. In addition, some cases might not have applied speech or clean desk rules, which may create a less supportive work environment [2]. Since this study evaluates as much as 11 cases, the questionnaires were not completely identical and were carried out at different times after the A-FO implementation. Furthermore, although reliability may have been reduced, results are based on subjective data. However, subjective ratings are a viable option for studying natural settings and a way to explore general patterNs.

5 Conclusion Employee preference for the A-FO varied between 13–92% for the case sites. Preference for the A-FO correlated with frequent performance of tasks involving innovation, however, inversely with frequent performance of concentration demanding tasks. Many A-FO sites seem also to be designed for supporting innovation rather than concentration demanding tasks. Secrecy tasks did not predict the preference ratings. Moreover, former open-plan ofﬁces occupiers preferred the A-FO to a signiﬁcantly higher extent than former cell ofﬁce occupiers. The employee’s ofﬁce type preference priori relocation was, however, the strongest predictor of their post-relocation preference. Almost all employees preferring the A-FO priori relocation continued to prefer the A-FO after relocation. To make employees more positive towards the A-FOs seem to require a thorough design and implementation process to make the change ﬁt the organizational context, which also need to be based on which ofﬁce design the employees are coming from, as well as the tasks frequently performed.

References 1. Appel-Meulenbroek, R., Groenen, P., Janssen, I.: An end-user’s perspective on activitybased ofﬁce concepts. J. Corp. Real Estate 13(2), 122–135 (2011) 2. Rolfö, L., Babapour Chaﬁ, M.: Policies for sharing workspaces in activity-based flex ofﬁces. In: ACE-ODAM (2017) 3. Riratanaphong, C., Van Der Voordt, D.: Performance Measurement of Workplace Change: A Comparative Analysis of Data from Thailand, the Netherlands and Finland. The Added Value of Facilities Management: Concepts, Findings and Perspectives. Polyteknisk Forlag, Lyngby, Denmark (2012) 4. Wohlers, C., Hertel, G.: Choosing where to work at work – towards a theoretical model of beneﬁts and risks of activity-based flexible ofﬁces. Ergonomics 60(4), 467–486 (2016) 5. Brunia, S., de Been, I., van der Voordt, T.J.: Accommodating new ways of working: lessons from best practices and worst cases. J. Corp. Real Estate 18(1), 30–47 (2016) 6. Rolfö, L., Eklund, J., Jahncke, H.: Perceptions of performance and satisfaction after relocation to an activity-based ofﬁce. Ergonomics, 61(5), 644–657 (2018). https://doi.org/10. 1080/00140139.2017.1398844 7. Jacobsen, D.I.: Organisationsförändringar och förändringsledarskap, 2(5), 347 (2013). Studentlitteratur, Lund 8. Rolfö, L.: Relocation to an activity-based flexible ofﬁce – design processes and outcomes. Appl. Ergon. 73, 141–150 (2018). https://doi.org/10.1016/j.apergo.2018.05.017

Predictors of Preference for the Activity-based Flexible Ofﬁce

553

9. Eklund, J., Daniellou, F.: Ergonomics and project management 1: Important aspects in the planning of the project. In: Proceedings of the 11th Congress of the International Ergonomics Association, Paris. Designing for everyone (1991) 10. Antonovsky, A.: The salutogenic model as a theory to guide health promotion. Health Promot. Int. 11(1), 11–18 (1996) 11. Jahncke, H., Halin, N.: Performance, fatigue and stress in open-plan ofﬁces: the effects of noise and restoration on hearing impaired and normal hearing individuals. Noise Health 14(60), 260–272 (2012) 12. Brennan, A., Chugh, J.S., Kline, T.: Traditional versus open ofﬁce design: a longitudinal ﬁeld study. Environ. Behav. 34(3), 279–299 (2002) 13. van der Voordt, T.J.: Productivity and employee satisfaction in flexible workplaces. J. Corp. Real Estate 6(2), 133–148 (2004) 14. Nielsen, K., Randall, R.: Opening the black box: presenting a model for evaluating organizational-level interventions. Eur. J. Work Organ. Psychol. 22(5), 601–617 (2013) 15. Babapour, M., Karlsson, M., Osvalder, A.-L.: Appropriation of an activity-based flexible ofﬁce in daily work. Nordic J. Work. Life Stud. 8(S3), 71–84 (2018)

Affective Design Approach to Mobile Security Authentication Daehee Park(&), Jaeyong Lee, Yenah Lee, and Scott Song Samsung Electronics, 56, Seongchon-gil, Seocho-gu, Seoul, Republic of Korea {daehee0.park,yenah.lee,jae-yong.lee, sangkon.song}@samsung.com

Abstract. For many years, affective design has been considered in the design of mobile phone functions for mobile phone manufacturers to differentiate their products compared to their competitors. Although, the newest Google Android OS mobile phone provides various biometric authentication methods to authenticate the mobile phone user, there are a few research regarding mobile authentication in the aspect of affect. In this project, we hypothesised that there might be different affects appearing between each mobile security authentication, especially biometric security authentication. Thus, participants were recruited to measure affect through the self-assessment manikin method using mobile security authentication from registering to executing on various situations. Then we analysed if there is a difference in the valence and arousal aspects. Finally, we propose affective values for each type of mobile security authentication and suggest how aspects of affective design could be improved. Keywords: Affective computing

Mobile authentication HCI

1 Introduction Although it is difﬁcult to fulﬁl user desires, most companies try to satisfy user needs [2]. Mobile phone brands are now trying to provide better user experience and affect to users instead of hardware distinctions; these positive experiences are vital for continued commercial success [10]. Although successful mobile phone user experience depends on a large number of factors, affective design is an important factor to consider. According to Jordan, users’ affective needs must be considered when designingproducts [7, 8]. For several years now, manufacturers have considered affective design to provide points of differentiation compared to their competitors. The newest Google Android OS mobile phone provides various ways to verify user identities such as ﬁngerprint identiﬁcation, iris recognition, face recognition, PIN number, password, pattern recognition, and Smart Lock [1]. In the aspect of user experience, it is important to ﬁnd out which method appears as priority to users and the reasons have not yet researched. Mobile phone users are required to register their security information to the mobile phone. The mobile phone then recognises the registered information to match the users’ identity and then allows them to access the mobile phone. These mobile authentication methods can be utilised in various areas in countries, for example, when using mobile © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 554–559, 2019. https://doi.org/10.1007/978-3-030-02053-8_84

Affective Design Approach to Mobile Security Authentication

555

banking services, payment services, and authenticating on speciﬁc applications. Thus, mobile security authentication can be expected to be used in many areas alongside user veriﬁcation. Affective design must be considered when designing user interfaces for mobile security authentication in order to provide more positive user experiences. According to Kuppens et al. [9], affect is regarded as a subjective experience in terms of emotion, mood, and other feelings. Kuppens et al. [9] proposed that these two properties are fundamental in describing the nature of affect. In this project, we hypothesised that different affects might appear between each mobile security authentication, especially biometric security authentication. We compared the volume of affect from biometric security authentication methods such as ﬁngerprint recognition, iris recognition, and face recognition to non-biometric authentication methods such as password. 1.1

Experiment Design

Ten participants who had been using various biometric mobile authentication for at least one year were recruited to measure their affective responses when doing so. They were familiar with all kinds of biometric mobile authentication. The independent variable for the experiment was the chosen method of biometric mobile authentication. The dependent variable was the SAM data [3–6, 11]. Participants were asked to complete several tasks using passwords, ﬁngerprints, face recognition and iris scanning. Initially, we compared biometric methods to passwords. Then we compared the SAM data between the various biometric authentication methods to see if there is any signiﬁcant difference between the methods. SAM used nine-point scales to measure the valence and arousal of participants when using biometric mobile authentication. 1.2

Methods

The participants were asked to perform several tasks in two contexts: a dynamic situation and a static situation. In a dynamic situation, the participants had to complete their task while walking. However, the static situation was more comfortable. In each situation, the participants were asked to register each mobile authentication method then authenticate the mobile phone three times using that particular method. They were then asked to send a text message to the experimenter. After completing this task in two situations, the participants were asked to choose their preferred mobile authentication method on the basis of personal preference, safety and recognition rate. They were also asked to describe the advantages and disadvantages of each authentication method.

2 Result The mean age was 30.85; gender wise, there were 6 male and 4 female participants. According to Table 1, there was a signiﬁcant difference between the authentication methods in the valence score when the participants registered each authentication methods whilst in a static state. It showed that face recognition provides the highest valence level (average 6.60).

556

D. Park et al. Table 1. Two-way ANOVA (Valence_Static_Registration) Source of variation SS df MS F P-Value F-crit Methods 21.68 3.00 7.23 4.23 0.01 2.96 Error 46.08 27.00 1.71 Total 111.8 39.00

Table 2 indicated that there was a signiﬁcant difference between the authentication methods in valence when the participants were using the mobile phone while they were in a static state. It showed face recognition to provide the highest valence level (average 6.10). Table 2. Two-way ANOVA (Valence_Static_Use) Source of variation SS df MS F P-Value F-crit Methods 30.90 3.00 10.30 3.10 0.04 2.96 Error 89.60 27.00 3.32 Total 171.1 39.00

Table 3 indicated that there was a signiﬁcant difference between the valence scores of the authentication methods when the participants registered each authentication methods while in a dynamic state. It showed that iris recognition provides the highest valence level (average 5.10). Table 3. Two-way ANOVA (Valence_Dynamic_Registration) Source of variation SS df MS F P-Value F-crit Methods 14.60 3.00 4.87 3.61 0.03 2.96 Error 36.40 27.00 1.35 Total 86.40 39.00

Table 4 indicated that there was a signiﬁcant difference between the valence scores of the authentication methods when the participants were using the mobile phone while in a dynamic state. It showed that face recognition provides the highest valence level (average 6.50).

Table 4. Two-way ANOVA (Valence_Dynamic_Use) Source of variation SS df MS F P-Value F-crit Methods 16.08 3.00 5.36 3.28 0.04 2.96 Error 44.18 27.00 1.64 Total 104.8 39.00

Affective Design Approach to Mobile Security Authentication

557

On the other hand, there was no signiﬁcant difference in the arousal scores when registering the authentication methods and using them. According to Table 5, participants chose ﬁngerprint recognition as the most preferable authentication method. This was also regarded as the safest method of authentication by the participants. In the recognition rate results, face recognition showed the highest rate. Table 5. Ranking of user preference, safety, recognition rate (self-assessment) Ranking 1 2 3 4

Preference Fingerprint Face Iris Password

Safety Fingerprint Password Iris Face

Recognition rate Face Password Fingerprint Iris

3 Discussion According to the results, there were signiﬁcant differences in all of the static situations in terms of valence. In the static situations, the participants were comfortable. This means that there was no pressure on them and they could focus on the authentications. The valence score of face recognition was highest, which is consistent with its high recognition rate. The participants judged face recognition as providing the highest recognition rate (Table 5). Fingerprint identiﬁcation presented the lowest valence score when it was registered in static situations. It was even lower than the password method. We asked the participants to try to register their ﬁngerprints to their mobile phone several times. To do this, the users had to press their ﬁnger against the ﬁngerprint scanner several times. This was a repetitive task, which led to a lower average score and higher variance score in valence. In the dynamic situation, iris recognition provided the highest valence score; however, it provided the opposite result in the Dynamic_Use situation. This was because the participants struggled to focus on the iris scanner while they were walking. Thus, authentication failed several times, which might have influenced their affective response. With respect to arousal score, there was no signiﬁcant difference between authentication methods. However, the arousal scores were higher in the dynamic situation than in the static one. Fingerprint recognition gave the lowest arousal score in the dynamic situation as it was a repetitive task, which tend to induce lower arousal scores. Through data analysis, we suggest a number of design improvements focused on the concept of affect. Firstly, a visual effect appears when users try to authenticate their identity using face recognition on their mobile phone. The Galaxy S8 does not provide any visual effect during face recognition. Galaxy phones only present the user’s home screen when authentication is successful. This led the participants to respond that although face recognition provides the highest recognition rate, it is preferred less than ﬁngerprint identiﬁcation. According to our in-depth interview results, most participants stated negatively that “We were not [made] aware [whether our] authentication was

558

D. Park et al.

successful.” Hence, we propose that designs should provide visual effects for users to recognise whether their authentication has been successful or not (Fig. 1).

[a]

[b]

Fig. 1. (a) Visual effect in face recognition. (b) Iris recognition guide using Emoji

Secondly, a face spot guide using emojis during iris recognition is provided. In Galaxy S8, the user should see two circles directly on the mobile phone display. If the mobile phone recognises the user’s iris successfully, authentication has been successful. At that stage, the mobile phone shows the face of the user. According to the results from the interviews, many participants complained that they worried that someone who unfamiliar people might see their face reflected on the mobile phone. It leads the user the feel shame. Thus, we propose that an emoji might replace the user’s real face. The user will see the emoji, and the phone will match his/her iris for authentication (Fig. 1). Thirdly, different authentication methods must be possible when a speciﬁc method fails. In the Galaxy S8, if the user fails to authenticate ﬁve times using their speciﬁc biometric authentication method of choice, such as iris recognition, then the mobile phone requires a different authentication method such as a PIN, pattern or password except biometric methods. The results showed that arousal score during dynamic use was higher than that during static use, even though there was no signiﬁcant difference between arousal score during dynamic-use and static-use. We assumed that it was difﬁcult to use iris scanning features while walking as users may struggle to match their eyes to the small circles on their mobile phone while walking. Thus, we propose that another biometric authentication method becomes available when the user fails to authenticate themselves using a speciﬁc biometric method. The newest Android OS platform provides various kinds of mobile authentication functions including several biometric methods. We hypothesised that there may be a signiﬁcant difference in the affective response between methods while users are using each method. We chose the SAM method to measure affective responses in several conditions regarding mobile authentication methods. Although there was a signiﬁcant different between the conditions in terms of valence, there was no signiﬁcant different between conditions in terms of arousal. We proposed three design improvements based on the results.

Affective Design Approach to Mobile Security Authentication

559

References 1. Smith, T.F., Waterman, M.S.: Authentication of users on mobile telephones–a survey of attitudes and practices. Comput. Secur. 24(7), 519–527 (2005) 2. Cross, N.: Engineering Design Methods: Strategies for Product Design, 3rd edn. Wiley, Chichester (2000) 3. Desmet P.: Designing Emotions. Ph.D Dissertation. Delft University of Technology, Nethelands (2002) 4. Desmet, P.M.A., Overbeeke, C.J., Tax, S.J.E.T.: Designing products with added emotional value; development and application of an approach for research through design. Des. J. 4(1), 32–47 (2001) 5. Forlizzi, J., Ford, S.: The building blocks of experience: an early framework for interaction designers. In: Proceedings of the 3rd Conference on Designing Interactive Systems: Processes, Practices, Methods, and Techniques, pp. 419–423 (2000) 6. Isomursu, M., Tähti, M., Väinämö, S., Kuutti, K.: Experimental evaluation of ﬁve methods for collecting emotions in ﬁeld settings with mobile applications. Int. J. Hum-Comput. Stud. 65(4), 404–418 (2007) 7. Jiao, J.R., Zhang, Y., Helander, M.: A Kansei mining system for affective design. Expert Syst. Appl. 30(4), 658–673 (2006) 8. Jordan, P.W.: The four pleasures-a framework for pleasures in design. In: Proceedings of Conference on Pleasure Based Human Factors Design, Groningen. Philips Design, The Netherlands (2000) 9. Kuppens, P., Tuerlinckx, F., Russell, J.A., Barrett, L.F.: The relation between valence and arousal in subjective experience. Psychol. Bull. 139(4), 917 (2013) 10. Kujala, S., Miron-Shatz, T.: Emotions, experiences and usability in real-life mobile phone use. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 1061–1070 (2013) 11. Morris, J.D.: Observations: SAM: the self-assessment manikin; an efﬁcient cross-cultural measurement of emotional response. J. Advertis. Res. 35(6), 63–68 (1985)

An Experimental Study on Relationship Between Intellectual Concentration and Personal Mental Characteristics Wakako Takekawa1(&), Kimi Ueda1, Shogo Ogata1, Hiroshi Shimoda1, Hirotake Ishii1, and Fumiaki Obayashi2 1

2

Graduate School of Energy Science, Kyoto University, Kyoto, Kyoto 606-8501, Japan {takekawa,ueda,ogata,shimoda, hirotake}@ei.energy.kyoto-u.ac.jp Panasonic Corporation, Kadoma, Osaka 571-8686, Japan [email protected]

Abstract. As a proposal of new diagnosis for mental diseases, this study focused on the relationship between intellectual concentration and personal mental characteristics. It is expected that the measurement of concentration characteristics may help the diagnosis of the mental disorders because the mental characteristics such as psychiatric disease, developmental disorder and behavioral feature are supposed to be closely related to their mental activity such as concentration. When analyzing the relationship, the characteristics of concentration are expressed as 36 feature values by analyzing answering time distribution of cognitive task, and the values of concentration were compressed to 5 main factors by principal component analysis. Then the combination of the factors and one of 36 parameters of mental characteristics were given to a decision tree analysis tool. Keywords: Intellectual concentration Decision tree analysis

Mental characteristics

1 Introduction In recent years, the number of patients with mental illness tends to increase [1]. In addition, it is difﬁcult to establish reliable criteria for the diagnosis of these diseases, and this difﬁculty is seen not only for mental disorders but also for developmental disorders and so on [2]. In this situation, it is desired to realize more reliable diagnosis which can provide proper treatment to potential patients. In this study, therefore, the authors focus on intellectual concentration and examine the possibility that it is closely related to their mental characteristics and its availability to one of the criteria for diagnosis of mental disorders. In addition, the relationships between the intellectual concentration and developmental disorders trend/personal characteristics are also investigates as one of the factors reflecting mental condition. Regarding quantitative evaluation of intellectual concentration, various studies have been conducted. One of them deﬁne that the intellectual concentration can be © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 560–566, 2019. https://doi.org/10.1007/978-3-030-02053-8_85

Relationship Between Intellectual Concentration and Personal Mental Characteristics

561

expressed as a state transition model as shown in Fig. 1 [3]. The model consisits of “Concentration state” and “Non-concentration state” [4]. In the concentration state, the cognitive resources are assigned to the target task, while they take rest for a while in the non-concentration state. The concentration state is also divided into two states which are working state and short-term pause state. Based on this concept, indices indicating the ratio of concentration time (Concentration Time Ratio; CTR) [5] and the depth of concentration (Concentration Depth Index; CDI) [6] were developed.

Fig. 1. Three state concentration model.

The purpose of this study is to conduct an experiment to gather the quantitative data of intellectual concentration and mental characteristics which are not only personal characteristics but also mental disorders and developmental disorders trend. And then the relationships between them are examined based on the collected data by statistical analysis.

2 Method In this study, both personal mental characteristics and answering time data of cognitive tasks which showed the characteristics of intellectual concentration were collected ﬁrst, and then statistical analysis to ﬁnd associations between the collected data was performed. This chapter describes the procedure of the experiment, the approximation of the concentration state using the answering time series of the cognitive task, and the outline of the statistical analysis. 2.1

Experimental Procedure

In this study, 236 university students of Kyoto University were recruited to join the experiment where their data of personal characteristics and answering time series of cognitive tasks were obtained. The participants received e-mails describing URLs to questionnaire survey pages about one week before the participation date of the experiment and they answered all the questionnaires via the internet in advance. The questionnaires will be explained in Sect. 2.2.

562

W. Takekawa et al.

The protocol of the experiment is shown in Fig. 2. “Comparison Task” developed by Ueda et al. was employed as the cognitive task in the experiment. It consists of problems of uniform difﬁculty requiring cognitive abilities to be used in ofﬁce work [7].

Fig. 2. Protocol of the experiment.

The data of 10 participants were omitted because they slept during the task and those of 226 participants were analyzed in total. 2.2

Survey on Personal Mental Characteristics

To investigate mental symptoms and personal characteristics, the following 6 surveys were conducted; (1) “BIS/BAS scale” that shows the tendency of behavioral inhibition and behavioral approach [8], (2) “Autism-spectrum Quotient (AQ)” that shows the degree of autism spectrum [9], (3) “Yatabe-Guilford Personality Inventory (YG)” that ﬁnely extracts personal characteristics [10], (4) “General Health Questionnaire (GHQ)” that shows whether to have tendency of neurosis [11], (5) ”Global Scale for Depression (GSD)” that shows the presence and the type of depression [12] and (6) “NEET/Hikikomori Risk Scale” that investigates psychological tendencies common to those having problems in social adjustment [13]. 2.3

Quantitative Evaluation of Intellectual Concentration

Figure 3 shows an example of the answering time histogram of the cognitive task. The distribution can be divided into concentration and non-concentration state, and then the former can be approximated by sum of two lognormal distributions; deeper concentration and shallower concentration, based on the concept of three state model [4] described in Sect. 2.1. In this approximation, the number of answers in the concentration state, the duration of concentration, the parameters of the lognormal distribution, and so on were calculated as feature values which expressed the intellectual concentration as shown in Table 1. In addition, some values were added to express the detail features of the intellectual concentration. First, the ratio of l2/l1 and r2/r1 were added since it could express the difference between deeper and shallower concentration. Then, for the values of *mark in Table 1, l2/l1 and r2/r1, the ratio between the value at SET1 and SET2 was added

Relationship Between Intellectual Concentration and Personal Mental Characteristics

563

Fig. 3. Histogram of answering time. Table 1. Feature values to express intellectual concentration. Feature value Meaning N The number of total answering N1 The number of answering while deeper concentration N2 The number of answering while shallower concentration T1 The time of deeper concentration T2 The time of shallower concentration CTR* Concentration time ratio MCTR Multi concentration time ratio CDI* Concentration depth index l1 ; r1 Paratemer of log-normal distribution showing deeper concentration l2 ; r2 Paratemer of log-normal distribution showing shallower concentration

since it could express the changes in intellectual concentration and could be quantiﬁed when resuming work before and after a short break. In this way, total 36 feature values were used as explanatory variables in the later analysis. 2.4

Decision Tree Analysis

As an analytical method to investigate the relationship, decision tree analysis was used in this study. The greatest advantage of this method is that it is easy to visually judge the elements which are important or not for the conclusion, and the result can be easily interpreted.

564

W. Takekawa et al.

The results of personal mental characteristics survey were used for the objective variable. Regarding AQ, GHQ and GSD, the presence or absence of symptoms judged from the score was set as objective variables, and regarding BISBAS, YG and NEET/Hikikomori, scores themselves were set. On the other hand, the indices of intellectual concentration were set as explanatory variable. They were standardized and then compressed into ﬁve indices by principal component analysis, because the number of total feature values, 36, is too numerous and effective results can not be obtained. The contents indicated by the ﬁve principal components were (1) answering time, (2) deeper concentration, (3) improvement in the work speed or concentration depth after the break, (4) shallower concentration, (5) decrease in the concentration after the break. The procedure of the above analysis is shown in Fig. 4. “Scikit-learn” was used for the analysis, which is open source machine learning library in Python developed by Pedregosa et al. [14].

Fig. 4. Procedure of analysis.

3 Result and Discussion The analysis result was expressed as an inverted tree-like ﬁgure. Figure 5 shows an example of the analysis result where AQ was set as an objective variable.

Fig. 5. Simpliﬁed ﬁgure of the result - AQ.

In this example, the branching condition is True when the 2nd principal component is 0.983 or less, and it is False when it is larger than 0.983. The 2nd principal

Relationship Between Intellectual Concentration and Personal Mental Characteristics

565

component is the component indicating the degree of deeper concentration as described in Sect. 2.3, so 82 participants that branched to the False side have relatively large degree of deeper concentration. Then focusing on the proportion with autistic tendency, 25 (11.1%) of 226 participants have the tendency in total, while only 3 (3.7%) out of 82 who branched to False side. In other words, it is supposed that a person with a relatively deeper concentration is likely not to have the autistic tendency. Other tendencies were also found as the results of the analysis as shown below; (a) People with relatively high concentration are likely to be obedient. (b) People with relatively low concentration are likely to have a temperament like job-hopping parttimers. (c) People who get slow in work speed after a break are likely to be emotional. (d) People who improve their concentration after a break are likely to be conﬁdent, outgoing and active. On the other hand, no notable relationship was found regarding with GHQ and GSD. From the above analysis, some personal mental characteristics which were likely to be related to intellectual concentration were extracted.

4 Conclusions In this study, an experiment to collect quantitative data of intellectual concentration and personal mental characteristics was conducted. And then the relationships between them were examined based on the collected data using statistical methods. As the results, it was found that some mental characteristics were related to the intellectual concentration. At the current stage, however, it is difﬁcult to practically use the indices of intellectual concentration as one of the diagnostic basis for symptoms or criteria for personal characteristics. Acknowledgments. This work was supported by JSPS KAKENHI Grant Number JP17H01777.

References 1. Ministry of Health, Labor and Welfare: Measures for Mental Health Depression. http:// www.mhlw.go.jp/kokoro/nation/dyp.html 2. Kimura, Y.: “Developmental disabilities” as medicalization: focusing the process of interpretation in the educational setting. J. Educ. Sociol. 79, 5–24 (2006) 3. Miyagi, K., Kawano, S., Ishii, H., Shimoda, H.: Improvement and evaluation of intellectual productivity model based on work state transition. In: 2012 IEEE International Conference on Systems, Man, and Cybernetics, pp. 1491–1496 (2012) 4. Shimoda, H., Ooishi, K., Miyagi, K., Uchiyama, K., Ishii, H., Obayashi, F., Iwakawa, M.: An intellectual productivity evaluation tool based on work concentration. In: 15th International Conference on Human-Computer Interaction, vol. 16, pp. 364–372 (2013) 5. Uchiyama, K., Ooishi, K., Miyagi, K., Ishii, H., Shimoda, H.: Process of evaluation index of intellectual productivity based on work concentration. In: Proceedings of ICSTE (2013) 6. Ueda, K., Shimonaka, S., Shimoda, H., Ishii, H., Obayashi, F.: Quantitative evaluation of intellectual productivity considering depth of concentration. In: 2017 IEEE International Conference on Systems, Man, and Cybernetics, pp. 758–763 (2017)

566

W. Takekawa et al.

7. Ueda, K., Shimoda, H., Ishii, H., Obayashi, F., Taniguchi, K.: Development of a new cognitive task to measure intellectual concentration affected by room environment. In: The Fifth International Conference on Human-Environment System (2016) 8. Yasuda, A., Sato, A.: Development of the behavioral inhibition system/behavioral approach system scales. Jpn. J. Psychol. 73(3), 234–242 (2002) 9. Wakabayashi, A.: Guidebook for Autism-Spectrum Quotient Adults’ version in Japan. Sankyobo (2016) 10. Yagi, T.: Diagnostic Manual of YG Test -Utilization of Personality Survey in Personnel Management. Institute for Psychological Testing, Inc. (1989) 11. Nakagawa, Y., Daibo, I.: Guidebook for the General Health Questionnaire in Japan (Augmented Version). Nihon Bunka Kagakusha Co., Ltd. (2013) 12. Fukunishi, I.: Inspection Guide of Global Scale for Depression. Chiba Test Center Co. Ltd., Japan (2012) 13. Uchida, Y., Norasakkunkit, V.: The NEET and Hikikomori spectrum: assessing the risks and consequences of becoming culturally marginalized. Front. Psychol. 6, 1117 (2015) 14. Pedregosa, F., Varoquaux, G., et al.: Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011)

Product Packaging Evaluation Through the Eyes of Elderly People: Personas vs. Aging Suit vs. Virtual Reality Aging Simulation Christina Zavlanou(&) and Andreas Lanitis Visual Media Computing Lab, Department of Multimedia and Graphic Arts, Cyprus University of Technology, 31 Archbishop Kyprianos Street, 3603 Limassol, Cyprus [email protected], [email protected]

Abstract. In the content of Universal Design, various methods and techniques are used to allow designers to step into elderly users’ shoes to understand their speciﬁc characteristics and needs and design elderly-friendly products. Examples include the use of personas, aging suits and virtual reality aging simulation. However, there is no comparison that documents the advantages and disadvantages of each method. In the present study, the use of personas, aging suit and a virtual reality-based approach are compared in the evaluation stage of a product design process. According to the results, both the aging suit and the virtual reality aging simulation have important advantages compared to personas in terms of effectiveness. Virtual reality is advantageous in terms of helpfulness and embodiment of the designer, while aging suit precedes in terms of easiness to use. Keywords: Aging simulation Universal design

Virtual reality Personas Elderly

1 Introduction Designing products that meet elderly population’s needs is of imperative importance, but often this implies a difﬁcult process for product designers, as they must deal with a diverse audience in terms of physical and mental condition [1]. Through the years, several methods and techniques have emerged to support elderly-friendly product design [2–4]. Practical approaches have also been developed to allow designers to take the role of elderly users and identify the physical and other challenges associated with aging [5]. However, to the best of the knowledge, there is no comparison of different methodologies used for supporting the design of elderly-friendly products. In the context of this research, three methods of interacting with speciﬁc products through the eyes of elderly users, are tested in terms of their effectiveness and appropriateness for supporting the design of elderly-friendly products. More speciﬁcally, a method based on personas of elderly consumers, a method based on aging suits and a virtual reality aging simulation are evaluated, to deﬁne which of the three methods is optimum for supporting elderly-friendly product design.

© Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 567–572, 2019. https://doi.org/10.1007/978-3-030-02053-8_86

568

C. Zavlanou and A. Lanitis

2 Experimental and Computational Details 2.1

Product Packaging

A fundamental element in product design is packaging design [6]. Oftentimes, the packaging of widely used products is inappropriate and the labelling is illegible by elderly consumers, making the use of products difﬁcult and potentially hazardous [7]. As part of this research, the case of assessing the packaging of widely used products in terms of suitability by elderly consumers with age-related vision problems is examined. The products were selected to include basic, everyday products, namely a milk bottle, a toothpaste box and a pill box packaging. 2.2

Methods of Product Labelling Evaluation

Personas. Personas are representative proﬁles of potential users, often used in product design to help designers recognize the characteristics of different users to whom a product is addressed [8]. In the context of this research, three different personas were created: A persona of a 75-year-old woman suffering from macular degeneration, a persona of a 69year-old man suffering from cataract, and a persona of a 78-year-old man suffering from glaucoma. Personas’ descriptions were based on information extracted from Mayo Clinic website1. Through the personas description, participants were informed about the symptoms associated with each disease and the main problems faced by these personas in their everyday lives, such as difﬁculty in reading small print on packages. Finally, participants were provided with pictures showing how each persona sees compared to someone with normal vision. In the case of the evaluation of the packaging using the personas method, the actual packages of the three products were used. Aging Suit. There are also practical ways to help designers put themselves in users’ place to understand their characteristics and needs. In the case of the elderly users, a typical example is that of aging suits [9] that use special accessories to simulate physical disabilities related to aging, like visual deﬁciencies and reduced mobility. For the product packaging evaluation with the use of the aging suit, the simulation goggles provided by the GERT aging suit2 were used (Fig. 1, left), as product labelling is mainly associated with visual stimuli. The simulation glasses used were dedicated for the diseases of macular degeneration, cataract and glaucoma. For the purposes of the product packaging evaluation using the simulation glasses, the real packages of the products were used. Virtual Reality Based Aging Simulation. For the needs of the experiments, 3D models of the packages were created using the Maya 3D modeling software. Actual product labels were scanned and placed on the 3D models in the form of materials. The dimensions of the 3D models corresponded to the dimensions of the real products and the materials of the 3D models approached the actual ones.

1 2

https://www.mayoclinic.org/. http://www.age-simulation-suit.com/.

Product Packaging Evaluation Through the Eyes of Elderly People

569

Fig. 1. Evaluation of product packaging design using the simulation goggles (left) and the VR aging simulation (right).

To implement the virtual environment and simulate the problems associated with vision, the Unity3D game engine was used3. The VR environment was composed of a 3D table on which the three products were placed, to allow the participants to focus on the objects and not be distracted by the environment. Participants could also view the virtual representation of the hand controllers through which they could manipulate the virtual packages (Fig. 1, right). The simulation of the age-related vision problems in the virtual environment, was based on information extracted from the Mayo Clinic’s website, other related work [10] and visual feedback from the GERT’s simulation glasses. Users could alternate the image distortion ﬁlters that simulate the three diseases: macular degeneration, cataract and glaucoma. The simulation of these diseases in the virtual environment was implemented using post-processing effects in the form of blurring, blooming and color degrading, available in Unity. The visualization of the virtual environment simulating age-related visual deﬁciencies was achieved through the Oculus Rift Consumer Version 1. 2.3

Experimental Evaluation

An interview mode experiment was conducted with 9 participants (7 females and 2 males) aged between 22 and 28 years old. All participants were postgraduate students in Multimedia Design and had attended at least one course related to Universal Design and participated in projects related to elderly-friendly design. The prior experience of the participants in similar projects enabled them to provide important views and opinions in relation to the suitability of different methods for product evaluation. Participants were equally divided into three groups, based on the method used to evaluate the product packaging. Group A - Participants using the personas method Group B - Participants using the aging suit Group C - Participants using the VR aging simulator The participants were asked to reply to open-ended questions regarding products packaging design elements. After evaluating the product packaging, participants were 3

https://unity3d.com/.

570

C. Zavlanou and A. Lanitis

asked to evaluate the method they used to help them consider elderly users’ characteristics, through a questionnaire. More precisely, they were asked to comment on the positive and negative aspects of the method tested and report how strongly they had experienced the role of an elderly user. Moreover, they were asked to rate the degree of usefulness, efﬁciency, ease of use and utility of the method on a Likert scale from 1 to 5 (1: Not at all, 2: Slightly, 3: Moderately, 4: Very, 5: Extremely). Finally, participants were given more information about the remaining two methods and they were asked to state which method they would prefer for the task of packaging evaluation.

3 Results and Discussion Regarding the evaluation of the packaging design of the three products, within the participants of group A, comments showed large deviations. The comments were abstract, confused and in no way uniﬁed. In addition, there were opposing views between the participants who used the personas method and those who used the other two methods. A difﬁculty in understanding the needs of the elderly personas was also reported within group A, although the description of the diseases and the problems associated with these diseases, as well as a picture of the symptoms were provided. Comments like “I cannot separate the peripheral from central vision in my mind” were reported. Generally, participants of group A found it hard to visualize the packaging through the eyes of elderly users, although they were aware of personas’ problems. As a participant stated, “You guess and you can imagine, but you cannot really understand”. On the contrary, both the simulation glasses provided by the aging suit and the VR aging simulator made the visualization possible, thus, the comments from participants of groups B and C were more targeted, more detailed and the results more tangible. Another important aspect is that participants from groups B and C tended to compare to previous cases. Many participants, used expressions like “as before”, meaning that having a visualization of the problems, they could hold back previous images and compare them. In the case of the VR aging simulation, this was further enhanced thanks to the ease of effects’ switching. This aspect is rather useful to ﬁnd a balance between users with diversities in terms of physical condition and package design optimization. The results from the questionnaire are consistent with the results of the interviews, demonstrating that the personas method is the least preferred in the evaluation of product design through the perspective of elderly consumers. The VR aging simulation is the most preferred one, because as participants stated, “VR provides a protected, more interesting and pleasant environment”. Figure 2 presents a comparison between the three methods as they arise from the answers of the participants to the questionnaire. VR performed better that the aging suit in terms of helpfulness, but in terms of easiness, the aging suit received the highest score, as one user did not have sufﬁcient experience with VR technology and therefore had difﬁculty in handling the 3D packages in the VR environment. Generally, the preliminary investigation showed that VR-based aging simulation has the same potential as aging suits in terms of usefulness and effectiveness in the

Product Packaging Evaluation Through the Eyes of Elderly People

571

Fig. 2. Comparison of the three methods in terms of helpfulness, effectiveness, easiness and usefulness, based on the responses to the questionnaire.

evaluation stage of product design. It also presents a slight precedency in terms of helpfulness offered to the designers, and in the general feeling of the aging simulation. Considering the additional possibilities that VR aging simulation can provide, it can outperform the aging suit in the process of assessing the product packaging through the eyes of elderly consumers.

4 Conclusions Three methods used in product design to allow designers to step into the shoes of elderly users, were examined and compared within a case of packaging evaluation. The methods considered in the experiments were based on personas, aging suits and VRbased aging simulation. Experimental results indicate that the personas method is not effective and not preferred by the designers. Both the aging suit and the VR aging simulation were effective and useful in the process of evaluating package design. A precedence was recorded in the VR aging simulation, as it provided a protected and pleasant environment through which elderly consumer’s problems are better perceived. Considering the other possibilities offered by VR technology, such as the flexibility in switching between different scenarios and the ability to create different environments, VR aging simulation prevails in the task of product packaging evaluation. Acknowledgments. Authors acknowledge funding from the European Union’s Horizon 2020 Framework Programme through NOTRE project (H2020-TWINN-2015, Grant Agreement Number: 692058).

References 1. Christensen, H., Mackinnon, A.J., Korten, A.E., Jorm, A.F., Henderson, A.S., Jacomb, P., Rodgers, B.: An analysis of diversity in the cognitive performance of elderly community dwellers: individual differences in change scores as a function of age. Psychol. Aging 14(3), 365 (1999) 2. Farage, M.A., Miller, K.W., Ajayi, F., Hutchins, D.: Design principles to accommodate older adults. Glob. J. Health Sci. 4(2), 2 (2012)

572

C. Zavlanou and A. Lanitis

3. Vermeulen, J., Neyens, J.C., Spreeuwenberg, M.D., van Rossum, E., Sipers, W., Habets, H., Hewson, D.J., De Witte, L.P.: User-centered development and testing of a monitoring system that provides feedback regarding physical functioning to elderly people. Patient Prefer. Adherence 7, 843 (2013) 4. Demirbilek, O., Demirkan, H.: Universal product design involving elderly users: a participatory design model. Appl. Ergon. 35(4), 361–370 (2004) 5. Lavallière, M., D’Ambrosio, L., Gennis, A., Burstein, A., Godfrey, K.M., Waerstad, H., Puleo, R.M., Lauenroth, A., Coughlin, J.F.: Walking a mile in another’s shoes: the impact of wearing an age suit. Gerontol. Geriatr. Educ. 38(2), 171–187 (2017) 6. Azzi, A., Battini, D., Persona, A., Sgarbossa, F.: Packaging design: general frame-work and research agenda. Packag. Technol. Sci. 25(8), 435–456 (2012) 7. Vredenburgh, A.G., Zackowitz, I.B.: Drug labeling and its impact on patient safety. Work 33 (2), 169–174 (2009) 8. Pruitt, J., Adlin, T.: The Persona Lifecycle: Keeping People in Mind Throughout Product Design. Elsevier, Amsterdam (2010) 9. Breiner, K., Wüchner, T., Brunnlieb, M.: The disability-simulator: simulating the influences of disabilities on the usability of graphical user interfaces. In: International Conference on Ergonomics and Health Aspects of Work with Computers, pp. 109–118. Springer (2011) 10. Zavlanou, C., Lanitis, A.: An age simulated virtual environment for improving elderly wellbeing. In: XIV Mediterranean Conference on Medical and Biological Engineering and Computing 2016, pp. 891–896. Springer (2016)

Design for Innovative Development in Construction Industry: Proposal Based on Comparative Analysis Kazuo Hatakeyama(&) Enterprise Consulting Ofﬁce, Av. Miguel Navarro y Cañizares, 31, apto 701, Salvador, BA, Brazil [email protected]

Abstract. Houses for low incomes residential construction sector is one of the most prominent niches of the building industry in Brazil. However, the construction process continues handcrafted, which creates many errors and quality problems. To reach greater efﬁciency is necessary a large-scale production with repeatability, standardization, high productivity and an accurate management system. The construction process remains handcrafted, using concrete made on building site with masonry laid brick over brick, plaster, ﬁnish and hand paint and other operations. Such practice is prone to quality errors, which means in budget deviation, and the majority of construction enterprise transfer the cost to sales price. To lessen the cost of production, innovate design is required to avoid the penalization of customers to pay high price for low quality product. The inefﬁciency and the lack of innovative process reflection is visible by mismatching the budgeted cost and the completion cost always higher than estimated. Keywords: Design innovation Design stages Construction cost Low income

Construction industry

1 Introduction The houses for low incomes residential construction sector is one of the most prominent niches of the building industry in Brazil. However, the construction process continues handcrafted, which creates many errors and quality problems. To reach greater efﬁciency is necessary to have mass production with repeatability, standardization, high productivity and an accurate management system. The construction process remains handcrafted, using concrete made on building site, a masonry laid brick over brick, plaster, ﬁnish and hand paint and other operations. Such practice is prone to quality errors, which means in budget deviation, and the majority of construction enterprise transfer the cost to sales price. To lessen the cost of production, innovate design is required to avoid the penalization of customers to pay high price for low quality product [1]. Recent studies published on strategy for full social and economic potential showed that between 2003 and 2011 the construction sector increased its production cost in labor by 85% and materials by 70%. It identiﬁed that the low productivity is one of the © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 573–578, 2019. https://doi.org/10.1007/978-3-030-02053-8_87

574

K. Hatakeyama

main challenges of the sector to overcome. The study carried out by [2] on Taxation, Industrialization and Technology in Construction, in the short and long terms run projections points to favorable growth for the activity in the next 10 years.

2 Objective Search for an innovated design to rationalize the construction cost to fulﬁl the demand of population of low income.

3 Methodology Literature survey based on the models mentioned in the article [3] takes the ﬁrst step to migrate the black box model towards a transparent box in which can see and know operation way, therefore better manage it (Fig. 1).

Fig. 1. The process of black box to the transparent box [3]

It can consider the stages and macro stages as the ﬁrst level of detail management process as shown in Fig. 2 regarding to stage gates and Fig. 3 on product development management.

Fig. 2. The stage gate system [4]

Design for Innovative Development in Construction Industry

575

Fig. 3. Product development process [5]

4 Theoretical Background 4.1

Stage Gates for Innovation - Process for New Products Development

The stage gate has been developed and ﬁrst suggested by [4] in his book “Winning at New Products”, published in 1986. The stage-gate development process divides into stages separated by gates. At each gate, the continuity of the development process led by a manager or by a committee. The decision based on information available at the time, including business cases, risk analysis, availability of necessary resources such as money, people with skills, among others. As [4] model uses tools when dealing with the development of new products. Gates or decision points are located at points of most beneﬁcial development for decisionmaking. The production areas, located between gates are: Idea generation; 2. Establishment of viability; 3. Test; 4. Validation; 5. Product launch. The schematic diagram of stage gate approach as shown in Fig. 2. Within each stage, a series of multifunctional activities is performed, and this crossfunctional team should coordinated by the project manager. In Fig. 2 one can see the 5 stages and 5 decision-making points to the commercial launch of the product. Stage 1 - Preliminary investigation - is considered to be a preliminary investigation of the project scope including a non-exhaustive work.

576

K. Hatakeyama

Stage 2 - Detailed investigation - comprises a detailed investigation, both technical and commercial project scope to build the business case of the project. Stage 3 - Development - is the development of the product, including its production and operation processes. Stage 4 - Validation and testing - are made technical tests for the resilience and quality reliability. Stage 5- Full production and market launch – After the several testing procedures and the validation the full production process carried out and market launch planned. 4.2

Reference Models for the Management of Product Development Process

The product development management (PDP) has as main objectives, control and quality assurance in the process of product development, linking all professionals involved from the design phase of a product to market and production processes. The deﬁnition proposed by [5] establishes PDP as a set of activities in a logical sequence in order to produce goods or services that have value to a speciﬁc group of customers. Reference [5] states that the “reference model” is essential to the product development process because: • Determines the ability of companies to control the PDP and interact with the market and with sources of technological innovation. • Includes strategic and operational management, the development cycles of problem solving, improving and learning, considering the entire product life cycle. • Structure the operational steps and activities of project development. • The formalization of the PDP management model enables everyone involved to have a common view of the process. Reference model is suitable for both in training as well as to updating professionals as the ideal model for the implementation of improvements in the process of developing products. Essentially the PDP comprises several stages, namely: idea generation, previous research, even development, pilot series and launch of the product on the market. There is no established rule for the division of the stages, also the boundaries between one to another are not clearly identiﬁed and known, but is the purpose of each stage to perform the tasks. The PDP split into three macro-stages: pre-development, development and postdevelopment. These macro-stages subdivided into stages that detail and specify activities within the process. These stages require resources and time to execute and transform inputs into outputs. The Fig. 3 briefly presents the reference model [5]. The pre-development macroprocess includes the product strategic planning stage. Such stage deﬁnes which products to develop, which are not and which markets will focused. It is still in the planning stage that the project scope is in deﬁnition, taking into account the economic viability and the organization’s ability to run the project. The development of macro process consists ﬁve stages: informational design, conceptual design, detailed design, preparation for production and product launch.

Design for Innovative Development in Construction Industry

577

During the stage of informational design is necessary information about the project, according [6] this stage turns the output of the previous stage design speciﬁcation, in addition to detailing product requirements. In the conceptual design is when the descriptions started as product modeling itself, with the functions that the product must present, regardless of how it will be structured such functions. Then they proposed possible solutions for product functions and made the schematic arrangement of the product. In the detailed design stage the necessary tests establishment including support materials, product’s packaging design is also stated. This stage has three cycles: detailing, acquisition, and optimization cycles. Concurrently with this stage occurs the production preparation stage. There are the development of product manuals and instructions for technical assistance, information to vendors and staff training, as well as the manufacturing devices. It is at this stage still occurs the production of a pilot lot or pre-series. At last, the development of macro process comes the product launch stage, when in addition to document best practices there are marketing activities, sales, product distribution logistics arrangement, the customer service and after service and technical assistance. So the product can released into the market. The post-development macro process contains two stages, namely: Stage in which follows up the product and process, an assessments of customer satisfaction and technical product performance, the audit processes, and enrollment to the lessons learned [7].

5 Survey The survey centered in the product and process stage (stage 5) based on [4], and post development gate [5] as the cost incurred in this stage presents several costly parameters that worth to improve and reduce. Most of approaches for reducing costs in production process are for mass production in manufacturing rather than for discrete product such as housing with limited quantity with distinct design. For the sector of housing construction business, it has noticed that the converging approach from the input to output framework, currently applied to the manufacturing goods, requires detailed study in each stage in the production process. Most of process bottleneck, in the stage of construction process, demands the stoppage to review the design to assure the correctness of the streamline route since the initial stage to the marketable ﬁnished product. In the ﬁeld survey from the three sizes of categories of construction ﬁrms: small, medium, and large, has noticed that most critical stage in terms of cost control falls in the production process. To discover the hidden cost appeared to be the most complex matter. The causes for the increase in the cost regarding to the original budget are complex, due fluctuation of the cost of materials, uneven cost of manpower due to the necessity of rework to comply with the quality standard demanded by the customers, among others.

578

K. Hatakeyama

6 Results The division of this complex process is an important step in establishing control points to ensure design efﬁciency. Innovative approach requires the close check on the initial stage beginning with ideas and market information. The conceptual design stage complementing the guidelines with a detailed deﬁnition of technical characteristics. The development and homologation of detailed product by the multifunctional team plus in speciﬁc areas experts. The production process still needs further improvements to accomplish the minimum cost. The launch to place the product on the market, together with the result of the previous stage, aiming at the acceptance of potential customers. Finally the closing for veriﬁcation of compliance with the targets of the project.

7 Conclusion The proposed model developed to innovate products for housing construction sector bringing an approach that blends typical theories for the manufacturing industry relying on generic models to models of speciﬁc design. It brings certain unknown, incipient and fragile side of the proposal. Nevertheless, should consider as a preliminary proposal, which lacks yet the depth and development. The hope that this study will serve as a basis for future studies and the sector of construction can create speciﬁc templates for their reality.

References 1. Nascimento, G.L., Hatakeyama, K.: Reference model for innovative product development in construction - a stages proposal based on comparative analysis. In: Proceedings of the Portland International Conference on Management of Engineering Technology, Honolulu, Hawaii (2016) 2. Fundacao Getulio Vargas: Taxing, Industrialization and Technology Innovation in the Civil Construction (2014). E-book www.abcici.org.br 3. Romano, F.V.: Reference model to manage the process in integrated design of buildings, p. 326. Doctor’s Thesis – PPGEP - UFSC, Florianopolis (2003) 4. Cooper, R.G.: Stage-gate systems: a new tool for managing new products. Bus. Horiz. 33, 44–54 (1990) 5. Rozenfeld, H., et al.: Management of Development of Products – One Reference for Process Improvement. Saraiva, Sao Paulo (2006) 6. Ribeiro, M.S.: The industrialization as requirement for rationalization of construction. Master’s Dissertation – Universidade Federal do Rio de Janeiro, PROARQ/FAU (2002) 7. PMI – Project Management Institute: A Guide to the Project Management Body of Knowledge (PMBOK), p. 216. PMI, Pennsylvania (2014)

An Evaluation Method for Intellectual Concentration Based on Concentration Depth Kimi Ueda1(&), Shota Shimonaka1, Hiroshi Shimoda1, Hirotake Ishii1, and Fumiaki Obayashi2 1

Graduate School of Energy Science, Kyoto University, Kyoto, Japan {ueda,shimonaka,shimoda, hirotake}@ei.energy.kyoto-u.ac.jp 2 Panasonic Corporation, Osaka, Japan [email protected]

Abstract. There have been lots of studies on the relationship between ofﬁce room environment and workers’ intellectual productivity such as room ventilation, temperature, humidity and sound noise, however, there are no established method of measuring the intellectual productivity. The authors have been developed CTR (Concentration Time Ratio) which expresses the time ratio of their concentration on the target intellectual work. In this study, a new index CDI (Concentration Depth Index) was developed to realize an analysis from a new viewpoint of concentration depth which is thought to have a relationship with the quality of concentration. Keywords: Intellectual concentration Working state model

Evaluation experiment

1 Introduction After starting information society in the beginning of the 21st century, intellectual work has had more value than other labors such as physical works, and they have been interested in its work efﬁciency and productivity. Many researchers and developers have been working on to improve the productivity of intellectual work in order to gain lots of beneﬁts [1]. Although the intellectual productivity is recognized as an important factor of ofﬁce performance and its measurement is required to discuss how to keep or improve it, there is no established measurement method. The generic measurement methods are roughly divided into two methods which are questionnaire surveys [2, 3] and cognitive tasks [4]. As one of the methods utilizing cognitive tasks, the authors have developed an evaluation index, CTR (Concentration Time Ratio) [5]. By using the CTR, they have showed that physical ofﬁce environment such as lighting and air-conditioning affects the workers’ intellectual concentration [6]. The CTR is a measurement index based on the concentration state model as shown in Fig. 1. This model assumes that cognitive states are divided into three states from the viewpoint of concentration, working state, short-term pause state and long-term rest state. According to Baddeley [7], however, the amount of their cognitive resources can be changed. This means that © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 579–584, 2019. https://doi.org/10.1007/978-3-030-02053-8_88

580

K. Ueda et al.

the quality of concentration may change depending on how much cognitive resources are assigned to the target cognitive work. Considering this idea, the authors have also developed another intellectual concentration index, CDI (Concentration Depth Index), and showed two example analyses [8].

Fig. 1. A cognitive state model and CTR calculation methods [9].

The purpose of this study is to show the availability of CDI comparing with CTR using another experimental data. As mentioned above, CTR expresses the amount of total concentration time while CDI shows the quality which can be expressed concentration depth in other words. Concretely, the results of the past experiment where cognitive tasks were conducted under different room conditions were analyzed by both CTR and CDI.

2 Cognitive State Model and Measurement Index CDI (Concentration Depth Index) The cognitive state model of CTR was introduced in Sect. 2. In this chapter, advanced cognitive state model which is the basic idea of CDI is briefly reviewed [8]. Figure 2 shows the cognitive state model of CDI. In this model, it is assumed that there are several concentration states which have different concentration depths. They have different amount of cognitive resources assigned to the target cognitive work. In each depth of concentration, the working state and the short-term pause state forms binomial Markov model, so that the distribution of answering time data of cognitive task forms a unique log-normal distribution. Figure 3 shows an example of a distribution histogram of answering time data. When the amount of assigned cognitive resources is different, the shape of the log-normal distribution is different. The more cognitive resource is assigned, the deeper the concentration is and the shorter the answering time is. As shown in Fig. 3, two log-normal distributions can be recognized. These two distributions express the ﬁrst deepest concentration and the second deepest concentration. In

An Evaluation Method for Intellectual Concentration

581

most of the answering time histogram, these two distribution can be seen. The distributions of the third and shallower concentrations cannot be seen because it is difﬁcult to distinguish them from non-concentration state. Based on the above consideration, CDI has been proposed as the time ratio of the ﬁrst deepest concentration time among the ﬁrst and second concentration time. For the detail deﬁnition and calculation method of CDI, refer to [8].

Fig. 2. An advanced cognitive state model considering depth of concentration.

3 Comparison Between CTR and CDI Based on Past Experimental Data In this chapter, the comparison between CTR and CDI is described based on the past experimental data where thermal condition of experimental room was controlled and cognitive task was given under different conditions. This experiment was conducted to evaluate the influence of intellectual concentration by controlling room temperatures of both a working room and a resting room. In the proposed condition, the temperature of the working room is cooler while the resting room is warmer. It is expected that the cooler thermal condition stimulates workers’ sympathetic nervous system and it leads them to concentrate on their work while the warmer thermal condition stimulates their parasympathetic nerve system and they quickly recover from fatigue. In total, it is expected that the proposed thermal condition can improve their intellectual concentration rather than non-controlled thermal environment. 3.1

Experimental Method

Figure 4 shows the experimental schedule. It was conducted in three days where the ﬁrst day was for a practice, and on the second and the third day two thermal conditions, the proposed condition (multi-controlled condition) and non-controlled condition (neutral condition) were set. Thirty-ﬁve graduate school students and undergraduate students participated in the experiment (all males, age of 19 to 26, average 21.8). They

582

K. Ueda et al.

Fig. 3. An example histogram and an analysis example.

were divided into several groups where 6 to 8 participants joined. In order to cancel the order effect of thermal conditions, the condition order given on the second day and the third day was counterbalanced in each group. In each day of the experiment, cognitive tasks were given ﬁve time as SET 1 to 5 and the result of SET 5 was not used because it may be affected by terminal effect. They conducted the task in the working room and took rest between the tasks in the resting room. The room temperature was controlled as mentioned above under the multi-controlled condition while they were set to the same value under the neutral condition. In each task set, answering time data of Comparison Task [7] were recorded in order to calculate CTR and CDI in the later analysis.

Fig. 4. Experimental protocol of environmental evaluation experiment.

An Evaluation Method for Intellectual Concentration

3.2

583

Results and Discussion

The averages and S.D.s of CTR and CDI are shown in Table 1. The data of 26 participants were used while others were rejected because they didn’t follow the experimental instructions or their physical conditions obviously changed. Table 1. Average CTR and CDI in neutral and multi-controlled condition. Index Neutral condition Multi-controlled condition Paired t-test CTR 56.4 ± 10.6% 57.5 ± 10.4% p = 0.140 CDI 56.3 ± 15.7% 59.4 ± 12.6% p = 0.033

The results showed that CDI was marginally signiﬁcantly larger in the multicontrolled condition by 3.1% while CTR had no signiﬁcant difference between neutral and multi-controlled conditions. This shows that CDI offers another viewpoint of analysis for the difference of intellectual concentration under the different thermal conditions. In other words, the collected answering time data can be analyzed in more detail from the viewpoint of concentration depth. As the results, it was found that the multi-controlled condition has the possibility to improve the depth of concentration even if it can’t improve CTR which shows the amount of concentration time. The results’ difference between CDI and CTR showed that the multi-controlled environment had stronger effect on the composition of intellectual concentration, which can be measured by CDI the ratio of the ﬁrst and second concentration state, expressing the quality of concentration than the amount of concentration.

4 Conclusions In order to show the availability of CDI, the results of the past experiment were analyzed by both CTR and CDI. Two thermal conditions were set in the experiment and cognitive tasks were conducted under the conditions. As the result of CTR and CDI, CTR under the multi-controlled condition showed no signiﬁcant difference from that under the neutral condition while CDI showed a signiﬁcant difference. This means that CDI has a possibility to offer another viewpoint of the analysis for the intellectual concentration affected by the physical room conditions. Acknowledgments. This work was supported by JSPS KAKENHI Grant Number JP17H01777.

References 1. Ministry of Land, Infrastructure, Transport and Tourism: Research Report by Study Group of Intellectual Productivity (2009). (In Japanese) 2. Sugiura, T., Hashimoto, T., Terano, M., Nakamura, M., Kawase, T., Kondo, Y.: Evaluation method of work place productivity, the ﬁrst report – organizing productivity evaluation methods and suggestion of the standard methods. Academic lecture papers of The Society of Heating, Air-Conditioning and Sanitary Engineers of Japan 123, pp. 11–22 (2007). (In Japanese)

584

K. Ueda et al.

3. Nishikawa, M., Nishihara, N., Tanabe, S.: The effect of moderately hot environment on performance and fatigue evaluated by subjective experiment of long time exposure. Arch. Inst. Jpn. J. Environ. Eng. 74(638), 525–530 (2009) 4. Wargocki, P., Wyon, D.P., Fanger, P.O.: Productivity is affected by the air quality in ofﬁces. Healthy Build. 2000, 635–640 (2000) 5. Uchiyama, K., Ooishi, K., Miyagi, K., Ishii, H., Shimoda, H.: Process of evaluation index of intellectual productivity based on work concentration. In: 5th International Conference on Software Technology and Engineering (2013) 6. Uchiyama, K., Miyagi, K., Ishii, H., Shimoda, H., Obayashi, F., Iwakawa, M.: Development of calculation method of intellectual concentration for evaluate intellectual productivities. J. Hum. Interface Soc. 16(1), 29–40 (2014). (In Japanese) 7. Baddeley, A.: Exploring the central executive. Q. J. Exp. Psychol. A, Hum. Exp. Psychol. 49, 5–28 (1996) 8. Ueda, K., Shimonaka, S., Shimoda, H., Ishii, H., Obayashi, F.: Quantitative evaluation of intellectual productivity considering depth of concentration. In: 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 758–763 (2017) 9. Ueda, K., Shimoda, H., Ishii, H., Obayashi, F., Taniguchi, K.: Development of a new cognitive task to measure intellectual concentration affected by room environment. In: The Fifth International Conference on Human-Environment System ICHES 2016, Nagoya, Paper No. 20144 (2016)

Application of the Principle of Conformal Symmetry in the Structure of Human Internal Organs Galina Spirina(&) Human Anatomy Department, Ural State Medical University, Yekaterinburg, Russian Federation [email protected]

Abstract. One of the important theoretical problems in living nature is shaping. The study of this problem has practical importance in connection with the demands of modern medicine. The study of laws and algorithms of formation is one of the important directions of developmental biology. Measurements of proportions of body parts traditionally carried out on the basis of afﬁne (simple) relations between the indices of two anatomical measurements. However, three – membered structures predominate in the human body. The criterion of conformal symmetry between blocks of three successive segments of lines is wurf. By analogy with the three-membered structures of the human body, author on large number of observations parameters of the heart and kidneys of people of all ages is in three segments related in conformal symmetry. Wurf value of the right and left kidneys, of the human heart are clustered around the numbers 1,3, closer in meaning to “golden” wurf. In each age group, the ratio of the three parameters heart, its ventricles is relatively constant, independent of age. Conformity reflects some strict and constant space-time parameters of macroscopic growth. Keywords: Human

Internal organs Conformal symmetry Growth

1 Introduction One of the important theoretical problems in living nature is shaping. There are a large number of publications about the impact on it of functions and environmental conditions. The study of this problem has practical importance in connection with the demands of modern medicine, agriculture, industrial production. The study of laws and algorithms of formations is one of the important directions of developmental biology [3]. Symmetry in the forms of biological bodies has long been of interest to researchers [1–3]. The idea of symmetry is the sum of the possibilities of combining ﬁgures and the presence of a certain operation that produces this combination [4, 5]. Measurement of proportions of body parts was traditionally carried out on the basis of afﬁne (simple) relations between the indices of two anatomical measurements. However, the human body can not completely reduced to proportions of two-membered structures. Three – membered structures predominate in the human body [1, 2]. This principle is manifested in the kinematic scheme of the body – in the structure of the ﬁngers, limbs, the whole body [6]. To calculate the conformal symmetries, three segments of the anatomical parameters are © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 585–588, 2019. https://doi.org/10.1007/978-3-030-02053-8_89

586

G. Spirina

used. The criterion of conformal symmetry between blocks of three successive segments of lines is wurf. The term “wurf”, replacing the term “double or complex, attitude” was introduced by Staudt, when creating a projective calculus without metric bases and denotes in translation from German “throw” [6]. The musculoskeletal system can be represented as a set of three-membered kinematic blocks. The value of the wurfs of all blocks in the distributed state is grouped around the value 1.3. The phenomenon of conformal growth is revealed: from the moment of the birth of the individual, the growth of the body occurs with the preservation of the size of the wurf of each three-membered kinematic block. If the value of the wurf of the measured structures is 1,31 ± 5%, it is the “golden” wurf. Petukhov [2] showed that the rule of the “golden” wurf obeys all three-termed kinematic chains of the human body. Petukhov̛ s research was continued by Nigmatullin, Gafarov, Salikhov [6], who believed that the person̛ s ̉face is conformally symmetrical, his proportions are harmonious. According to the authors, subordination to the laws of conformal symmetry is manifested in all age periods. This conclusion, formulated on the example of the face, conﬁrms the conclusion other authors about the spiral twist of the human fetus around the main axis, which causes the morphological mirror asymmetry of the human body [4, 7].

2 Aim of Work Morphometric analysis of the parameters of the kidneys and heart, determining their interrelation and interdependence.

3 Materials and Methods The material of the work was 20 preparations of the right and left kidneys of adults of both sexes, 501 a preparation of the heart of human fetuses, children, adults of both sexes. By analogy with the three- membered structures of human body, the parameters of the kidneys and heart were represented in the form of three –membered blocks. On preparations of the right and left kidneys of adult people measurements were made of their width, length, and thickness. After the measurements, the wurf of each kidney was determined according to the formula: W¼

ðC AÞ ðD BÞ ðC BÞ ðD AÞ

A

B

C

D;

ð1Þ

A-B – length of the kidney B-C – width of the kidney C-D – thickness of the kidney. We determined the heart parameters, the length and width of the ventricular input and output. When studying the parameters of the kidneys and heart proceeded from the assumption that the riddles of the formation of biological bodies lie in the conformal of space.

Application of the Principle of Conformal Symmetry

587

4 Results According to the data obtained, the parameters of the kidneys of adults can be represented in the form of three segments connected by conformal symmetry. The value of the wurf of the right and left kidneys is close in value and grouped around the value 1.3. To successfully solve the problems of preventing anomalies, congenital disease and reducing perinatal mortality, depending study of the process of heart formation is necessary. The linear dimensions of the heart, its ventricles, can be represented in the form of three segments connected by conformal symmetry. The change is one of the parameters is in a certain way coordinated with the change of other two. In each age group, the ratio of three indicated parameters of the heart, its ventricles (the length of the inflow and outflow section, the total width of each ventricle) is relatively constant, independent of age and heart index [5]. The average value of the wurf of the ventricles, of the heart itself ranges from 1.24 to 1.30, approaching 1.309.

5 Conclusion The study of literature sources convinces us that the morphogenesis processes and morphogenesis in the embryo organism are often controlled by essentially identical regulatory mechanisms, implementation schemes that are of universal signiﬁcance and applied to derivatives of various embryonic tabs. The revealed linear relationship between the three parameters of the kidneys, heart and its ventricles can be considered as a biomechanical constant that regulates the proportionality of the indicated parameters. Parameters of the internal organs of a person (kidneys and heart, its chambers) represent a set of trinomial kinematic blocks in the straightened state. The value of the wurfs of all blocks is kept constant regardless of age, which is clearly seen in the analysis of heart products. The rule of the “golden” wurf is not only the three-termed kinematic chains of the musculoskeletal system, but also the parameters of the internal organs (kidneys, heart and its chambers). In the opinion of Petukhov [2] in the kinematic blocks of the human body, two kinds of proportions – wurf and afﬁnity – manifest themselves differently in the process of ontogenesis. Wurf proportions are rigidly kept constant; demonstrate a connection with Fibonacci series. Afﬁne proportions do not show a stable connection with the Fibonacci series and change in ﬁlogeny. These changes are dependent on lifestyle, function and other factors. Features of the kinematic scheme of man are determined by the parallel existence in it of unaltered wurf proportions, the magnitude of which is determined by the phyllotaxis laws of morphogenesis and aphillotaxis afﬁne proportions through which the organism adapts to a special way of life. These features can be used the model the internal organs of a person.

References 1. Petukhov, S.V.: Biomechanics, Bionics and Symmetry. Science, p. 240 (1981) 2. Petukhov, S.V.: Geometry of living nature and algorithms of self-organization, vol. 6, p. 46 (1988)

588

G. Spirina

3. Petukhov, S.V.: Symmetries in biology. The appendix to the book: Shubnikov A.B., Coptik V. A. “Symmetry in Science and Art”, 3rd edn., pp. 489–546 (2004) 4. Spirina, G.A., Doronin, A.I.: Structural organization of the heart of fetuses in the normal and pathological course of pregnancy. Complex-adaptive mechanisms of internal organs and brain in norm, pathology and experiment, Tyumen, pp. 148–150 (1996) 5. Spirina, G.A.: Realization of the principle of conformal symmetry in the structure of the heart. Advances in Human Factors in Simulation and Modeling. In: 2017 Proceeding of the AHFE International Conference of Human Factors in Simulation and Modeling, Los Angeles, California, USA, vol. 26, pp. 175–183. Springer (2017) 6. Nigmatullin, R.T., Gafarov, V.G., Salikhov A.Y.: Soft core of a person ̍ s face. Aspects of surgical and functional anatomy. Ufa, 99–119 (2003) 7. Doronin, A.I.: Morphometric characteristics of the human heart in the prenatal period of development. Abstract dissertation Ph. D. (Medicine), Ufa, p. 23 (2001)

Design of a Framework to Promote Physical Activity for the Elderly Alexandre Calado(&), Pedro Leite, Filomena Soares, Paulo Novais, and Pedro Arezes Algoritmi Centre, University of Minho, Guimarães, Portugal [email protected], [email protected], [email protected], [email protected], [email protected]

Abstract. Physical inactivity is estimated to be one of the leading risk factors for global mortality and it is associated with several illnesses, such as type 2 diabetes, cardiovascular diseases and various types of cancers. To tackle this issue and promote physical activity amongst the elderly, a system that computes automatically, in real-time, the score of a Boccia game was developed. The objective of this paper is to infer the best design possible for the User Interface (UI) that displays this information. To achieve this, two surveys were conducted involving 45 participants. In the ﬁrst survey, the participants were asked what features they would like to see in the UI. Based on these remarks, the authors designed an UI, along with several variations. The preferences between these variations were afterwards evaluated in the second survey. Thus, the ﬁnal design of the UI was validated before being shown to the elders. Keywords: Human factors

Boccia Interface design Physical inactivity

1 Introduction The overall level of physical activity declines with age [1], consequently making older adults one of the most sedentary age groups. According to Matthews et al. [2], adults aged more than 60 years have been identiﬁed to be the most physically inactive group in the United States. Physical inactivity is the fourth leading risk factor for global mortality [3] and it is associated with an increased risk of various diseases and disorders, such as type 2 diabetes, coronary heart disease, hypertension, obesity and different types of cancer [4]. Taking into account the aforementioned risks, it is paramount to develop a strategy for promoting physical activity in the elderly, which can have a panoply of beneﬁts at an advanced age, such as prevention of functional loss [5], blood pressure control [6] and reducing depression symptoms [7] and mortality rate [8]. The approach taken by the authors of this paper was the development of a system based on a computer vision algorithm that automatically computes the real-time score of a Boccia match and displays it in a User Interface (UI) [9, 10]. The purpose of this system is to make the overall Boccia game experience more enjoyable for the player and to encourage him/her to play more often, thus promoting physical activity practice. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 589–594, 2019. https://doi.org/10.1007/978-3-030-02053-8_90

590

A. Calado et al.

The present paper focuses on the design of the UI that displays the score of the game. Before showing the UI to the elders, two surveys were conducted to infer the general public’s opinions and preferences and validate the overall design. These surveys included questions regarding the usefulness of the proposed system, understanding of the UI’s content by the user and preferences between different pairs of UIs, varying in text font size, colour and organization of graphic elements.

2 Background Design, in many instances, holds the key for an efﬁcient interaction between man and technology. Bearing this in mind, it is crucial to certify that the design of any UI is satisﬁable for the user and that its content is intuitive and comprehensible. Due to the current increase of the elderly population, emphasis should be given to the design directed for this age group, taking into account impairments that are characteristic of late age. However, considering the system proposed by this paper, special attention should be given to impairments related with vision. There are several suggestions and guidelines regarding designing for older adults. Most of the suggestions found in literature related to vision comprise text characteristics, such as font size, type and contrast with the background [11]. However other features, such as the colour of the graphic elements, should also be taken into account. For instance, Lee and Park [12] studied how the colour environment of a display can influence its usability on older adults. The survey conducted concluded that larger font size, brightness contrast between the text and the background, along with clear, dynamic, high-chromatic colour combinations were preferred by the seniors. Regarding digital games, there as also been a growing acceptance by the elderly [13]. For example, Gerling et al. [14] developed a game that allowed older adults to go on virtual walks through a park while competing in small mini-games, using Nintendo’s Wii Remote and the Balance Board for control. This acceptance further encourages the development of innovative solutions, such as the system proposed by the authors of this paper. 2.1

The Boccia Game

Boccia is a simple precision ball game which was selected by the authors due to its easy adaptability to individuals with different types of motor skills impairments. This game is often played in the nursing home where the system proposed by the authors is intended to be tested. However, in this case, the rules were altered in order to be understandable to some of the residents that have cognitive limitations. According to the caretakers from the nursing home, the players are divided into two teams, the red team and the blue team. Both teams are composed by three players, who are given two balls each, of the respective team colour. The game itself is divided into three segments called “ends”. In each end, the players take turns at throwing their balls in order to hit a white ball, called jack, which is placed in the centre of the court before the game is started. Every time a player hits the jack, he/she earns one point for his/her team. When all three ends ﬁnish, the points earned by each team in each end are

Design of a Framework to Promote Physical Activity for the Elderly

591

summed and the team with the higher total of points is declared the winner. In the eventuality of both teams ﬁnishing with an equal amount of points, the team that placed the ball closer to the jack during the ﬁnal end is the winner.

3 Methods The study to validate the design of the UI was based on two surveys. The participants included caretakers from a nursing home and day care centre for the elderly in S. Torcato, Guimarães, in Portugal, along with students from the University of Minho. Both surveys were ﬁlled by the same participants. Of the 45 participants, 75% were males and 25% female and the average age was 26, ranging from 20 to 58 years. All participants were informed about the objectives of the surveys and their identity was kept anonymous. Moreover, none of the participants was familiar to the UI, thus all answers were unbiased. The ﬁrst survey questionnaire consisted in a series of statements regarding two main areas: general motivation towards implementing a system that displays the score of a Boccia game, and which are considered the most important visual components to be included in the user interface. For each of these statements, participants indicated how much they agreed by using a Likert scale scale that ranged from “very irrelevant” (score of 1) to “extremely relevant” (score of 5). On the other hand, the second survey was conducted with the objective of assessing the system’s usability and user’s satisfaction towards it, based on a UI designed considering the answers to the ﬁrst survey. Furthermore, the participants were asked to choose between three pairs of different versions of UI, considering different text font sizes, darker or brighter colours and graphic elements organization (elements displayed horizontally or vertically).

4 Results Regarding data analysis, an alpha level of 0.05 was set for the computer analysis (SPSS, version 24) of statistical signiﬁcance of the results. Besides, the data was found to be approximately normally distributed. Regarding the results from the ﬁrst survey, in Table 1, means, standard deviations and t-test statistics on the attitude towards preferred visual elements are represented for both caretakers and students’ groups. Agreement was very high for general motivation towards implementing a system that displays the score of a Boccia game (agreement of 3.93) and that it would motivate the elderly to participate more often (agreement of 3.67). Furthermore, the participants considered the most important elements to be the ones displaying information about the Boccia Game score followed closely by visual aid when throwing the ball. On the other hand, the least important elements were considered to be the “Date & Time”, “Game Duration” and “End number being played”. Besides, no signiﬁcant differences were found between the caretakers and students’ answers. Based on the aforementioned remarks, an UI was designed with focus on information about the scores. This UI is depicted on Fig. 1.

592

A. Calado et al.

Table 1. Means, standard deviation and t-test statistics on the attitude preferred visual elements Items

Current score Individual score Total team score Team score from each end Rules description Number of the end being played Indication of the current player Game duration Court ﬁeld view Recommendation of the amount of force to be used in the throwing and its orientation Date & time

Caregivers

Students

M

SD

M

SD

4.00 4.00 4,33 4.00 3.00 2.67 4.00 2.33 3.67 3.67

0.58 0.58 0.67 0.58 0.58 0.33 0.58 0.33 0.33 0.88

3.93 3.4 4.00 3.43 2.67 3.62 3.76 3.45 3.45 3.31

0.13 0.17 0.13 1.33 0.20 0.17 0.15 0.16 0.19 0.188

Independent Sample t-test statistic Sig. 0.89 0.36 0.51 0.28 0.67 0.14 0.67 0.08 0.61 0.63

2.67

0.33

2.60

0.18

0.91

Fig. 1. UI designed based on the information extracted from the ﬁrst survey.

Information regarding the red and blue team are depicted on the right and left side of the UI, respectively. At the top, from each side, it is possible to observe the current score of the end being played. In the tables below, one can see the points obtained at the ﬁnal of each of the three ends. Finally, it is represented, at the bottom, the total points of each of the team. The results from the second survey are represented in Table 2, where means, standard deviations and t-test statistics on the user’s satisfaction towards the developed UI are represented for both caretakers and students’ groups. Regarding the participants choices between three pairs of different versions of the UI depicted in Fig. 1, it was inferred that the majority preferred an UI with larger text font size (77.1%), darker colours (67,4%) and the organization of graphic elements displayed vertically as illustrated in Fig. 1 (89,1%). Once again, no signiﬁcant differences were found between the caretakers and students’ answers.

Design of a Framework to Promote Physical Activity for the Elderly

593

Table 2. Means, standard deviation and t-test statistics regarding preferred visual elements Items

It is preferable to use a different way to display the information The information presented in the graphical interface is useful The presented tool is satisfactory The information is displayed in a clear and understandable way I did understand the information related with the team score I did understand the information related with the team during the three ends I did understand the information related with the total team score

Caregivers

Students

M

SD

M

SD

3.00

1.00

2.76

0.14

Independent Sample t-test statistic Sig. 0.69

4.00

0.33

4.02

0.11

0.95

3.00 3.00

1.00 0.58

3.81 3.81

0.12 0.16

0.12 0.20

4.00

0.33

4.12

0.13

0.81

4.00

0.33

4.12

0.12

0.80

4.00

0.33

4.10

0.12

0.82

5 Final Remarks This paper focused on the design of a UI to display the score of a Boccia match in real time. A study based on two surveys was performed in order to evaluate what users wished to see in such a UI and to evaluate the user satisfaction towards an UI design based on the latter. The output from both surveys helped designing and validating an UI to be tested during a Boccia game played by the elderly. Overall, the participants preferred an UI focused on just displaying information about the Boccia Game, with large text font size, darker colours and graphical elements displayed vertically as depicted in Fig. 1. According to data from the second survey, the participants also showed to be relatively satisﬁed with the preliminary version (Fig. 1) and properly understood its content. The majority of the participants were of relatively young age, which may not be representative of the opinion from different age groups. However, it was important to have a way of validating the UI before showing it to the elders. Regarding future work, this system will be integrated within the iBoccia framework [15], which features the use of various sensors for monitoring elderly physical activity and display the extracted data to the caregiver. It is planned to test the whole system during a Boccia match at the nursing home and evaluate the level of satisfaction from elders. Regarding further development of the UI, based on the results from the ﬁrst survey, it is also planned to display visual aid when throwing the ball, regarding the orientation and force that should be applied by the player. An indication of the current player’s turn will also be implemented by displaying the player’s photo on the screen. These additions are predicted to further motivate the elderly to engage in the game and, therefore, practice physical activity more regularly.

594

A. Calado et al.

Acknowledgments. This article is a result of the project Deus ex Machina: NORTE – 01 – 0145 – FEDER - 000026, supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF).

References 1. Troiano, R.P., Berrigan, D., Dodd, K.W., Mâsse, L.C., Tilert, T., McDowell, M.: Physical activity in the United States measured by accelerometer. Med. Sci. Sports Exerc. 40, 181– 188 (2008) 2. Matthews, C.E., Chen, K.Y., Freedson, P.S., Buchowski, M.S., Beech, B.M., Pate, R.R., et al.: Amount of time spent in sedentary behaviors in the United States, 2003–2004. Am. J. Epidemiol. 167, 875–881 (2008) 3. WHO: W.H.O. Global recommendations on physical activity for health, World Health Organization, Geneva (2010) 4. WHO: Factsheets on Health-Enhancing Physical Activity in the 28 European Union Member States of The WHO European Region, pp. 1–27 (2015) 5. Stessman, J., Hammerman-Rozenberg, R., Cohen, A., Ein-Mor, E., Jacobs, J.M.: Physical activity, function, and longevity among the very old, vol. 169, pp. 1476–1483 (2009) 6. Westhoff, T.H., Franke, N., Schmidt, S., Vallbracht-Israng, K., Meissner, R., Yildirim, H., et al.: Too old to beneﬁt from sports? The cardiovascular effects of exercise training in elderly subjects treated for isolated systolic hypertension. Kidney Blood Press. Res. 30, 240– 247 (2007) 7. Barbour, K.A., Blumenthal, J.A.: Exercise training and depression in older adults. Neurobiol. Aging 26, 119–123 (2005) 8. Hakim, A.A., Petrovitch, H., Burchﬁel, C.M., Ross, G.W., Rodriguez, B.L., White, L.R., et al.: Effects of walking on mortality among nonsmoking retired men. New Engl. J. Med. 338, 94–99 (1998) 9. Calado, A., Leite, P., Soares, F., Novais, P., Arezes, P.: Boccia Court Analysis for Promoting Elderly Physical Activity (2018, submitted) 10. Leite, P., Calado, A., Soares, F.: Boccia Court Analisys for Real-Time Scoring(2018, submitted) 11. Fisk, A.D.: Designing for Older Adults (2007) 12. Lee, M., Park, J.: Senior user’s color cognition and color sensitivity features in visual information on web-interface. In: LNCS (LNAI and LNBI), vol. 8515, pp. 129–137. Springer, Cham (2014) 13. De Schutter, B.: Never too old to play: the appeal of digital games to an older audience. Games Cult. 6, 155–170 (2011) 14. Gerling, K.M., Schulte, F.P., Masuch, M.: Designing and evaluating digital games for frail elderly persons. In: Proceedings of the 8th International Conference on Advances in Computer Entertainment Technology - ACE 2011, vol. 1 (2011) 15. Silva, V., Ramos, J., Soares, F., Novais, P., Arezes, P., Sousa, F., Silva, J., Santos, A.: iBoccia: a framework to monitor the Boccia gameplay in elderly. In: LNCVB, vol. 27 (2018)

Construction of Multi-purpose Japanese Sign Language Database Yuji Nagashima(&) Faculty of Informatics, Kogakuin University, 2665-1 Nakano-Machi, Hachioji-Shi, Tokyo 192-0015, Japan [email protected]

Abstract. The study will build a database that can serve as the basis for common use and discussion by researchers by addressing research tasks. We will discuss the best source format, spatio-temporal resolution, format of data ﬁles, and storing method for academic ﬁelds such as linguistics and engineering. We are now considering the introduction of the 3-D motion data, multi-view images, and depth images. As for the 3D motion data, we will use the optical motion-capture technology, which has world-leading precision. The outcomes of this study have the potential to signiﬁcantly contribute to improved quality of life for the hearing impaired, for instance by establishing a new guarantee of information accessibility for those who use sign language as their primary language. Keywords: Japanese Sign Language (JSL) Collect JSL Optical motion capture (MoCap) 3D computer graphics

Database

1 Introduction Sign language is a means of communication used by the hearing impaired, and is a unique, interactive natural language with a grammatical system that differs from spoken languages. Although it is a language, research on sign language remains underdeveloped in linguistics, engineering, and other related ﬁelds when compared to spoken languages. One reason for this is the lack of a multi-purpose database for shared use by linguists, engineers, and researchers in other ﬁelds. Yet, it is likely that the data format desired by linguists is not the same as that desired in the ﬁelds of engineering and cognitive science. To facilitate research into sign language in many ﬁelds, we need a variety of data formats, including images depicting signing from multiple perspectives, 3D motion data, and depth images, that can be alternated depending on what is analyzed or parsed. In this paper, we consider methods for constructing a database of sign language vocabulary. First, we consider methods for selecting the sign language vocabulary that is recorded in the database. We then consider methods for selecting those who provide the language material. Next, we discuss methods for recording actual sign language footage as well as the results of our test recording in 2017. Lastly, we discuss our future research plan. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 595–599, 2019. https://doi.org/10.1007/978-3-030-02053-8_91

596

Y. Nagashima

2 Methods for Selecting Vocabulary and Informant Selection 2.1

Methods for Selecting Vocabulary

There exists a large number of sign language dictionaries. The dictionary with the largest number of entries is the Nihongo–shuwa jiten (Japanese–Sign Language Dictionary) published by the Japanese Federation of the Deaf [1]. This dictionary contains about 6,000 Japanese words. For our proposed database, we take the entries appearing most frequently in the sign language databases of the NTT Nihongo no goi tokusei (Lexical Properties of Japanese) [2], the Corpus of Spontaneous Japanese [3], and the sign language news on NHK Educational TV [4], and mark them as candidates for the vocabulary to be included. We then compile the ﬁnal list of entries for the database by picking those candidates that are included in the Nihongo–shuwa jiten. We aim to have selected about 5–6,000 entries by 2020. In 2017, we selected 400 entries and decided on sign motions as a trial. We decided on how to record the sign motions by analyzing existing dictionaries and discussing it among ourselves, research partner deaf native signers, and CODA (Children of Deaf Adults) native signers. 2.2

Informant Selection

Taking into account that the database will be open to the public, we will record the sign language vocabulary as performed by one man and one woman. We chose the following conditions for selecting informants. • • • •

Native signers in a native signer family Some experience with being ﬁlmed Sign language ease of readability Consent to making the footage public

After conducting interviews with native signers involved in this database construction project, we decided on Male M (38 yrs.) and Female K (39 yrs.).

3 Methods for Recording Sign Language Vocabulary To facilitate research into sign language in many ﬁelds, we need a variety of data formats, including images depicting signing from multiple perspectives, 3D motion data, and depth images, that can be alternated depending on what is analyzed or parsed. When constructing the database, we aim to synchronize three types of framerate data: 120 fps ultra-precision 3D motion using MoCap, 60 fps full-HD high-resolution footage, and 30 fps depth sensor and infrared footage. In order to achieve this, we decided to conduct the recording at the motion capture studio of Toei Zukun Lab, who possess the expertise and know-how. Figure 1 shows the conﬁguration of the recording equipment and synchronization conceptual diagram used in 2017. Figure 2 shows the layout of cameras and sensors. Table 1 shows the details for lengths and angles in Fig. 2.

Construction of Multi-purpose Japanese Sign Language Database

597

Fig. 1. The conﬁguration of the recording equipment and synchronization conceptual diagram used in 2017

Fig. 2. The layout of cameras and sensors.

598

Y. Nagashima Table 1. The details for lengths and angles in Fig. 2. Equipment Camcorder #01 Camcorder #02 Camcorder #03 Camcorder #04 Kinect #01

Angle [degree] Distance from signer [m] Direct distance x axis 0.0 3.27 yCc = 3.27 hL = 30.5 rL = 3.41 yL = 2.94 hR = 28.8 rR = 3.40 yR = 2.98 hCl = 7.4 rCl = 3.17 yCl = 3.14 0.0 2.51 yCk = 2.51

y axis 0.00 xL = 1.73 xR = 1.63 xCl = 0.41 0.00

Height 1.21 1.20 1.20 1.28 1.08

4 Viewer Development The viewer needs to be able to play the synchronized data described in the previous section despite variations in frame rate. For this, we developed a Unity-based viewer capable of synchronized playback. Our plan when developing the viewer was that it should be able to synchronize and play any four types of data. The viewer we developed is capable of synchronized playback by synchronizing 3D CG (BVH data) with point data (C3D) consisting of retro reflective markers for optical MoCap. Yet the limitations of Unity functionality caused lagging HD video playback when synchronizing full-HD video with 3D motion data. We are working on the development of a new viewer to resolve this playback lag issue. Figure 3 shows playback of BVH data (3D CG), C3D data (point data), frontal view, and Kinect-sensor depth data for sign language word {oNAJI(SAME)} as synchronized by the viewer.

Fig. 3. Playback of BVH data, C3D data, frontal view, and depth data for sign language word {oNAJI(SAME)} as synchronized by the viewer.

Construction of Multi-purpose Japanese Sign Language Database

599

5 Remaining Tasks In this paper, we discussed the concepts for our multi-purpose database under construction. This database uses synchronized recordings, which will allow analyzers to employ their preferred data formats as they perform synchronized analysis of data with varying frame rates along a time axis. Through this, it becomes possible for several researchers to analyze the same sign from their own different perspectives, which we anticipate should yield fresh insights. Next, our plan is to develop an annotation support system capable of synchronized analysis of various data formats in the database. Acknowledgments. This work was supported by JSPS KAKENHI Grant Number 17H06114.

References 1. Japanese/Japanese Sign Language Dictionary. Japan Institute for Sign Language Studies (ed.) Japanese Federation of the Deaf, Tokyo (1997) 2. Amano, S., Kasahara, K., Kondo, T., (written and eds.): NTT Database Series [Lexical Properties of Japanese] Volume 9: Word Familiarity Database (Addition). Sanseido, Tokyo (2008) 3. Corpus of Spontaneous Japanese. http://pj.ninjal.ac.jp/corpus_center/csj/ 4. Katou, N.: Japanese Sign Language Corpus on NHK News. In: 2010 NLP (the Association for Natural Language Processing) Annual Meeting, pp. 494–497 (2010)

Product Function Analysis: Reducing Cost of Production Service Line with Work Teams Velia Castillo-Pérez(&), Liliana Carrasco-Arméndariz, Mario Corral-Chacón, and Ramon Elizondo-Rios Division of Graduate Studies and Research, Tecnológico de Ciudad Juárez, Tecnológico Nacional de México, Av. Tecnológico 1340, Col. Crucero, 32000 Ciudad Juárez, Chihuahua, Mexico {vcastillo,lcarrasco,mcorral,relizondo}@itcj.edu.mx

Abstract. The study identiﬁes diverse methodologies related to new designs development and implementation, in conjunction with product function analysis for each component interacting into system assembly, through multidisciplinary work teams. Automotive gasoline module and its components evaluated looking forward improvement opportunities and cost reductions, affecting directly components shape and quantity. Results show up: material components change, assembly components modiﬁcation, components quantity goes up and down based on customer needs. Several methodologies combined as concurrent engineering, reverse engineering, and design for manufacturing among others are used. Prototypes constructed in 3D printer using selective laser sintering. Statistical tests were positive. Keywords: Design process Function analysis Work team Gasoline module

Systems integration

1 Introduction Automotive enterprises are looking for new projects establishing a Customer-Supplier strengthen economic ties. New project could be a long-term commitment producing service parts. Technology is evolving, bringing new alternatives to assembly same product, reducing material cost, innovative designs. Firms should continue producing new and service parts according to product life cycle. This study is focus in gasoline module; this product has its own set of components, a system of many parts working together to deliver gasoline demand into a vehicle under certain pressure, with certain flow, and nowadays the same system can measure gasoline internal volume contained in the tank. Several companies produce broaden variety of gasoline modules with different characteristics, such as quantity of components, shapes, sizes, material types, among thers, but at end assemble as a whole should meet the function required plus customer needs. Figure 1 Samples of gasoline modules, presents eight different modules, all have cover, reservoir is a container with certain characteristics allowing the rest of components to be assemble properly, gasoline full to keep bomb refrigerated. Diverse level sensors, arms and float valves. Different hoses and different ways to connect them. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 600–605, 2019. https://doi.org/10.1007/978-3-030-02053-8_92

Product Function Analysis: Reducing Cost of Production Service Line

601

Fig. 1. Samples of gasoline modules presents eight different modules, all have cover, reservoir is a container with certain characteristics allowing rest of components assembled properly, gasoline full to keep bomb refrigerated. Diverse level sensors, arms and float valves. Different hoses and different ways to connect them.

Competitive analysis performed to evaluate modules and their components with objective of ﬁnd out improvement opportunities and cost reductions. 20 different components are used and eight are used twice. However, gasoline modules tendency maintains signiﬁcant similarity. Study objective is reducing internal components of gasoline service module without affect system function. Family analyzed, ﬁrst time, has 55 different modules part numbers, 70 thousand pieces produced annually. Gasoline Module main functions are: (a) Supply gasoline, thru a gasoline bomb assembled in module and with an electrical sign, module will send gasoline through injection lines directly aimed to motor; (b) Controlling gasoline demand, regulate quantity required, at the time required. (c) Taking readings of flow level in tank and supply information to module.

2 Literature Review The proposal design will have: (1) same system than actual, as entry an electric signal, as output flow and pressure, (2) Less components without affect product function, (3) Gasoline bomb will function under same operation and assembly conditions.

602

V. Castillo-Pérez et al.

Design steps. First obstacle to design a product is lack of methodology, there is not a path to follow and reach product idea. Table 1 Design steps shows author Actual proposals developed as solution in ﬁeld to design. Table 2 presents different theories about product sensible to variety design. Table 1. Design steps Author [1] (Coha, Kostelic, & Rasmussen, 1990)

Year 1990

Topic Assembly contain diverse components supporting electric bomb

[2] (Löfqvist, 2009)

2009

Cyclical process is not follow Companies follow lineal flow at start and at ﬁnish as part of the job

[3] (Cross, N. 1997)

1997

Proposed a general diagram of design steps

Methodology Exploration Generation Evaluation Communication Problem exploration Concepts generation Evaluate and select Concept development Mutation Emergence Analogy Combination Basic principles

3 Components Selected to Improve Based on methodology presented by several authors a combination of them genera the one to used, steps are: (1) product function deﬁnition, (2) Components classiﬁcation, (3) Components justiﬁcation, (4), components to improve analysis and selection, (5) generation of proposals and alternatives, (6) Proposal analysis and selection, (7) prototypes construction, (8) tests and measurement. (1) Product deﬁnition: Service Gasoline Modules, Gasoline module has among 20 and 25 components. (2) Components classiﬁcation categories: buy/make, cost, material, manual o mechanic assembly, assembly/component, ﬁx/functional. (3) Components justiﬁcation: evaluate each component-using matrix of AIAG manual to identify critical components and deﬁne combinations, modiﬁcations or component removal. (4) Analysis and selection of components to improve: 1. Hydraulic Connectors. Value added for these connectors is low. Looking forward to remove them, combine them is recommendable. 2. 2. Plastic adaptor, presented on Fig. 3 It got 52 points, It goes assembled on metallic cover, its function is to be intermediary to assemble others components, receives among 3 to 5 distinct components according to model.

Product Function Analysis: Reducing Cost of Production Service Line

603

Table 2. Product sensible to variety design Author [4] (Bryan, A.; Ko, J.; Hu, S. J.; Koren, Y.;, 2007) [5] (Kimural, F.; Nielsen, J., 2005). [5] (Kimural, F.; Nielsen, J., 2005)

Year 2005 2007 2005

Topic Product families Co-evolution

2005

Demonstrate relationship between design and manufacture

[6] (Wagner, U.; AlGeddawy, T.; ElMaraghy, H.; Müller, E.;, 2014)

2014

Interchangeable products

[6] (Wagner, U.; AlGeddawy, T.; ElMaraghy, H.; Müller, E.;, 2014)

2014

Concurrent engineering

[7] (Aguilar, Jaime A.; Valencia, Manuel V.; Martínez, Manuel F.; Quiceno, Carlos A.; Sandoval, Claudia M.;, 2012)

2008

TRIZ

Methodology Design and process are not sensible to time Customer requires similar products Product variation Volume variation Complex facilities Complex product delivery Complex cost Concurrent engineering implement design and manufacturing simultaneously Idea generation Projection Concept development, test and evaluation Prototype development, test and evaluation Pre-Launching Launching Synergic job

Fig. 2. Hydraulic connector (a)/(b). Their only function is connect to ﬁlter assembly

604

V. Castillo-Pérez et al.

Fig. 3. Plastic adaptor It got 52 points, It goes assembled on the cover

3. Retainer, showed on Fig. 4, its punctuation is 17, lowest from analysis result, doesn´t affect most of functions, however its proposal is to place internal components such us bomb and ﬁlter among others. This component has remain for long time.

Fig. 4. Retainer, its punctuation is 17, lowest from analysis result, doesn´t affect most of functions, however its proposal is to place internal components such us bomb and ﬁlter among others. This component has remained for long time.

Pressure sensor also selected to be improved. Components selected analyzed, based on their characteristic and correlated with the rest. Next step was proposals and alternatives, using brainstorming.

Product Function Analysis: Reducing Cost of Production Service Line

605

4 Results

Fig. 5. Cover with geometry removed, redesigned replace hydraulic connectors and assembly hoses is done directly (left) Connectors combination facilitates assembly and reduces costs (right)

5 Conclusions Flow test and pressure internal component simpliﬁcations meet functionality principles. Statistical analysis veriﬁed proposal reach similar behavior than reference module. Estimated savings 51,000 dollars, but project became very attractive to the company.

References 1. Aguilar, J.A., Valencia, M.V., Martínez, M.F., Quiceno, C.A., Sandoval, C.M.:. Uso de la Teoría de Solución de Problemas Inventivos (TRIZ) en el análisis de productos de apoyo a la movilidad para detectar opotunidades de innovación. Ingeniería y Competitividad, pp. 137– 157 (2012) 2. Beyer, S.E., Jones, D.R., Catlin, M.L., Sawert, U., Hilderbrant, M., Wolfender, M.K., y otros. Patente nº US2003/0094458 A1. United States 3. Bryan, A., Ko, J., Hu, S.J., Koren, Y.: Co-Evolution of Product Families and Assembly Systems. CIRP, pp. 41–44 (2003) 4. Coha, T.F., Kostelic, R., Rasmussen, G.K.: Patente nº Patente no.4,945,884. United States (1990) 5. Cross, N.: Creativity in design: analyzing and modeling the creative leap. Leonardo 30, 651– 656 (1997) 6. Kimural, F., Nielsen, J.A.: Design method for product family under manufacturing resource constraints. CIRP Annals Manufacturing Technology, pp. 139–142 (2005) 7. Löfqvist, L. Design processes and novelity in small companies: a multiple case study. In: International Conference on Engineering Design, pp. 24–27. Stanford University, Stanford (2009) 8. Wagner, U., AlGeddawy, T., ElMaraghy, H., Müller, E.: Product family design for changeable learning factories. In: CIRP, pp. 195–200 (2014)

Systemic Approach for Inclusive Design of Low-Income Dwellings in Popular Settlements at Guayaquil, Ecuador Boris Forero(&), Jesus Rafael Hechavarría Hernández, Silvia Alcivar, and Virginia Ricaurte Facultad de Arquitectura y Urbanismo, Universidad de Guayaquil, Guayaquil, Ecuador {boris.forerof,jesus.hechavarriah,silvia.alcivarm, maria.ricaurter}@ug.edu.ec

Abstract. Traditionally low-income households are deﬁned as “low-cost housing”, however, the concept of Social Interest Housing -Vivienda de Interés Social- (VIS) deﬁned in this research is made to emphasize that it is a priority to consider human beings together with their habitat. In Ecuador, the population growth for 2050 will increase by 35%, added to this is the polarization in terms of construction of new housing: where for every 10 solutions in Guayaquil, 70% correspond to self-built dwelling environments. In the application of Quality Function Deployment (QFD) methodology, teachers and students of the Faculty of Architecture and Urbanism of the University of Guayaquil performs ﬁeldwork surveys in order to develop multidisciplinary research aimed to designing low-income housing through research projects. Among the speciﬁc objectives are the analysis and propose of prefabricated construction systems, the design of low-income housing prototypes and the interior design and multifunctional furniture by ergonomic, anthropometric and social analysis. Keywords: Social inclusion Concurrent engineering Bioclimatic design Low-income housing

Participatory design

1 Introduction The systemic approach allows approaching the object of study in an integral way where various indicators taken into account, within which patterns, flows and relationships that articulate them are recognized. In architecture a global vision is interpreted where each component part of the housing and its habitat corresponds to an organic whole in respect of multi-dimensionality and complexity, that must be studied together to ﬁnd a reasonable compromise between costs, comfort and care for the environment, in order to offer healthy and safe dwelling prototypes for the less favored sectors of the population [1]. The different architectural and urbanistic plans of housing projects for low-income groups, historically focus on this character or integrative, multidisciplinary and participatory vision, which far from being a solution, has been part of the problem of popular housing [2]. © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 606–610, 2019. https://doi.org/10.1007/978-3-030-02053-8_93

Systemic Approach for Inclusive Design

607

It is imperative that cities and other territorial entities provide for reasonable development in their territories; and speciﬁcally in the case of housing, where an increase in population density is required (also with vertical solutions to take better advantage of the urban space), thus aiming at a better quality of life and at a sustainable development where the social, environmental and economic ﬁnd its balance, where multiple efﬁciency criteria that will be modeled through heuristics based on multiobjective optimization [3, 4]. The above, hand in hand with participatory design methods where inhabitants can be part of the decision making to guarantee the sense of belonging with projected solutions [1].

2 Goals Main goals of this research are to develop prototypes of social interest housing in the search for a reasonable compromise between cost, comfort and care for the environment using prefabricated building systems that allows increasing not only productivity in construction but densiﬁcation of housing by multi-storey projects for low-income settlements of zones 5 and 8 of Ecuador. In order to achieve this, research and propose of prefabricated construction systems, and design prototypes of low-income housing with interior design and multifunctional furniture made by ergonomic, anthropometric and social analysis [3]. In this order, the National Secretariat of Planning and Development (SENPLADES) reiterates the propitiation of adequate conditions for access to a safe and inclusive habitat. In this project the interdisciplinary and interinstitutional work is proposed by linking the University, private companies and government agencies responsible for housing management to ensure the generation of more comprehensive proposals, where the economic, social and care factor to the environment have the same importance as inclusive design does [3, 5, 6].

3 Activities and Results The University in response to its social commitment must propose study and research proposals on these housing problems to contribute with residential solutions according to the diverse human, geographic and historical realities of the popular settlements. With regard to this topic, the document that traces the road map of the Faculty of Architecture and Urbanism of the University of Guayaquil has the name “Territorial Planning, Urbanism and Constructive Technology Systems”, where it raises this problem of the housing deﬁcit, and the expansion of urban borders. The commitment of this educational institution will be the contribution to the country to these issues of great national interest also endorsed by Republics’ Constitution of 2008, which in its Article 30, states that every citizen regardless of their social or economic situation will be entitled to an adequate and digniﬁed housing [6]. The results obtained so far, include architectural prototypes designed from statistical analysis, and the implementation of active and multidisciplinary learning

608

B. Forero et al.

methodologies that allow obtaining more coherent answers, responding to the different settlements, cultures and human groups that inhabit them [1, 3]. Through inclusive design where the criteria of territory and climate in addition to humans, are part of the solution, which has been put into practice in the suburb of Guayaquil, in addition to “Cerro Las Cabras” and “Mision Manuela Espejo” [6] in Duran. Currently, work is being carried out at “Buijo Historico” and “Sabanilla” in Daule’s rural area where semi-structured interviews held in ﬁeld. The dynamics developed in the different localities of Zone 8, become learning and knowledge that is replicated in each new place of study and are carried out with the collaboration of under-graduate students of architecture and interior design as part of their activities of pre-professional practices and community work (Fig. 1).

Fig. 1. Participatory design with the community at Guayaquil and VIS Project. This method, involves people in its own solutions and realities, in search of “home of one’s own in the city” (left, center and right).

Scientiﬁc production developed by VIS Project during 2016–2018 period as result of the various activities and processes indicated below, reflect the systemic and multidisciplinary character of this research project [7]. Regarding construction systems, the research has focused on the dimensional characterization of building models of lowincome housing projects built at Guayaquil. In addition to the design of prefabricated panels of cane-steel-mortar micro vibrated with rice husk ash [8]. This as part of a Doctoral research. Studies of projected or designed spaces and of the space required within the dwellings and the spatial changes suffered during the occupation by the original design and its impact on the spatial transformation that this brings to the urban landscape were made at low-income settlement of El Recreo in Duran [9]. Studies conducted on the user’s perception of their home and the environment in social interest programs, carried out in housing program Manuela Espejo, in Duran. It is analyze under variables of space, comfort, construction, safety, signiﬁcance, mobility and equipment [6]. From a bioclimatic perspective, studies were conducted for the improvement of thermal conditions in low-income state-led program called SocioVivienda, with

Systemic Approach for Inclusive Design

609

analysis of construction systems, architectural design, ﬁnishes and colors that allow us to approach a climatic reality that has not been taken into account for architectural design [10, 11]. It has also carried out participatory design activities, whose ﬁrst stage and through the results of surveys conducted in the ﬁeld, is better known the reality of the place and its inhabitants, allowing an optimal design. This, in activities with direct participation of the community in workshops with students and teachers of the University of Guayaquil applying Inclusive Design Methods and active learning [1].

4 Conclusions The design of prototypes of social housing starts from the study of the social, economic and environmental context, thus characterizing the territory where the project will be placed and the household, who is the protagonist in the decision making process. With this knowledge, it is possible to give coherent answers integrating the design of the interior furniture to the houses of social interest, taking into account the ergonomic and anthropometric analysis of the community; all as part of a systemic design process. Participatory design, in the other hand, is a key tool of ownership and sense of belonging of communities, and in the case of low-income settlements, it acquires special meaning because its roots and past are recognized. In such a way that you learn to design with more human eyes from these experiences in community workshops. Acknowledgments. The authors would like to thank to School of Architecture and Urbanism, at University of Guayaquil for the support to VIS Project.

References 1. Hechavarría, J., Forero, B., Al-Terkawi, J.: Enfoque sistémico como propuesta metodológica para el diseño de Viviendas de Interés Social en estudiantes de arquitectura de la Universidad de Guayaquil. Opuntia Brava, vol. 9 (2017) 2. Hechavarría, J., Forero, B., Portilla, Y.: Propuesta metodológica para el diseño de Viviendas de Interés Social en la Carrera de Arquitectura de la Universidad de Guayaquil. 4° Congreso Internacional de las Ciencias Pedagógicas de Ecuador. Guayaquil: Instituto Superior Tecnológico Bolivariano (2018) 3. Hechavarría, J., Forero, B., Arzola, J.: Enfoque sistémico para la optimización multiobjetivo del diseño de viviendas de interés social. Aporte Santiaguino 8, 157–170 (2015) 4. El-Anwar, O., Aziz, T.: Integrated urban-construction planning framework for slum upgrading projects. J. Constr. Eng. Manag. (2014) 5. Forero, B., Hechavarría, J., Sandoya, R.: Diseño bioclimático de viviendas de bajo costo. Una propuesta metodológica para estudiantes de la Carrera Arquitectura, Guayaquil. Opuntia Brava, vol. 9 (2017) 6. Ricaurte, V., Hechavarría, J.: La percepción del usuario sobre su vivienda y el entorno en programas de interés social en Durán, Ecuador. (U. Ecotec, ed.) Ecociencia (2017). ISSN 1390-9320

610

B. Forero et al.

7. Hechavarría, J., Forero, B.: Aplicación de la metodología de Análisis y Síntesis de Sistemas de Ingeniería en la búsqueda de soluciones a problemas de la sociedad. In: III Congreso Internacional Tecnología, Universidad y Sociedad (TUS). Universidad de Especialidades Espiritu Santo, Guayaquil (2015) 8. Almeida, S., Anaya, J., Hechavarría, J.: Caracterización dimensional de modelos constructivos de Vivienda de Interés Social construídos en Guayaquil, Ecuador. In: Libro de Memorias Cientíﬁcas del V Conrgreso Internacional de investigación y actualización en ingenierías, pp. 234–249. CIDE, Guayaquil (2017) 9. Alcívar, S., Morales, S., Forero, B.: Análisis funcional de la Vivienda de Interés Social: El Recreo, Cantón Durán 2017. Conrado 14, 99–104 (2018) 10. Forero, B.: Mejoramiento de las condiciones térmicas de las viviendas del complejo habitacional Socio Vivienda II Etapa I, en la ciudad de Guayaquil, Ecuador. Master’s thesis, Universidad de Colima (2015) 11. Forero, B., Hechavarría, J.: Análisis de las condiciones de confort térmico en el interior de las viviendas del complejo habitacional Socio Vivienda 2, Etapa 1, en la ciudad de Guayaquil, Ecuador. In: III Congreso Internacional Tecnología, Universidad y Sociedad (TUS). Universidad de Especialidades Espiritu Santo, Guayaquil (2015)

IHSED 3: Artificial Intelligence and Intelligent System Design

VibroTac S: An Electronic Assistive Device for Blind and Visually Impaired People to Avoid Collisions Simon Schätzle(&), Thomas Hulin, and Benedikt Pleintinger German Aerospace Center (DLR), Oberpfaffenhofen, Germany {Simon.Schaetzle,Thomas.Hulin, Benedikt.Pleintinger}@dlr.de

Abstract. This paper presents a new concept for an electronic assistive device for blind and visually impaired people. Due to the integration of tiny distance sensors into the vibrotactile wristband VibroTac®, this novel device enables to detect obstacles in the user’s vicinity and thus to avoid collisions. The simultaneous monitoring of areas in different directions up to about one meter as well as an intuitive notiﬁcation of the direction and the distance of detected objects through corresponding vibrating actuators is unique and makes its use very intuitive. A prototype was tested with three blind people in a mobility and in a table scenario. We received very positive and promising feedback and blind people rated the device to have great potential to be a useful aid in their daily life. Keywords: Electronic assistive device for blind and visually impaired people Collision avoidance Vibrotactile feedback Electronic aid

1 Introduction Usually, blind or visually impaired people use white canes to scan their surroundings for obstacles and thus to cope in unfamiliar environments. This kind of mobility aid is well suited to detect obstacles on the floor and to get information of the ground’s surface. However, the classical white cane cannot prevent from hazardous collisions with obstacles in chest or head height. Therefore, white canes have been upgraded with sensors and corresponding feedback units in order to extend the scanned area. For instance, the Ultracane incorporates two ultrasonic sensors for detecting obstacles on the ground within 2 or 4 m and up to 1.5 m ahead in chest and head height [1]. Feedback is given through two vibrating buttons on the handle. The company Vistac developed another modiﬁed white cane [2]. Based on laser, the beam detects obstacles located just above the stick in the chest and head area of the user. Detected obstacles are indicated through vibrations in the handle. Surprisingly, such modern and beneﬁcial assistive devices are rarely used in real life. The following potential reasons for the low distribution were mentioned from blind persons: (a) high purchase price, (b) heavy and impaired center of gravity and/or (c) unfavorable location of feedback (vibrations resulting from the stick end when gliding over the ground gives clue about surface structure but these vibrations are distorted by the vibrotactile feedback). © Springer Nature Switzerland AG 2019 T. Ahram et al. (Eds.): IHSED 2018, AISC 876, pp. 613–619, 2019. https://doi.org/10.1007/978-3-030-02053-8_94

614

S. Schätzle et al.

Apart from modifying the white cane, there are other approaches to realize collision avoidance in chest and head height. Cardin et al. introduced an obstacle detection system consisting of two sonar sensors attached to the shoulders [3] and demonstrated its usability to reach a step forward in the integration of visually impaired people. The “Sunu Band” is a wristband to complement the white cane and uses a sonar cone to detect objects in one direction up to 5.5 m away. The objects’ distances are indicated through vibration feedback [4]. Another example is “Tacit”, a hand mounted haptic feedback sonar obstacle avoidance assistive device. It is designed to navigate complex environments by measuring in distal direction. The distance to objects is translated into a pressure on the wrist [5]. The upper body protection device such as the chestmounted system “Ultra-Body-Guard” [6] scans in one direction and indicates obstacles by vibrations either on the neck or in the hand. “Ray” is an ultrasonic mobility aid, developed to complement the white cane [7]. It is a handheld device comparable in its use like a torch. Recognized objects in the ultrasonic cone are displayed to the user via vibrations or acoustic signals. The direction of the detected obstacle is known from the direction in which the device is held. Many devices or concepts of the state of the art require active scanning movements because of a one-dimensional or fan-like detection zone. We are not aware of any mobility aid that allows for simultaneous multi-dimensional scanning that is neither hand-held nor stigmatizing (e.g. sensors placed around the head) and that provides intuitive direction and distance feedback of detected obstacles.

2 System Description In the following, we introduce a new concept for an electronic assistive device for blind and visually impaired people which aims to assist in two different ways: (a) as mobility aid for the detection of obstacles in chest or head height and of sideward or backward approaching obstacles and (b) as an device to avoid knocking over objects in environments that are unknown or shared with other people and where the position of objects are altered unnoticed by the blind person (for instance at a table when an object with known position is to be grasped but meanwhile other objects were positioned in the way). 2.1

Requirements

We deﬁned the following requirements in order to accomplish the aforementioned goals: (1) simultaneous scanning of several areas in different directions, (2) detection of areas within the motion range of the person/arm, (3) detection distance up to approx. 0.8 m (this is approx. equal to the radius of the detected area with a white cane; with a presumed walking speed of 1.5 m=s, a reaction time of t ¼ 1:5 s results) whereas for the table scenario, a detection distance of about 0.3 m should be enough, (4) intuitive tactile feedback of direction and distance of detected objects, (5) hands should be free for other tasks e.g. manual task or holding the white cane, (6) compact and lightweight, (7) easy to use and unrestricted motion range, (8) measurement rate above 10 Hz in order to avoid noticeable delay (and noticeable distance s ¼ v t ¼ 1:5 m=s 0:1 s ¼ 0:15 m at 10 Hz).

VibroTac S: An Electronic Assistive Device

2.2

615

Vibrotactile Feedback Device

We modiﬁed the vibrotactile feedback device VibroTac that was developed at our institute. The VibroTac, depicted in Fig. 1, is a patented ergonomic wristband that provides position- and direction-speciﬁc output of information thanks to six vibration segments (tactors) that are linked elastically with each other. The intensity and signal shape of each tactor can be controlled individually so that versatile information such as collision feedback for Virtual Reality or telerobotic applications, motion guidance of the human hand, or distance data can be generated [8].

Fig. 1. The vibrotactile feedback device VibroTac and the novel concept of an electronic assistive device to avoid collisions

The integration of distance sensors into the VibroTac enables the simultaneous scanning of several areas in different directions as shown in Fig. 1. If an object is within the detection range of a sensor, the corresponding tactor is activated in order to inform intuitively about its direction and distance. Detailed information about the generation of direction and distance information with the VibroTac can be found in [8]. 2.3

Potential Sensor Technologies

Common sensors for the detection of objects and their distance typically use inductive, capacitive, acoustic or photoelectrical operating principles. Based on the following criteria, the suitability of the different principles for the present application was assessed: material and surface independence, detection range, size (should not exceed the size of a tactor), weight (10 Hz). Based on literature and on datasheets of commercial products, three different ratings (“+” well, “o” possibly and “−” not suited) were used in Table 1 rather than concrete value intervals as the focus is on the rating of the general suitability of the different principles in the context of the aimed application. Ultrasonic sensors and photoelectric reflex sensors with infrared light are principally appropriate, though the maximum distance of the latter one depends on the surface’s degree of reflection and on sunlight. P Operating principles that measure the amount of reflected light ( energy) might lead to signiﬁcant variations of distance measurement results for objects in the same distance but with different surface properties. Regarding infrared sensors, only working principles with triangulation or time of flight (TOF) should be considered in this application.

616

S. Schätzle et al. Table 1. Rating of different distance sensor working principles

2.4

Evaluation of Five Sensors

We chose three ultrasonic and two infrared (one triangulation, one TOF, see Table 2) sensors in order to compare real measurement results with expected results based on the datasheets with representative table scenario objects (different size and surface properties: plastic bottle with label, transparent plastic bottle, black fabric, white cardboard, aluminum can, glass bottle without label). For the sake of brevity, only a summary is given but details such as a description of measurement procedure, plots and analysis of data can be found in [9]. Table 2. As appropriate considered sensors that were chosen for detailed evaluation. Nr. Example

Range (m) Resolution Update Influence reflectance LxWxH (mm) Cost (€) (cm) rate (Hz)

S1 S2 S3 S4 S5

0.15–6.45 0.15–6 0.04–6 0.1–0.8 0–0.6

ultrasonic ultrasonic ultrasonic infrared: triang. infrared: TOF

2.54 3 3 – 0.3

20 15 15 19 15

– – – small small

22x20x16 24x20x17 32x15x10 30x13x14 5x3x1

28 14 37 10 5

Two of the three ultrasonic sensors were tolerant to surface properties and showed reasonable distance results up to 0.8 m. Sensor S2 had difﬁculties to detect the plastic bottles and the can (see Fig. 2). Possible reasons might be a total (signal is deflected at smooth surface) or double (indirect reflections are received) reflection. Furthermore, the sensor size is adverse. Surprising results were obtained with the infrared sensor S4. Despite of its triangulation working principle, the measured distances were signiﬁcantly

VibroTac S: An Electronic Assistive Device

617

influenced by the object’s surface properties: similar output values were generated e.g. for the glass bottle in 0.19 m distance and the green transparent plastic bottle in 0.8 m distance.

0,9

2,0

0,8

1,5 0,7

1,0 0,6

0,5

0,5

0,4

0 0

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1,2 1,4 1,6 1,8 2,0 0,3

100 % 0,2

75 % 50 %

0,1

25 % 0

0% 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1,2 1,4 1,6 1,8 2,0

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

Fig. 2. Measurement results of ultrasonic sensor S2 (left) and infrared sensor S4

Very precise results were achieved with the time-of-flight sensor S5: while the distance correctness is not affected by the reflectance properties, the maximum measurable distance decreases with increasing absorption coefﬁcient (dmax ¼ 0:35 m). For the evaluation of the patented “VibroTac S”-concept [10], a prototype (see Fig. 3) was built with time-of-flight sensor S5 due to its advantages: reliable distance measurement, little power consumption (

Human Systems Engineering and Design

Recommend Stories

Idea Transcript

Helpful Links

Smile Life

Get in touch