Parallel Services: Intelligent Systems of Digital Twins and Metaverses for Services Science 3031253329, 9783031253324

Citation preview

SpringerBriefs in Service Science Lefei Li · Fei-Yue Wang

Parallel Services Intelligent Systems of Digital Twins and Metaverses for Services Science

SpringerBriefs in Service Science Series Editor Robin Qiu, Division of Engineering & Information Science, Pennsylvania State University, Malvern, PA, USA Editorial Board Members Saif Benjaafar, Industrial and Systems Engineering, University of Minnesota, Minneapolis, MN, USA Brenda Dietrich, Cornell University, New York, USA Zhongsheng Hua, Zhejiang University, Hefei, Anhui, China Zhibin Jiang, Management Science, Shanghai Jiao Tong University, Shanghai, China Kwang-Jae Kim, Pohang University of Science and Technology, London, UK Lefei Li, Department of Industrial Engineering, Tsinghua University, Haidian, Beijing, China Kelly Lyons, Faculty of Information, University of Toronto, Toronto, ON, Canada Paul Maglio, School of Engineering, University of California, Merced, Merced, CA, USA Jürg Meierhofer, Zurich University of Applied Sciences, Winterthur, Bern, Switzerland Paul Messinger, Alberta School of Business, University of Alberta, Edmonton, Canada Stefan Nickel, Karlsruhe Institute of Technology, Karlsruhe, Baden-Württemberg, Germany James C. Spohrer, IBM University Programs World-Wide, IBM Almaden Research Center, San Jose, CA, USA Jochen Wirtz, NUS Business School, National University of Singapore, Singapore, Singapore

SpringerBriefs present concise summaries of cutting-edge research and practical applications across a wide spectrum of fields. Featuring compact volumes of 50 to 125 pages, the series covers a range of content from professional to academic. Typical publications can be: A timely report of state-of-the art methods A bridge between new research results, as published in journal articles A snapshot of a hot or emerging topic An in-depth case study A presentation of core concepts that students must understand in order to make independent contributions SpringerBriefs are characterized by fast, global electronic dissemination, standard publishing contracts, standardized manuscript preparation and formatting guidelines, and expedited production schedules. The rapidly growing fields of Big Data, AI and Machine Learning, together with emerging analytic theories and technologies, have allowed us to gain comprehensive insights into both social and transactional interactions in service value co-creation processes. The series SpringerBriefs in Service Science is devoted to publications that offer new perspectives on service research by showcasing service transformations across various sectors of the digital economy. The research findings presented will help service organizations address their service challenges in tomorrow’s service-oriented economy.

Lefei Li • Fei-Yue Wang

Parallel Services Intelligent Systems of Digital Twins and Metaverses for Services Science

Lefei Li Department of Industrial Engineering Tsinghua University Beijing, China

Fei-Yue Wang Institute of Automation Chinese Academy of Sciences Beijing, China

ISSN 2731-3743 ISSN 2731-3751 (electronic) SpringerBriefs in Service Science ISBN 978-3-031-25332-4 ISBN 978-3-031-25333-1 (eBook) https://doi.org/10.1007/978-3-031-25333-1 © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed 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, expressed 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

Starting from the beginning of this century, service science/engineering has been a standalone and fast-emerging discipline that declares the dominance of service economy. Traditionally, the complexity of any service system comes from the nature of service: simultaneity, intangibility, heterogeneity, and perishability. Nowadays, however, the rapid adoption of technologies, especially ICT, makes information go beyond just one aspect of a service system, but a major, and even the dominant one. AI, big data, cloud computing, blockchain, VR/AR, etc. make the design and management of service systems far more challenging and far more fascinating than ever in the history of mankind. Parallel intelligence, evolving from parallel control and management, has been recognized as a promising methodology for the modeling and analysis of complex systems, especially for complex CPSS (Cyber-Physical-Social Systems). Based on years of studies of parallel intelligence and its applications in various service industries, we establish a new approach for service system: parallel service. The theoretical foundation of parallel service lies in the actual-artificial duality of services, which comes from the Karl Popper’s model of reality and corresponding three axial ages. With the ability to model service systems as artificial service systems, perform computational experiments and parallel learning. Parallel service becomes the bridge that connects the "Cognitive Gap" between the actual and artificial world. The enabling technologies, including decentralized technology, multiagent simulation, and data fusion, etc., provide a handful "toolbox" that supports the realization of parallel service. In the application studies that cover various typical service industries like transportation, retailing, healthcare, etc., parallel service demonstrates three unique advantages: interdisciplinary integration, smarter service decision making, and enabling human-centered service innovations. In recent years, along with the trend of digital transformation, digital twin and meta-verse appear to be the next "engine" Parallel service naturally serves as an enabling mechanism for digital twin and metaverse in service systems. Our true DAO model in parallel service explains the unique power toward that direction. v

vi

Preface

The target of this book is to provide the readers a new way of thinking of a service system. For academic readers, it provides a new research area within the service science/engineering domain, incorporating vast of interdisciplinary advancements. For practitioners, with the help of dedicated methods and sample cases, the book shall provide opportunities to enhance the design and management of all types of modern service systems. Beijing, China October 22

Lefei Li Fei-Yue Wang

Acknowledgments

For many years, we have been conducting research on parallel intelligence and applying the concept to various service industries. In this book, we build on our past research to develop a new approach to services systems: Parallel Services. Parts of Chaps. 3 and 7 on parallel intelligence have been published in our other papers. The parallel services framework in Chap. 4 was published as a PSM concept in my paper “Parallel Service Management Framework and Application to Railway Station Layout Planning” in IEEE Intelligent Systems. The authors would like to acknowledge all those who participated in the process of compiling and reviewing this book for their help, without which this project could not have been successfully completed. And this work was supported by my students. They are Tsinghua University PhD students Ridong Wang, Yudan Lu, Yuchen Liang, and master’s student Zibing Zhan. They contributed throughout the process from the initial idea to the final publication. In addition, this work was supported by the National Key R&D Program of China under Grant No. 2020AAA0103804.

vii

Contents

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 6

2

Motivation: Complexity of Service in the Digital Age . . . . . . . . . . . . . . . . . . . . 2.1 Trends of Services in the Digital Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Smart Services with Smart Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Retailing, Logistics, and Financial Services Based on Artificial Intelligence Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3 Technology Applications in Services for Emergencies . . . . . . . . 2.2 Complexity of Services System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Challenges in the Digital Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9 9 9 11 12 16 17 19

3

Opportunity: The Actual-Artificial Duality of Services . . . . . . . . . . . . . . . . . . 3.1 Three Worlds and Three Axial Ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 The “Cognitive Gap” Between Two Worlds . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Parallel Services as a Bridge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 From CPS to CPSS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 The Future of Parallel Services Based on True DAO. . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

21 21 23 24 25 26 26

4

Framework of Parallel Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Definition and Vision of Parallel Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Framework of Parallel Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

29 29 31 33

5

Enabling Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 ACP Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Artificial Services System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.1 The Services Need–Demand Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.2 The Services Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.3 Parallel Learning and Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Design Thinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

35 35 36 36 37 37 38 ix

x

Contents

5.4 Systems Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 6

Enabling Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Decentralized Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Multi-Agent Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Data Fusion Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

41 41 43 44 45

7

Research on Parallel Services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Parallel Transportation Management Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.2 Parallel Transportation Management Systems . . . . . . . . . . . . . . . . . 7.1.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Parallel Healthcare Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.2 Design of Hybrid Services System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.3 Computational Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.4 Parallel Execution of the Internet Hospitals . . . . . . . . . . . . . . . . . . . . 7.3 Parallel Retailing Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.2 Design of the Artificial Services Systems . . . . . . . . . . . . . . . . . . . . . . 7.3.3 Computational Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.4 Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Parallel Logistics Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.2 Parallel Logistics Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49 49 49 50 51 54 54 55 58 59 59 59 61 63 63 65 65 66 67

8

Parallel Services and Digital Twins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 Introduction of Digital Twins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Parallel Services and Digital Twins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69 69 70 70

9

Parallel Services Metaverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 Introduction of Metaverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1.1 The Basic Concept of Metaverses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1.2 The Value Proposition Behind Metaverses . . . . . . . . . . . . . . . . . . . . . 9.2 CPSS for Metaverses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.1 Parallel Intelligence for Metaverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.2 The Essence of Parallel Services Metaverses . . . . . . . . . . . . . . . . . . 9.3 DAOs for Parallel Services Metaverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.1 “TRUE DAO” Toward Deep Intelligence. . . . . . . . . . . . . . . . . . . . . . . 9.3.2 Enabling Technologies for DAOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

71 71 71 73 75 75 76 77 77 79 81

Chapter 1

Introduction

With the growth of service economy, service as a science has long gained the continued attention of academia and industry. Across the business spectrum, services have generally became a growth engine. According to the data of World Bank[1], the value added of global service industry has continued to increase in the past three decades, and now it accounts for more than 60% of the GDP. Among developed countries such as the United States, Japan, Germany, and the United Kingdom, the value added of service industry has reached more than 70% of the GDP, and that of developing countries such as Brazil, South Africa, and Mexico has also reached about 60%. China has also entered the era of service economy, and the contribution of service industry to economic growth and employment has been increasing. In terms of GDP ratio, the average annual growth rate of China’s service sector value added exceeded 7% from 2013 to 2021, and the share of service sector value added in GDP rose from 45.5 to 53.3% from 2012 to 2021. In terms of contribution rate to economic growth, the contribution rate of service sector to economic growth rose from 45 to 63.5% from 2012 to 2019, higher than the share of service sector value added in GDP in the same period. In terms of employment contribution, the cumulative increase in employment in the service sector is 83.75 million, with an average annual growth rate of 3.0% from 2013 to 2021 (Fig. 1.1 [1]). Services are indispensable for economic and social functioning and improving the quality of life, besides of basic survival needs required by people. Figure 1.2 scopes some kinds of the services [2]. The combination of digital technology and vertical scenes has created new service business models, derived new industry forms, reshaped the industrial ecology, and provided a new exploration path for the high-quality development of the service industry. Services landscape and areas are changing rapidly, creating numerous opportunities for service innovation. Due to online and offline co-production, the same physical space can achieve more diversified consumption activities, more convenient and targeted service mode. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_1

1

2

1 Introduction

Fig. 1.1 World services, value added (% of GDP) [1]

Along with the spillover effect of technological innovation, in daily life, new businesses and new models such as online education, webcasting, e-sports, crossborder e-commerce, and fresh e-commerce are rapidly gaining popularity. In the field of manufacturing, many manufacturing enterprises have embarked on the path of servitization. There are a lot of terms to demonstrate this significant trend. Studies show that the terms “servitization” [9] and “service transformation”[3, 6] are commonly used to denote the process of service growth. They are defined as the transformation process from a product-centric business model and logic to a service-centric approach[4]. Suppliers are more focused on customers and value creation. While, the term “product-service systems” (PSS) [5] is usually used to describe the innovative combination of goods and services. IBM changed from a business model that relied on selling computer equipment and software to a model that relied on service delivery and service innovation for competitive advantage and growth [8]. Xerox introduced pay-per-copy services to support its new products. Xerox’s business model then shifted from pay-per-use to an annuity-based business model that focused on generating recurring revenue and

1 Introduction

3

Fig. 1.2 Role of services in an economy [2]

cash through bundled contract services, equipment maintenance, consumables, and financing [18]. Product lifecycle management changes the traditional single product sales model and utilizes information technology for the whole process of product management from R&D, production to sales and maintenance, which can extend the service system and innovate the way of value-added product services.

Caterpillar’s Equipment Lifecycle Management Caterpillar provides full lifecycle services from equipment delivery, through routine operation and maintenance, to major repairs and rebuilds. It continues adding value to equipment, helping customers to reduce overall holding and operating costs, and ultimately improving asset utilization. Caterpillar offers its customers maintenance services that include equipment maintenance plans and long-term maintenance agreements. In addition to traditional troubleshooting and on-site maintenance, the company and its customers are able to monitor operational status of their equipment in real time through the use of intelligent components, sensors, and monitoring platform as well as diagnostic software. By doing so, Caterpillar is able to provide operators and service personnel with information about potential problems associated with the machine, as well as instructions on proper (continued)

4

1 Introduction

operation, whether modifying machine operation, notifying the shop that maintenance is required, or performing a safety shutdown of the machine. It can be controlled remote. In addition, Caterpillar offers a wide range of financing and leasing services, such as credit applications for equipment, multiple discounts for used machines, parts, and services, and convenient online account management tools. Moreover, for products at the end of their lifecycle, Caterpillar is able to restore the specifications and performance of old parts through remanufacturing technologies to create remanufactured parts and remanufactured products. Caterpillar’s highly efficient service network through its agent model also enhances its service capacity. Caterpillar selects local small- and mediumsized enterprises as agents and establishes long-term and stable exclusive agency relationships with them. First, professional, stable, and reliable aftersales service is provided through agents. With a large distribution system, Caterpillar’s rigorously selected agents not only sell and lease products, but also provide a series of services such as technical support, maintenance, training, finance, and insurance, which is called “one-stop service.” Because of their proximity to customers, agents are able to provide quick after-sales service to minimize losses caused by machine breakdowns. Caterpillar is committed to providing the parts and after-sales service needed for Caterpillar products anywhere in the World within 24 hours.

Value-added information service mode uses big data, cloud computing, and Internet of Things platform to provide customers with online support and digital content value-added services through intelligent information products. JD home is one of the examples.

JD Home Cloud Design Platform JD Home launched its cloud design platform, Jingdong Designer, to build a digital and scenario-based system ecology with design as the core of the whole chain, through a multi-product matrix of VR model rooms and Dongdong intelligent design tools. It is aimed at building bridges between consumers and professional service personnel, constructing whole service process, realizing the accurate matching of goods and consumer needs, and meeting the consumer’s guarantee for delivery results. Jingdong Designer creates VR model rooms with different styles to assist users’ purchase decisions with immersive scenario-based experiences. For the consumers without clear idea of what they want to do, it can also use AI design tool to generate on-demand design solutions with simple click.

1 Introduction

5

In theory, services are recognized as applications of expertise, skills, and experience to create value for both consumers and suppliers [8]. Proposed in 2004, service-dominant logic focuses on intangible and dynamic resources, emphasizing on value co-creation and social attributes [10, 11]. In its perspective, all businesses can be perceived as services businesses [12]. Compared to good-dominant logic in the era of manufacturing industry, services are people-centered instead of physical good centered [12]. The development of services poses new problems and challenges to service operation and management, with a wider range of consumer participation and more real-time requirements for decision making. Intrinsically, services system is a complex adaptive system. The entities in a services system are connected and interacted dynamically to co-create value. It is also people-centric, informationdriven, and e-oriented[7]. As a social-technical system, it is integrated with people, technologies, infrastructures as well as engineering processes[8]. The complexity of service systems increases the difficulty of services operations management and services innovation. Moreover, in the digital age, as the scale and requirements increase, the establishment, operation, and evaluation of service systems become more challenging. Privacy and security issues, system reliability issues, resistance on service innovation, etc., have appeared. As the complexity of the system increases, the differences between the systems to be modeled and their corresponding models become larger, resulting in a huge gap beyond the capabilities of conventional tools and methods, namely “Cognitive Gap” (Fig. 1.3) [16]. Therefore, many complex systems behave like Merton systems, such as stock market systems and social systems[17]. Consequently, with the development of technology and rising complexity of service systems, new philosophy and methodology are required to be applied to

Fig. 1.3 Complexity vs. intelligence: the cognitive gap [16]

6

1 Introduction

research on service systems in order to solve the potential problems and bridge the gap between real systems and artificial systems. The ACP-based parallel intelligence initialed by us since 2004 has provided a new perspective and new approach to solve complex system problems[13–15]. ACP refers to artificial systems, computational experiments, and parallel execution. Artificial systems simulate and represent actual system. Computational experiments are conducted to test, validate, analyze as well as predict for decision-making. With dual feedback and closed-loop workflow between the real and artificial systems, parallel execution of control and management is realized. Parallel intelligence generates big data to fill gaps and builds end-to-end bridges to connect different parts of two worlds, from physical to mental worlds, and vice versa[15, 17]. Based on years of research on parallel intelligence and its application in various service industries, we have established a new approach for service systems: Parallel Services

References 1. Bank, T.W.: Services, value added (% of gdp) | data. https://data.worldbank.org/indicator/NV. SRV.TOTL.ZS (2022). Accessed 19 Oct 2022 2. Bordoloi, S., Fitzsimmons, J.A., Fitzsimmons, M.J.: Service Management: Operations, Strategy, Information Technology. McGraw-Hill Education, New York (2018) 3. Fang, E., Palmatier, R.W., Steenkamp, J.B.E.: Effect of service transition strategies on firm value. J. Marketing 72(5), 1–14 (2008) 4. Kowalkowski, C., Gebauer, H., Kamp, B., Parry, G.: Servitization and deservitization: overview, concepts, and definitions. Ind Mark Manag 60, 4–10 (2017) 5. Mont, O.K.: Clarifying the concept of product–service system. J Cleaner Prod 10(3), 237–245 (2002) 6. Oliva, R., Kallenberg, R.: Managing the transition from products to services. Int J Service Ind Manag 14, 160–172 (2003) 7. Qiu, R.: Service science fundamentals. In: Service Science, pp. 92–126. Wiley, Hoboken (2014). https://doi.org/10.1002/9781118551820.ch4 8. Spohrer, J., Maglio, P.P., Bailey, J., Gruhl, D.: Steps toward a science of service systems. Computer 40(1), 71–77 (2007) 9. Vandermerwe, S., Rada, J.: Servitization of business: adding value by adding services. Eur. Manag. J. 6(4), 314–324 (1988) 10. Vargo, S.L., Lusch, R.F.: Evolving to a new dominant logic for marketing. J. Mark. 68(1), 1–17 (2004). https://doi.org/10.1509/jmkg.68.1.1.24036 11. Vargo, S.L., Lusch, R.F.: Service-dominant logic: continuing the evolution. J. Acad. Mark. Sci. 36(1), 1–10 (2008). https://doi.org/10.1007/s11747-007-0069-6 12. Vargo, S.L., Lusch, R.F.: Evolving to a new dominant logic for marketing. In: The ServiceDominant Logic of Marketing, pp. 21–46. Routledge, Milton Park (2014) 13. Wang, F.Y.: Artificial societies, computational experiments, and parallel systems a discussion on computational theory of complex social-economic systems. Fuza Xitong yu Fuzaxing Kexue(Complex Syst. Complexity Sci.) 1(4), 25–35 (2004) 14. Wang, F.Y.: Parallel system methods for management and control of complex systems. Control Decision 19, 485–489 (2004) 15. Wang, F.: Parallel philosophy and intelligent technology: dual equations and testing systems for parallel industries and smart societies. Chin. J. Intell. Sci. Technol. 3(3), 245–255 (2021)

References

7

16. Wang, F.Y., Zhang, J.J., Wang, X.: Parallel intelligence: toward lifelong and eternal developmental AI and learning in cyber-physical-social spaces. Front. Comput. Sci. 12(3), 401–405 (2018). https://doi.org/10.1007/s11704-018-7903-5 17. Wang, F.Y.: Parallel intelligence in metaverses: welcome to hanoi! IEEE Intell. Syst. 37(1), 16–20 (2022) 18. Xerox: Xerox, 2021 Annual Report (2022). https://s3.amazonaws.com/cms.ipressroom.com/ 84/files/20223/Xerox+2021+Annual+Report+Complete+--+Final.pdf

Chapter 2

Motivation: Complexity of Service in the Digital Age

2.1 Trends of Services in the Digital Age Technological advances are transforming services. With the development of emerging technologies, especially big data, artificial intelligence, Internet of Things, cognitive computing, augmented reality, virtual reality, and human–computer collaboration, traditional forms of service delivery have changed and new service business models have gradually emerged, forming new industry forms.

2.1.1 Smart Services with Smart Sensors Smart services are delivered through or to smart objects, which have the features of awareness and connectivity and allow real-time data collection, continuous communication as well as interactive feedback[13]. Personal health monitoring, family home monitoring, and industrial equipment monitoring are typical examples of smart services applications. Service providers can use the information collected through smart objects to improve their service offerings and allow customers to benefit from customized services[1]. Service providers monitor, optimize, remotely control, and autonomously adapt smart objects to enable smart services. Predictive Maintenance A typical case is the predictive maintenance of production machines. Companies can mine and analyze the data collected by sensors to enable remote monitoring and fault detection of equipment. Thus, it enables a transition from service delivery through remote service centers to a seamless omni-channel experience while minimizing service costs and downtime. At the same time, maintenance activities are planned and continuously adjusted based on machines, production processes, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_2

9

10

2 Motivation: Complexity of Service in the Digital Age

current production schedules, and other factors. More importantly, mathematical models and artificial intelligence associated with data in real time enhance the ability to personalize and deliver dynamic solutions. Suppliers can develop and continuously adjust and optimize the applied predictive maintenance services through technology. In addition, the scope of services is also adjusted in the process based on customers’ feedback, such as supplying spare parts on demand.

Rolls-Royce’s Intelligent Engine Rolls-Royce is a supplier of aircraft engines to major manufacturers such as Boeing and Airbus. Instead of selling engines directly to these manufacturers, it provides leased service time that includes maintenance, repair, and other services during the lease period. In case of engine failure, Rolls-Royce takes responsibility for repair and maintenance, and stations repair staff in major airports. This has led to an increase in service-based revenue and the establishment of strong relationships with customers through service contracts. Modern aircraft engines are equipped with multiple sensors that generate large amounts of data. For instance, an Airbus A350 aircraft has 6,000 sensors that produce 2.5 Tb of data per day. This data can be used to measure the engine’s health and performance conditions by tracking fuel flow, pressure, temperature, altitude, speed, weather, and temperature. By using real-time and historical data, airlines can make informed decisions on maintenance schedules and enhance fleet management. Predictive maintenance prevents service interruptions and groundings, and improves navigation safety. With this approach, low-cost airlines can also benefit without the need for a dedicated maintenance team[8].

Healthcare Services System Another important category of cases is the application of smart services in the healthcare services system. In chronic disease management, smart healthcare is beginning to replace the traditional hospital- and doctor-centered health management approach, which is achieved through real-time patient self-monitoring and health data feedback. Wearable or implantable smart devices, IoT technology, and health management platforms provide continuous sensing, monitoring, and feedback on patients’ physiological indicators, and medical intervention will be supplied when necessary. At the same time, patients can also use apps and platforms to self-manage their health conditions[10]. Another category is the anticipated application of smart home and home health monitoring technologies in the context of the aging trend. The term “smart home” refers to a specific type of home or

2.1 Trends of Services in the Digital Age

11

dwelling equipped with sensors and actuators that are integrated into the home’s infrastructure and designed to monitor the resident’s environment in order to improve the experience of older adults in their homes[10].

2.1.2 Retailing, Logistics, and Financial Services Based on Artificial Intelligence Technology In terms of the application of artificial intelligence in services sector, retailing, logistics, and financial industries have adopted a wealth of artificial intelligence technologies to carry out new services model, services scenes innovation, and services efficiency improvement. Retailing In retailing industry, AI-related technologies, including computer vision, intelligent voice, natural language processing, machine learning, knowledge mapping, etc., are applied to various aspects. These technologies contribute to reduce costs and increase efficiency as well as improve consumer experience. Application scenarios include precise marketing, product identification and analysis, consumer identification and analysis, intelligent operation, unmanned retail, intelligent customer service, etc. Specifically, e-commerce platforms use AI to carry out personalized recommendations, forecast sales, adjust price dynamically, and improve the efficiency of supply chain networks. Brick-and-mortar retailers use AI to locate store, analyze and arrange their shelf displays, gain insights into consumer behaviors, apply selfcheckout devices, carry out unmanned retailing, and improve the efficiency of their supply chain networks. Brands use AI to carry out precision marketing by various approaches and develop product sales mix based on market feedback, etc. Artificial intelligence and machine learning are embedded throughout the Walmart. Walmart has launched next-generation fulfillment centers. Walmart uses artificial intelligence to go from analyzing how much has been sold in stores to predicting consumer demand by analyzing cross-channel data from Google searches to Tik Tok to uncover what customers actually want to buy. During the pandemic, the tricky demand problem also turned into a tricky supply problem. An important algorithmic application for analyzing which top pieces are out of stock and what should replace it, deep learning AI, considers hundreds of variables in real time—size, type, brand, price, aggregated shopper data, personal preferences, and current inventory, among others—to determine the best next available item.

12

2 Motivation: Complexity of Service in the Digital Age

Logistics In logistics, artificial intelligence can be used for route optimization, task assignment, scheduling and transportation, automated warehousing, etc. In addition, the expected application of self-driving cars and drones in the field of logistics and transportation is still highly expected.

AI in SF Technology A Chinese logistics company named SF Technology integrates artificial intelligence into actual business scenarios, opens up each process, and further promotes information interconnection and interoperability of the entire logistics chain with business accumulation and technological innovation. Through machine learning, computer vision, operations research and global optimization, and other artificial intelligence technologies, SF realizes logistics system state perception, real-time analysis, scientific decision-making, and accurate execution and builds the “intelligent brain” of SF logistics system. SF’s AI Argus provides its transportation network with basic data such as vehicle loading rate, vehicle dispatching, capacity monitoring and energy efficiency of site personnel, providing real-time feedback and optimizing capacity. Its video structured analysis platform also detects and identifies violent sorting to reduce the probability of effectively reducing broken and lost pieces.

Financial Services In the financial services, AI has been applied in areas such as banking transactions, payments, and anti-fraud. In banking, for example, AI algorithms are used for user profiling and marketing services in retailing sector. Natural language processing is empowered in customer services robots. Computer vision is used in face recognition to verify face payment.

2.1.3 Technology Applications in Services for Emergencies In addition to changing the way of services delivery and enriching service scenarios, technology has enhanced the ability and resilience of services to cope with emergencies. From 2020, another important impact of the epidemic on services is the shrinking of offline scenarios and the explosion of online scenarios. Many service providers have enhanced their service capabilities through technology.

2.1 Trends of Services in the Digital Age

13

Experience Change in Brick-and-Mortar Stores Retailers in the post-epidemic era are accelerating the transformation from offline stores to online, building an omni-channel ecosystem, and providing better services to consumers by developing multiple consumption scenarios. An important aspect is the digital transformation of retailers, enhancing the online experience of consumers through digital technology. Retailers are improving users’ interactive and immersive experiences online through multimedia content, personalization, gamification, and other features. Examples include online shopping guides for automotive brands, VR viewing, and video advisors. Or virtual makeup fitting and virtual clothes fitting are provided with the help of AR technology. In addition, through online digital stores and offline stores linkage, users can find the nearby stores through positioning and complete the shopping by placing orders online. Another aspect is the enhancement of offline experience, such as Lululemon providing community activities for brand lovers in physical stores. Another example is “AR Red Packet” campaign launched by Uniqlo and Alipay in China. Customers use AR technology to participate in treasure hunting activities and find red packets while shopping in stores, enhancing the fun of shopping and the sense of participation. The use of contactless payment technology improves the convenience of shopping and consumer purchase through self-service payment methods such as face recognition.

E-commerce Innovation As customers’ online consumption becomes more frequent, e-commerce sites are also facing pressure on goods management and delivery efficiency. Faced with this situation, service providers have also pioneered solutions using technology, as is the case with Walmart.

Walmart’s Technology Innovation During the epidemic, online orders surged and soared. Walmart took advantage of artificial intelligence to make new innovations on both the dispensing side and the sales side. On the sales side, Walmart developed a recommendation algorithm for alternative products in out-of-stock scenarios. The AI-driven system learns each consumer’s personal preferences over time, and if an item is out of stock, the system will recommend alternatives according to the consumer’s preferences. On the distribution side, in June 2021, Walmart developed the Me@Walmart APP, with key features including picking path optimization, conversational communication, etc. The APP helps pickers (continued)

14

2 Motivation: Complexity of Service in the Digital Age

create a picking path by grouping similar orders together through routing and batching algorithms, enabling orders to be completed in a single pick to save time and improve efficiency.

Another trend is the rise of live commerce. The continued iteration of communication technology and the development of terminal technology provide technical support to live commerce. In terms of characteristics, live video is a richer content, more efficient dissemination of media. Compared with graphic propaganda, it can accomplish richer and more complex content marketing, especially some shopping experience. Compared to shelf-based e-commerce, it facilitates online shopping by stimulating interest in shopping, rather than shopping led by customer needs. Consumers may not have a clear need or even know about a product, but happen to find an item of interest while browsing the content with vivid speak and show.

Education Services Innovation In addition to the traditional consumption sector, technology has also enriched the application scenarios in educational services in the face of the epidemic. Many teaching services have been moved from offline to online. Online or blended education is becoming popular. Before the epidemic, the online learning environment was more like a store for course materials. And now this online space is gradually expanding as a way of learning, with Video and interactive media are now part of how students learn, and discussion boards allow for continued conversation and recording of ideas outside of the classroom[3]. Indian online education unicorn BYJU provides educational content by offering a combination of animation, games, and other visual inputs. BYJU has developed the Magic Workbooks product using Osmo’s artificial intelligence technology. It uses the iPad’s camera and computer vision to determine what the child is doing with the exercise book. If a child makes a correct answer, the camera captures that information and rewards it in the form of an animation or sound. The computer vision technology gives children real-time feedback as they practice in their own natural environment. Google also develops a platform for online learning. Google Teach from Home is a platform that provides online course training information and technology tools designed to help teachers and students successfully teach, learn, communicate, and collaborate remotely. In addition, with the rise of metaverse, the concept of education metaverse is coming into view. It can be perceived as the unification of cloud-based smart education, where teachers and students participate in the classroom as digital identities and interact in a virtual teaching place. The introduction of VR devices in the metaverse classroom can fully reshape the presentation of teaching content,

2.1 Trends of Services in the Digital Age

15

allowing students to be immersed in knowledge. In addition, the plasticity of the virtual space has given rise to scenarios such as virtual labs and virtual assemblies, extending the metaverse from the classroom to after-school activities.

Public Services Innovation Considering public services in China, in some scenes with high traffic flow such as traffic, security, buildings, and public places, it is necessary to analyze, identify, and control the flow of people passing by. Machine vision and remote thermometry are important supporting techniques. It effectively realizes the rapid capture and recognition of human faces, combined with no-contact infrared thermometry, providing crowd management solution during the outbreak with minimal manpower. Inside the building, through voice recognition algorithm and automatic control technology, contactless elevator service realizes voice call for elevator keys, which greatly reduces the risk of elevator contact virus transmission.

Food Services Innovation: Case of Haidilao Haidilao International Holding Ltd. is one of the largest hot pot chains in China, which has expanded its business globally. During its establishment and development, Haidilao hot pot has established a unique competitive advantage through its services. Before the meal, each restaurant has a customer waiting room containing various areas where customers can enjoy various types of services such as free fruits, snacks, drinks, board games, hand massage, manicure, and children’s play area while waiting for the meal. During the meal, the waiters will promptly respond to the multiple needs of customers by providing free supplies such as hot towels to wipe hands, aprons, and cell phone bags. Even, if customers come to the restaurant during their birthdays, the restaurant will provide birthday parties and special gifts. After the meal, waiters will provide free snacks, breath-freshening mints, and deodorant for customers to take away. Although Haidilao hot pot has excellent store service, it was also hit hard during the epidemic. Recently, Haidilao hot pot has announced the establishment of a new business module “Community Operation Division,” which intends to form a community operation model of “takeaway + community + live streaming + online shopping mall” through the integration of Haidilao’s internal and external resources. It focuses on diversified food service innovation beyond dine-in service. At present, the operation center focuses on take-out and take-away business, integrates and transforms online and offline traffic through community operation and O2O, and continues to launch new products and packages that are different from the dine-in experience.

16

2 Motivation: Complexity of Service in the Digital Age

During the epidemic, there was a clear migration of consumers’ dining behavior, from offline dining shifted to online ordering and home cooking. At the same time, consumers are increasingly diversified in the form of online food consumption. Take-out, take-away, and pre-made dishes are all under consideration. Meanwhile, they pay particular attention to the nutritional mix of meals and the quality of ingredients and are more partial to trusted new products. After concentrated observation and research, Haidilao hot pot has locked a number of valuable dining scenes, such as the one-person dining scene, housewife dining scene, etc. It will provide different types of pre-made packages and product options according to different segments of the customers. Besides the current hotpot-related dishes, it is expected to research and develop various kinds of foods, such as pre-made dishes, skewers, lo mein, beer, etc.

2.2 Complexity of Services System Services systems are complex systems. Traditionally, the complexity of any services system comes from the nature of services: simultaneity, intangibility, heterogeneity, and perishability. Intangibility refers to the fact that services are mostly behaviors rather than goods, which are difficult for consumers to perceive and evaluate in advance. Synchronicity refers to the fact that the production and consumption of services products take place simultaneously, and it is difficult to be separated in space and time. Heterogeneity refers to the fact that services products are not easily standardized and the quality is difficult to maintain stable and consistent. Nonstorability refers to the fact that most services are perishable and insufficiently tradable[2]. Additionally, the complexity of services systems comes from human involvement. Uncertainty and diversity increase due to the high level of customer contact in services production. Under the S-D logic, services are applications of expertise, skills, and experience for benefit of others[11]. As well, a services system is the configuration of people, technologies, and other resources that interact with other services systems to create mutual value[9]. Customers are important value co-creators. Networks and interactions play a central role in value creation and exchange[12]. The more knowledge-intensive and customized services are, services processes rely more on customer’s participation and input. It is hard to predict and model customers’ behaviors in value co-creation and exchange. The evolution and dynamics of the services systems also increase the complexity of services. Services processes and services evaluations are constantly changing. The first is the dynamic flow of resources. Resources flow through the services systems and are expected to be reallocated over time. It is probable that there are multiple combinations and adaptation mechanisms of services systems[5]. In addition, customers’ perception in the services process is subjective and changing. Customers perceive services through services encounters that occur during services

2.3 Challenges in the Digital Age

17

delivery and beyond[7]. Services encounters between customers and suppliers occur in different ways, face-to-face or virtual, directly or indirectly[7]. The value of a service is the total perceived value of the results co-created by suppliers and customers in a series of services encounters throughout the services lifecycle[7]. It enhances the complexity of service quality management and evaluation. Moreover, as service economy develops, customers’ expectations change gradually. Increased willingness of customers to engage in service innovation boosts the need for personalization and customization. Although as technology advances, suppliers are capable of utilizing data to understand customer needs and preferences. It is still hard to predict needs precisely under the randomness and dynamics. Services systems involve a wide range of stakeholders with different objectives. Due to the increasing complexity of objectives, decision-making is increasingly a multidisciplinary process that requires interactions and collaboration between various stakeholders with different resources. Taking the medical service system as an example, major stakeholders include patients, hospitals, employers, insurance companies, pharmaceutical firms, and government. Interactions between stakeholders in turn increase the complexity of the system. Functions of the services systems are generated through these interactions. The properties arising in the systems are hard to predict and irreducible. In conclusion, service systems are evolutionary, complex adaptive systems [5]. Services systems’ operations and management require interdisciplinary knowledge, including computational and network sciences, social sciences, management sciences, and other related fields. It is required to explore the interaction within and among services systems as well as the whole effects on systems dynamics under the service-dominant logic. A scientific method and approach is needed to address the challenges of services systems complexity.

2.3 Challenges in the Digital Age Information technology plays an important role in the formation and operation of service ecosystems and in service innovation. Resources such as information, skills, and knowledge are combined and exchanged in new ways to create value for participants in the exchange [4]. Technology advances greatly enrich the services scenarios and enhance services capabilities as well as efficiency. However, as the scale and requirements increase, the establishment, operation, and evaluation of services systems becomes more challenging. The Readiness of Technology Affects the Reliability of Its Application For new technologies with bright prospects, there is still a gap between laboratory results and actual service scenarios. Taking autonomous vehicles as an example, new technology has huge potential in logistics, travel, and various services scenarios. However, its large-scale applications are still far away, constrained by the maturity of the technology.

18

2 Motivation: Complexity of Service in the Digital Age

5G, the use of the Internet of Things and smart devices, enhances users’ convenience, experience, and participation in services systems. Nevertheless, it is inevitable that data collection, sharing, storage, management as well as transmission security and other privacy security and technical ethics occur. Taking AI as another example, since AI based on deep learning is currently highly dependent on big data, an enterprise’s database is often the cornerstone that determines whether an AI project can be successfully implemented. Data quality, information silos, data support for decision-making, infrastructure, and other factors constrain the application of AI technology. These lead to challenges in the services operations management. With technology increasing efficiency, the requirements for operational reliability also rise. In some large services systems, such as communications and logistics services, a large amount of technologies provide additional services capabilities. While technology may offer additional service capabilities, it also comes with risks such as service interruption and unpredictable problems. Therefore, in the process of services design, services integration, and services innovation, the flexibility and reliability of the systems must be considered in advance, rather than introducing new technologies directly. The Adoption of New Technologies Presents Challenges Some early adoptors utilize new technologies without clear business scenario. The vast majority of early technology applications are dotted and experimental in nature. The layout of the scale, commercialization, and operational state are lacking. There are difficulties in the integration of technology and scenarios. Sometimes there are contradictions between technological advances and adaptability of suppliers. Technological advances and market changes drive services systems upgrades. However, from an organizational point of view, technology solutions upgrades are not suitable for all institutions in terms of obstacles of corporate culture and capabilities. Thus, suppliers need to consider integration and application of new technologies into their services systems during the design or transformation stage. In addition, suppliers face barriers of consumers’ resistance to new technological innovations applied in services systems. Research analyzed the factors of customers’ resistance in IoT environment, including functional barriers, psychological barriers, and individual barriers [6]. Therefore, when applying technologies to service innovation, suppliers need to design suitable service mechanisms to address potential challenges. Overall, new technologies provide opportunities and challenges to services systems innovation, but it does not mean that some potential problems are inevitable. Thorough research, putting the emerging technology into a continuous business scenario, test and verification are required before the utilization of new technologies. It is expected that through scientific services design and experiment approaches, obstacles are able to be avoided in advance.

References

19

References 1. Beverungen, D., Müller, O., Matzner, M., Mendling, J., Vom Brocke, J.: Conceptualizing smart service systems. Electron. Markets 29(1), 7–18 (2019) 2. Bordoloi, S., Fitzsimmons, J.A., Fitzsimmons, M.J.: Service Management: Operations, Strategy, Information technology. McGraw-Hill Education, New York (2018) 3. Cowell, P.: Which 5 recent changes in university teaching should continue? | World Economic Forum. https://www.weforum.org/agenda/2021/02/covid-19-pandemic-highereducation-online-resources-students-lecturers-learning-teaching. Accessed 19 Oct 2022 4. Lusch, R.F., Nambisan, S.: Service innovation: a service-dominant logic perspective. MIS Quarterly 39(1), 155–176 (2015). https://doi.org/10.25300/MISQ/2015/39.1.07 5. Maglio, P.P., Vargo, S.L., Caswell, N., Spohrer, J.: The service system is the basic abstraction of service science. Inf. Syst. e-Business Manag. 7(4), 395–406 (2009). https://doi.org/10. 1007/s10257-008-0105-1 6. Mani, Z., Chouk, I.: Consumer resistance to innovation in services: challenges and barriers in the internet of things era. J. Product Innov. Manag. 35(5), 780–807 (2018). https://doi.org/10. 1111/jpim.12463 7. Qiu, R.: Service science fundamentals. In: Service Science, pp. 92–126. Wiley, Hoboken (2014). https://doi.org/10.1002/9781118551820.ch4 8. Rolls-Royce: Data, insights and action | Rolls-Royce. https://www.rolls-royce.com/countrysites/india/discover/2018/data-insight-action-latest.aspx#predictive-maintenance. Accessed 19 Oct 2022 9. Spohrer, J., Maglio, P.P., Bailey, J., Gruhl, D.: Steps toward a science of service systems. Computer 40(1), 71–77 (2007). https://doi.org/10.1109/MC.2007.33. 784 citations (Crossref) [2022-08-30] Conference Name: Computer 10. Tian, S., Yang, W., Le Grange, J.M., Wang, P., Huang, W., Ye, Z.: Smart healthcare: making medical care more intelligent. Global Health J. 3(3), 62–65 (2019) 11. Vargo, S.L., Lusch, R.F.: Evolving to a new dominant logic for marketing. J. Marketing 68(1), 1–17 (2004). https://doi.org/10.1509/jmkg.68.1.1.24036 12. Vargo, S.L., Lusch, R.F.: Service-dominant logic: continuing the evolution. J. Acad. Marketing Sci. 36(1), 1–10 (2008). https://doi.org/10.1007/s11747-007-0069-6 13. Wünderlich, N.V., Heinonen, K., Ostrom, A.L., Patricio, L., Sousa, R., Voss, C., Lemmink, J.G.: “Futurizing” smart service: implications for service researchers and managers. J. Serv. Marketing 29(6/7), 442–447 (2015)

Chapter 3

Opportunity: The Actual-Artificial Duality of Services

3.1 Three Worlds and Three Axial Ages The huge success of the computer Go program “AlphaGo” has awakened the world to a new thesis for the times, “AlphaGo thesis.” We present it in order to guide people into the intelligent society of the future in a healthy way[8].

AlphaGo Thesis The emergence of AlphaGo represents a parallel interaction between the virtual and the real. What used to be a Newton’s paradigm of “big laws, small data” is now shifting to a Merton’s paradigm of “big data, small laws.” In addition, the birth and evolution of AlphaGo embodies the “Small Data—Big Data—Deep Intelligence” process. AlphaGo expands the “small data” of over 800,000 human Go games into “big data” of over 70 million new Go games by playing against itself. Finally, it used artificial intelligence methods such as reinforcement learning to condense its knowledge of Go decisions into two diagrams, which it eventually used to defeat human Go masters. Later, during AlphaGo Zero’s training, the “small data” was reduced to zero, the “big data” was less, and the “small intelligence” was condensed into one diagram. And this evolutionary process took less than 3 days. With the increase in computing power, we believe that in the future this process will soon not take 3 hours, or even 3 seconds. And human players will have to become “parallel players” in the future in order to be eligible to participate in Go tournaments. At the same time, the new era is not only about new technologies, but also about new philosophies. This is the power of paradigm shift and the warning of AlphaGo thesis.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_3

21

22

3 Opportunity: The Actual-Artificial Duality of Services

Fig. 3.1 Three parallel worlds and three axial ages

What changed human society a hundred years ago was industrial technology, which was the “Old IT.” Until AlphaGo, IT stood for information technology, which was “Previous IT.” But after AlphaGo, IT must be given a new meaning of the times: intelligent technology, which is the “New IT”[1, 2]. We must use these three technologies to develop the three parallel worlds that Popper identified[3]. Extending Jaspers’ “Axial Age”[4], we are entering the Third Axial Age, as in Fig. 3.1. The First Axial Age The First Axial Age is centered on the physical world. It is the “Axial Age” proposed by Jaspers. Both the mental world and the artificial world must have had a corresponding Axial Age. The First Axial Age, centered on the physical world, took place between 800 BC and 200 BC. At this time, a large number of philosophers emerged in the Middle East, the Middle Kingdom, and India, and human selfconsciousness gradually took shape. This period saw an awakening of human nature and the advancement of philosophy. At the same time, the social form and the basic idea of “service” took shape.

3.2 The “Cognitive Gap” Between Two Worlds

23

The Second Axial Age The Second Axial Age, with its focus on the mental world, mainly comprises the Renaissance. It was a time of awakening to reason, as represented by Kopernik, Galileo, and Newton. At the same time a major breakthrough in scientific knowledge was made, various theorems of services began to emerge gradually and the services science was developed.

The Third Axial Age The Third Axial Age centered on the artificial world began 90 years ago with Goedel’s incompleteness theorem and continued with the “bounded rationality principle” of Herbert A. Simon, one of the founders of artificial intelligence. Human spirituality and intellect must awaken again, and new technological breakthroughs must be made in order to enter a new era dominated by intelligent science and technology. At the same time, services dominant logic was put forward. The services dominant logic emphasized that all economies are services economies. Since then, services science has been taken more seriously by society.

By developing the Third World, a new globalization movement can be created. At the same time, the artificial world can be created “out of nothing,” and everyone can have it, so it is naturally “positive-sum.” This proves that the third globalization movement needs to be centered on the development of an artificial world, with “positive-sum” and “win-win” as its aim.

3.2 The “Cognitive Gap” Between Two Worlds Fifty years ago, James Lighthill’s report on AI made a very pessimistic prediction about most of the core areas and approaches in AI research in the early 1970s and before, in which the research and development of AI could be summarized as “ABC.” “A” stands for advanced automation, “C” stands for computer simulation and analysis of the CNS, and “B” stands for building robots to connect the two separate worlds of “A” and “C,” or what Popper called the physical and mental worlds. We has described it as the “Cognitive Gap.” The emergence of the gap has forced us to shift our attention from modeling through Newton’s Laws to modeling through Merton’s Laws. Here, “Newton’s Laws” refer to the various traditional laws and formulas of physics, mechanics, chemistry, and biology that can describe the behavior of a system analytically and precisely. “Merton’s Laws” refer to the various “Merton’s Self-Fulfilling Prophecy

24

3 Opportunity: The Actual-Artificial Duality of Services

Laws” named after the American sociologist Merton, which can guide the behavior of a system. A statement could alter actions and therefore come true. Given the current state of the system and the conditions of control, the next state of the Newton system could theoretically be predicted accurately by solving the equation, whereas the Merton system could not. Therefore, the Newton system could directly design the appropriate control method to control the system. The Merton system, however, requires the design of “Merton’s laws” that can effectively influence or guide the behavior of the system in accordance with the goal. On this basis, it is necessary to build artificial systems, which indirectly change behavioral patterns. In turn, through the parallel interaction between the real system and the artificial system, the real system is driven to operate under the desired goal. Moving from Newton to Merton systems is a challenge for the ages, but also a key to developing the artificial world and a central issue to be faced by the reform of systems engineering processes and management.

3.3 Parallel Services as a Bridge Parallel intelligence is therefore a bridge across the cognitive gap in building intelligent systems, as Fig. 3.2. ACP-based parallel intelligence was studied in the late 1990s and formally introduced in 2004. Parallel intelligence provides parallel control and management of real systems, artificial systems, computational systems for validation analysis, and predictive systems for decision-making, through virtual–real interaction and dual feedback between real and artificial spaces. Parallel intelligence provides an effective mechanism for transforming small data into big data and then big data into task-specific deep intelligence. Parallel intelligence generates big data to fill the gap between real and artificial systems and builds endto-end bridges to connect different parts of the two worlds.

Fig. 3.2 The cognitive gap for intelligence

3.4 From CPS to CPSS

25

For services systems, as we discussed in Chap. 2, services systems are very complex. A huge gap forms between real and artificial services systems that cannot be addressed by conventional methods. A services system is as complex as a Merton system. External analysis immediately affects the internal behavior. Therefore, parallel services are the means to address the complexity of services systems.

3.4 From CPS to CPSS In the future, the Big 5G will integrate three worlds to form a five-dimensional cyber-physical-social system, or CPSS, with two parallel spaces, real and artificial. CPSS places “social” at the core, where “social” refers to human behavior and relationships. “The power of data,” “the power of computing,” “the power of algorithm,” “the power of network,” and the newly added “the power of blockchain,” these five forces combined, will drive industry to “Industry 4.0” and “Industry 5.0,” completing the two main stages of the third industrial revolution. Figure 3.3 shows the basic components and functions of CPSS. These three worlds are supported and operated by physical space and cyberspace. From CPS, which follows Newton’s law, to CPSS, which follows Merton’s law, CPSS extends its research to social network systems, focusing on the combination and coordination of human brain resources, computational resources, and physical resources. The “S” is human social systems and the activities of human societies, including human habits, thinking, economics, management, and sociology, but also artificial societies using technologies such as multi-intelligences. The science of services is integral to the activities of human social systems. Human needs are created and need to be met by corresponding acts of services. All economies are services economies, and all political activities are inseparable from services. The introduction of CPSS therefore re-emphasizes the importance of services science. Parallel services systems are also needed in artificial social systems, for data analysis and decision support of real services systems. Fig. 3.3 Framework of CPSS

26

3 Opportunity: The Actual-Artificial Duality of Services

3.5 The Future of Parallel Services Based on True DAO We have identified the theoretical basis and importance of developing parallel services science and establishing parallel services systems, which is an inevitable trend for the future. The technical requirements for the future of parallel services can be inspired by blockchain technology. We integrated blockchain, distributed autonomous organization technology, and distributed autonomous operation methods into Leibniz’s philosophy of “monad”[5], which drives intelligent technology to reach “true DAO” in the future[6]. True DAO “True DAO” means that True = Trust + Reliable + Useful + Effective/Efficient DAO = Distributed/Decentralized + Autonomous/Automated + Organizational/Operational

“Dao” in Chinese means “journey” or “monad.” It is the central concept of Chinese philosophy. The ancient Chinese philosopher and writer Lao Zi was the founder of Daoist philosophy. In his Dao De Jing, he said, “Dao gives birth to one, one to two, two to three, and three to all things.” As we have just discussed, this philosophical thinking is becoming a technological process with the help of parallel intelligence in CPSS, as well as blockchain, smart contracts, cloud computing, edge computing, and DAOs. “Dao gives birth to one” represents the physical world, “one gives birth to two” represents the acquisition or generation of small data, “two gives birth to three” represents small data generating big data, and “three begets all things” represents the formation of deep intelligence, and finally, all things are grouped into Metaverse, or “Yuan Yuan Tai Chu.” A new philosophy of intelligence emerges, and with it, the transformation of our world into a “6S” society with “6I.” The “6S” includes safe in the physical world, secure in the cyberworld, sustainable in the ecological world, sensitive to individual needs, serves for all, and smart in all. And the “6I” includes cognitive intelligence, parallel intelligence, crypto intelligence, federated intelligence, social intelligence, and ecological intelligence. To address the complexity of services systems, parallel services help us to create such societies [7].

References 1. Hu, Y., Wang, J.X.: Post Machine Age. CITIC Press, Beijing (2019) 2. Jaspers, K.: The Origin and Goal of History. Yale University Press, New Haven (1953) 3. Popper, K.: Three Worlds, the Tanner Lecture on Human Values. The University of Michigan, Ann Arbor (1978)

References

27

4. Wang, F.Y.: New IT and new axial age: the origin and goal of the future. J. Exploration and Free Views 2017(10), 23–27 (2017) 5. Wang, F.Y.: Parallel philosophy and intelligent science: From Leibniz’s monad to blockchain’s DAO. Moshi Shibie yu Rengong Zhineng/Pattern Recog. Artif. Intell. 33(12), 1055–1065 (2020) 6. Wang, F.Y.: The true and DAO in the age of blockchain intelligence. Regional Economic Review 2020(3), 6–9 (2020) 7. Wang, F.Y.: Parallel philosophy and intelligent technology: dual equations and testing systems for parallel industries and smart societies. Chinese J. Intell. Sci. Technol. 3(3), 11 (2021) 8. Wang, F.Y., Zhang, J.J., Zheng, X., Xiao, W., Yang, L.: Where does AlphaGo go: From churchturing thesis to AlphaGo thesis and beyond. Acta Automatica Sinica 3(2), 113–120 (2016)

Chapter 4

Framework of Parallel Services

4.1 Definition and Vision of Parallel Services In Chap. 3, we have clarified that in the digital age services are complex. We has identified two characteristics of complex systems. Indivisibility One is the assumption of indivisibility, which elaborates that the whole behavior of a complex system cannot be fully determined by an individual analysis of its parts. The analysis of an urban transport system, for example, involves a larger number of entities and the close connections between them. Urban transport modes include metro, bus, private car, ride-hailing service, cycling, walking, etc. Urban transport systems also involve the interaction of transport infrastructure, people, and vehicles, where people are also divided into local people and those travelling on business, and time can be divided into busy time and leisure time. If the metro system, ride-hailing service competition system, and so on are analyzed separately and then combined, it is impossible to describe the whole city’s transport system clearly.

Unknowability The second is the unknowability assumption, which states that the overall behavior of a complex system cannot be fully determined in advance on a large scale. The more complex the system the greater the uncertainty. For example, a combined online to offline (O2O) hospital with many operations cannot be modeled by a simple Poisson process arrival model or other uncertainty models.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_4

29

30

4 Framework of Parallel Services

Service systems cannot be broken down into parts for study due to their complexity, nor can comprehensive predictions be made. However, in the services domain, classification and prediction are necessary if good services are to be provided or if re-search in services science is to be conducted, a paradox. In Chap. 3, we recounted our proposal that parallel intelligence could be used to bridge the gap between the physical and mental worlds in order to deal with the complexity of the Merton system. In the services domain, we can similarly apply the concepts and methods of parallel intelligence to services in order to address the complexity of service systems, leading to the concept of “Parallel Services.” After dismantling and analysis, we define “Parallel Services” as an approach to the design and management of complex services systems based on parallel intelligence. From this definition, we can see several aspects of the properties of parallel services. Definition of Parallel Services “Parallel Services” is an approach to the design and management of complex services systems based on parallel intelligence.

First, as stated above, parallel services were created to address the complexity of services in the digital age. Thus, as a methodology, the parallel services problem addresses complex services systems. Parallel services create an artificial services system that learns the data and environment in a real services system and provides decision support to the real services system. When the system is very complex, containing multiple entities and multiple modes of interaction, traditional services analysis tools are no longer able to solve such complex problems. Building a corresponding model to solve each problem is impractical and wastes a lot of startup costs. Therefore, an artificial services system, parallel to the real services system, is the only way to solve the problem of services complexity. Second, parallel services are in essence simultaneously services engineering, services systems, and services science. As a services system, parallel services is an engineering framework for systematically solving service problems, incorporating theories from the field of services science. As such, it is a methodology for design and management, where “methodology” includes both theory and method. We proposed the ACP approach in 2004, establishing the disciplinary system of parallel intelligence, which we are now applying to the field of services. For parallel services, there are many challenges ahead. First, service observation is key to understanding customer preferences and behavior, and it requires close monitoring of customers’ daily activities. This can create privacy issues as well as data collection and processing loads and costs. Second, we need to measure customer satisfaction, and in doing so, we cannot ignore the difficulties of measuring subjective emotions such as pleasure, curiosity, and fun. Despite the difficulties ahead, the enormous potential of parallel services makes them a valuable area of research. The emergence of parallel services opens up

4.2 Framework of Parallel Services

31

many opportunities and possibilities, particularly with regard to social networks, sustainability, and well-being.

4.2 Framework of Parallel Services Figure 4.1 is the basic framework of parallel services [1]. The basic framework of parallel services consists of a real system and an artificial system. The real system has three main components: the real services system, the customer, and the environment. The artificial services system is opposed to the real services system. The artificial services system can learn from the real services system, but also understand the customer through big data analysis, the environment through IoT, and social computing, and finally provide decision support to the real services system through parallel execution. In contrast to current services management research, parallel services have several different features. First, service providers now typically collect data only when customers actually arrive and use their services. Customer demand analysis is mostly based on transactional data. But parallel services use data from a variety of sources, including click data, camera data, location data, and even social network data, which means that we are not just looking at transactional data, but also

Fig. 4.1 Framework of parallel services

32

4 Framework of Parallel Services

Fig. 4.2 Three-tier scale model of parallel services [1]

taking into account potential customer needs. Second, despite the use of techniques such as ambush marketing by many leading service providers, most are unable to dynamically track changes in consumers across different environments. The same customer may have very different service perceptions in different environments. As a result, most service providers today are unable to provide services that are best suited to consumers. Parallel services create intelligent parallel services systems that can input environmental data and consumer behavioral data for learning and simulation, thus enabling artificial services systems to provide intelligent decision support to the real services system. Another unique feature of parallel services is the flexibility of scale. We model the customer network as a three-layer scale structure when studying parallel services management, which includes person, group, and crowd as Fig. 4.2, focusing on the macroscopic emergent behavior arising from the interaction between people and the environment. As the scale increases, the complexity of the structure increases accordingly. This directs the focus to the interconnections and interactions between people, but minds, motivations, and preferences of human are very complex to study. At the person level, parallel services support us in modeling the behavior of customers and service providers in detail. The person level requires us to study a person’s thoughts, motivations, and preferences, which are related to the person’s characteristics, emotions, needs, and beliefs. At the group level, parallel services support our use of techniques such as group dynamics to understand the unique behaviors of customers. Issues in services such as the level of services satisfaction of a group and how a group makes choices in services need to be studied. At the crowd level, parallel services require the use of social computing techniques to observe and describe complex social behaviors, similar to social network analysis. In current services management methods (e.g., SERVQUAL), individual customers are usually surveyed and studied. Researchers are rarely concerned with changes in satisfaction within a customer in a group. The structure, communication, and relationships of a group can affect customer satisfaction. When a group grows in size, it eventually becomes a crowd, and its satisfaction is another key issue to

Reference

33

explore. Parallel services take into account the person, group, and crowd dimensions of services, which brings a broader application to services. For example, the study of public welfare issues requires consideration of the services satisfaction of a crowd. To support computational experimentation and parallel execution of parallel services, the design of artificial services system requires three functional components: the services need–demand model, the services network, and parallel learning and optimization. The design methodology of the artificial services system is detailed in Sect. 5.2. In addition, the parallel services framework has its own unique features compared to the ACP method in Sect. 5.1 of common complex systems. The parallel services framework is customer-centric. The parallel services framework goes beyond real services systems. The parallel services framework is very flexible and open.

Reference 1. Li, L., Lyu, C., Luo, J., Yang, S., Dai, C.: Parallel service management framework and application to railway station layout planning. Intell. Syst. IEEE 30(2), 54–61 (2015)

Chapter 5

Enabling Methodology

5.1 ACP Method As mentioned above, the ACP approach is a core methodology for solving complex system problems. Inherently, in terms of any finite resource, the overall behavior of complex systems is not able to be determined by analysis of independent parts. In addition, the overall behavior of complex systems cannot be determined in advance on large scales (such as time or large spaces) [2]. The first characteristic of complex systems represents holism that is the solving approach instead of reductionism. The second characteristic of complex systems claims that possibility is the nature of systems instead of certainty[2]. Based on the above recognition, it is necessary to establish the theory and method of complex systems research by artificial systems, computational experiments and parallel execution theory, and integrated qualitative and quantitative methods. Artificial Systems As the structure of complex systems is ambiguous and the boundaries are uncertain, traditional system analysis methods are difficult to characterize the interrelationships between components of the systems. The proposition of artificial systems creates new opportunities[1, 2]. More specifically, by utilizing multi-agent modeling and knowledge automation techniques, copy dynamic behaviors of physical systems in order to research on the underlying mechanisms [1–3]. Computational Experiments Complex systems are dynamic. An accurate and complete holistic analytic model of such systems is not practical. The dynamic change of people and society requires deepening approaches constantly. The computational experiments are the potential solutions to complex system problems[1, 2]. In detail, computational experiments are substitutes of simulations when they are not feasible to conduct, test, and validate strategies. The methods are controllable, observable, and repeatable, which satisfy the requirements of scientific methods[2– 5].

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_5

35

36

5 Enabling Methodology

Parallel Execution Complex system problems do not have an optimal solution and a unique optimal solution in a sense. Optimal solutions based on analytical model are highly relied on hypotheses. However, the hypotheses on complex systems always have a gap between real conditions. Moreover, a unique solution to complex system problem is not feasible due to feature of unpredictability. Establishing a method of parallel execution for artificial systems and real systems is effective with dynamic adaptability [1, 2]. Precisely, artificial systems and real systems operate, interact, and refer in parallel in real time, achieving control and management of complex systems[2–5]. To sum up, the ACP approach with the following steps provides solid theoretical basis for parallel services approach and applications. At first, model physical complex systems with artificial systems. Then, develop, analyze, and evaluate strategies with computational experiments. Finally, realize control and management through the constant interaction between artificial systems and real systems.

5.2 Artificial Services System Design Parallel services rely on the artificial services system for computational experimentation and parallel execution. The artificial services system design approach consists of three parts: the services need–demand model, the services network, and parallel learning and optimization.

5.2.1 The Services Need–Demand Model The services need–demand model is shown in Fig. 5.1. Services generation usually starts with people generating a need and eventually becomes the customer’s demand in services. The services need–demand model simulates the customer’s demand generation process. When service providers analyze market demand, they usually use sales data. However, such an analysis ignores unmet demand and the unrealized demand from the unobserved need. In the artificial services system of parallel services, we propose a periodic need and demand model that uses a stochastic model to represent the periodicity of customer’s need to generate demand. There are two important stages between need and demand: motivation and services selection. The motivation stage will transform the need into a potential demand based on the intensity of the demand, mood, and other factors. The services selection stage will consider customer preferences and characteristics of the services, such as price and reputation. The potential demand then becomes the actual demand for the services provider. A potential factor in the services selection stage is the recommendation between customers or between customers and the environment.

5.2 Artificial Services System Design

37

Fig. 5.1 The services need–demand model

5.2.2 The Services Network The services network in the artificial services system is the integration of service providers and customers. It has two components: the operational network from the service provider’s perspective and the customer network from the customer’s perspective, as shown in Fig. 5.2. The operational network includes service providers and their suppliers and partners outside the enterprise. Inside the enterprise, it includes services process, information system, employee, and facility. The customer network has three scales: person, group, and crowd. Person follows the need– demand model, while group and crowd define the interactions between customers. In the artificial services system, the interactions between services providers and customers help us to simulate the services process and anticipate the risk of services failure.

5.2.3 Parallel Learning and Optimization Parallel execution requires that the artificial services system has the ability to learn from the real services system in real time, which is a big challenge for existing machine learning algorithms. With the current computing power, a semiparallel approach may be a feasible solution.

38

5 Enabling Methodology

Fig. 5.2 The services network

Providing decision support for real services systems is a fundamental task of artificial services systems. Lefei Li proposes an agent-based bi-level optimization model [6], and computational experiments show that the model is an effective method to solve the optimization problem of urban bus services. In the two-layer optimization model, we can use the artificial services system in the lower layer to predict customer response and use heuristic algorithms in the upper layer to find a better solution to the services problem. With the two-layer structure, we can easily merge different OR techniques in the upper layer and use the artificial services system as an evaluation function.

5.3 Design Thinking Design thinking is an important enabling method for services. Parallel services are designed to solve service problems and therefore must require the use of design thinking. Design thinking originated in the United States and was defined and popularized as a creative way of doing things by Rolf Faste of Stanford University in the 1980s and 1990s. It was also adopted by David Kelley and Tim Brown of IDEO, which is now the World’s largest business innovation consultancy. Design thinking involves three main steps: first, fully and deeply understand the problem. Second, explore various possible solutions to the problem. After that, iterate on the solution by designing prototypes, testing, and evaluating. Finally, go to implement the problem solution.

References

39

Design thinking is a human-centered approach to innovation. In the process of exploring solutions, it pays special attention to the needs of users, such as using empathy map to think about the problem from the user’s perspective. Another key point of design thinking is prototyping and iteration. The testing process of prototypes also involves users to give feedback. In an artificial services system, we can perform lower cost prototype simulation simulations and thus iterate the solution more quickly. And because of parallel services, we have more information about the customer and the environment, so we can analyze the needs of the customer in a specific environment in a more detailed and comprehensive way. Parallel services need to be supported by design thinking, and the emergence of parallel services brings higher performance applications to design thinking.

5.4 Systems Engineering A set S of objects is said to be a system if it satisfies that it contains at least two different elements and the elements are related to each other in a certain way. The system is diverse, relevant, and holistic. The components of a system are diverse, simultaneously connected, and together form a unified whole. Systems engineering is a way to theorize and model systems thinking so that it can be mastered and applied in a standardized way to solve complex system problems, which is the definition of systems engineering. Systems engineering studies systems of systems that are often more complex. INCOSE expresses systems engineering as an interdisciplinary approach of enabling systems to be successfully implemented. Systems engineering focuses on the design and application of the whole rather than the individual parts, which involves looking at the problem from a holistic perspective, taking into account all aspects and all variables of the problem, and relating the social and technical aspects, which requires the use of holistic thinking perspectives. Systems thinking, systems science, and systems engineering are intertwined and distinct, and together they support the development, study, and practice of systems.

References 1. Wang, F.Y.: Artificial societies, computational experiments, and parallel systems a discussion on computational theory of complex social-economic systems. Fuza Xitong yu Fuzaxing Kexue (Complex Syst. Complexity Sci.) 1(4), 25–35 (2004) 2. Wang, F.Y.: Computational theory and method on complex system. China Basic Sci. 6(5), 3–10 (2004) 3. Wang, F.Y.: On the modeling, analysis, control and management of complex systems. Fuza Xitong yu Fuzaxing Kexue(Complex Syst. Complexity Sci.) 3(2), 26–34 (2006)

40

5 Enabling Methodology

4. Wang, F.Y., Zhang, J.J., Zheng, X., Wang, X., Yuan, Y., Dai, X., Zhang, J., Yang, L.: Where does AlphaGo go: from church-turing thesis to AlphaGo thesis and beyond. IEEE/CAA J. Autom. Sinica 3(2), 113–120 (2016). https://doi.org/10.1109/JAS.2016.7471613 5. Wang, X., Li, L., Yuan, Y., Ye, P., Wang, F.Y.: Acp-based social computing and parallel intelligence: societies 5.0 and beyond. CAAI Trans. Intell. Technol. 1(4), 377–393 (2016) 6. Zhang, G., Zhang, H., Li, L., Dai, C.: Agent-based simulation and optimization of urban transit system. Intell. Transpor. Syst. 15, 589–596 (2014)

Chapter 6

Enabling Technology

6.1 Decentralized Technology Distributed autonomy, free from centralized management and global data storage, represents the future development trend of parallel services and parallel intelligent systems [28, 31]. Decentralized parallel services systems are more potential and flexible than traditional centralized systems, which could maximize the operational efficiency of organizations and intelligent societies [30]. In this case, decentralized technologies play an important role in promoting the implementation of parallel services systems, including blockchain, smart contracts, cloud-edge computing, federal learning, and decentralized networks [8, 28]. This section describes some of the important technologies and their application prospects in parallel services. Blockchain and Smart Contracts Blockchain provides a basic decentralized technical architecture for parallel services systems, which implements anonymity protection, consensus protocol, data anti-tampering, and encryption functions in a distributed system [7, 17]. The structure of an example of blockchain that consists of a continuous sequence of blocks is shown in Fig. 6.1. With the peer-to-peer infrastructure, blockchain could form an effective, secure, transparent, traceable, and open network architecture that enables ACP methods and artificial intelligence technologies to gradually build programmable intelligent assets, systems, and societies [14, 30]. In addition, based on the technical architecture of blockchain, the use of smart contracts to encapsulate the complex behavior of nodes can serve as the application interface of blockchain in different complex and variable scenarios [20]. Intelligent-ware based on blockchain and smart contracts is a novel paradigm for distributed intelligence [33]. It is a package of smart contracts with rules and task- or scenario-specific computing capabilities. The core elements of blockchain are encapsulated in intelligent-ware, forming a library of intelligent-ware that can

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_6

41

42

6 Enabling Technology

Parent Block Hash

Block Header

Transaction Counter

TX

TX

Block i-1

TX

Parent Block Hash

Block Header

Transaction Counter

TX

TX

Block i

Parent Block Hash

Block Header

Transaction Counter

TX

TX

TX

TX

Block i+1

Fig. 6.1 An example of blockchain that consists of a continuous sequence of blocks [36]

be plugged in and flexibly adapted in decentralized applications, such as DAOs (Decentralized Autonomous Organizations) and ACP methods. Federated Learning and Federal Ecology With the rapid development of mobile Internet and Internet of Things, the demand for information technology in various industries is increasing, especially for information management and analysis, data storage and transmission, and data security [34]. In parallel services systems, data are usually stored in a distributed manner, with different units storing and maintaining data independently, which could lead to problems including data silos and data distrust if an effective and unified data sharing mechanism is missing, such as possible scenarios in intelligent transportation systems [14]. Fortunately, distributed storage-based edge computing, machine learning, and federated learning have provided new breakthroughs in data processing and model training in this context [6, 16, 27]. Federated learning with blockchain is possible and able to enhance mobileedge computing as a decentralized secure and private system [19] and provide parallel systems with computing abilities to deal with distributed data and security requirements. An example of FLchain design for vehicular networks is shown in Fig. 6.2 [19]. Furthermore, originated from the idea of intelligent ecosystem research [29], federated ecology [32] considers the complete cycle from data generation to data usage, services, and intelligence and constructs a mechanism for collaboration and privacy protection of each federated node, which provides a framework for secure data management in parallel services systems. 5G and Web 3.0 The development process of smart society is accompanied by changes in global networks, from transportation networks, energy networks, information networks represented by the Internet, to the Internet of Things, and to the Internet of Minds [28]. With stronger mobile broadband, high reliability, and low latency communication provided by 5G, parallel services systems could create a closer connection between the artificial and physical system. Meanwhile, the combination of rapidly developing Web 3.0 technology and blockchain [9]

6.2 Multi-Agent Simulation

43

Fig. 6.2 FLchain design for vehicular networks [22]

could promise a better open and decentralized culture movement in parallel services systems. Relying on the significant progress of network infrastructures, parallel services could bring their distributed operations and management patterns further to life.

6.2 Multi-Agent Simulation Simulation is one of the most useful design support technologies in the strategic and tactical level decision-making process, especially when the target systems are complex, dynamic, and stochastic [23]. Unlike analytical or optimization models, a simulation model is a decision-support technology but not a decision-making tool. For the complex real-world decision processes or systems, it can be an approximation that runs and provides insights into the system dynamics but not analyzes or solves. Among several widely used simulation approaches, multi-agent simulation (i.e., one of the agent-based simulation modeling methods) is well suited to modeling systems with heterogeneous, autonomous, and proactive actors, such as human-centered systems [23].

44

6 Enabling Technology

Fig. 6.3 A MAS model: the micro-level entities, their actions and interactions, and the environment

As shown in Fig. 6.3, multi-agent simulation (MAS) addresses systems that are composed of micro-level entities/agents, their actions and interactions, and the environment. The micro-level entities/agents have an autonomous and proactive behavior and interact through an environment, thus producing the overall system behavior that is observed at the macrolevel [26]. The operations of MAS rely on the following concepts: • • • •

Autonomous activity of an agent The sociability of an agent Interactions that connect the two preceding concepts The situatedness of the agents

MAS is a unique way to design, test, and study both theories and real systems. MAS has not only been widely applied in services science and service industry, such as transportation [2], healthcare [21], retailing [24], and logistics [10].

6.3 Data Fusion Techniques In the era of big data, a variety of multiple data have been generated, collected, processed, and analyzed in distinct areas. In order to obtain deeper insights and make reliable decisions, the value of data is required to be activated and mined. For example, for the purpose of management and operation of service systems, voice of customers is required to be analyzed. Customers’ behavior, preferences, needs, and requirements are expected to be mined from different sources with a wide diversity. The types of multiple data include text, audio, or video data from surveys, interviews, or social media, trajectory data from Internet of Things devices or websites, transaction data from information systems, and so on. It is clearly that these datasets have the features of semantic heterogeneity, multi-modality, variable representations, scales, etc. [35]. To solve the potential problems, both academia and industry have been denoting to the techniques on big data fusion. Consequently, the

References

45

development of data fusion methodologies and techniques provides opportunities for the realization of ACP framework-based parallel service. As following, typical techniques of data and knowledge fusion are going to be introduced. Matrix and Tensor Factorization When datasets share the same factorization model and with linear relationship, the single matrix or tensor decomposition is applicable, like singular value decomposition and non-negative matrix factorization [3, 4, 12, 15]. When dealing with heterogeneous datasets with distinct arrays, latent models, or types of uncertainty, coupled matrix factorization is adopted [1, 3]. It is common in recommendation system applications. Machine Learning (ML)-Based and Deep Learning (DL)-Based Approaches Approaches based on machine learning are usually appropriate for pre-processing or classification. Typical techniques include Support Vector Machine (SVM), KMeans, etc. They are widely used in healthcare, transportation systems, manufacturing, human–machine interaction domains, and so on. In terms of feature extraction and representation, approaches based on deep learning are more compelling. The well-known techniques comprise convolutional neural networks (CNNs), recurrent neural networks (RNNs), deep Boltzmann machine (DBM) and their variants, etc. DL-based techniques perform well on discriminative features learning. These techniques introduced above are proposed and proved to be beneficial for the fusion of text, image, audio, video, or physiological signals data [5, 11, 13, 18, 25].

References 1. Abney, S.: Bootstrapping. In: Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pp. 360–367. Association for Computational Linguistics, Philadelphia, Pennsylvania (2002). https://doi.org/10.3115/1073083.1073143 2. Balmer, M., Nagel, K., Raney, B.: Large-scale multi-agent simulations for transportation applications. In: Intelligent Transportation Systems, vol. 8, pp. 205–221. Taylor & Francis, Milton Park (2004) 3. Bao, J., Zheng, Y.: Location-based and preference-aware recommendation using sparse geosocial networking data. In: Proceedings of the 20th International Conference on Advances in Geographic Information Systems (2012) 4. Baxter, J.: A model of inductive bias learning. J. Artif. Intell. Res. 12(1), 149–198 (2000) 5. Cambria, E., Das, D., Bandyopadhyay, S., Feraco, A.: Affective computing and sentiment analysis. In: A Practical Guide to Sentiment Analysis, pp. 1–10. Springer, Berlin (2017) 6. Chamikara, M.A.P., Bertok, P., Khalil, I., Liu, D., Camtepe, S.: Privacy preserving distributed machine learning with federated learning. Comput. Commun. 171, 112–125 (2021) 7. Chen, Y., Lu, Y., Bulysheva, L., Kataev, M.Y.: Applications of blockchain in industry 4.0: A review. Inf. Syst. Front. 1–15 (2022). https://doi.org/10.1007/s10796-022-102 8. Ding, W.W., Liang, X., Hou, J., Wang, G., Yuan, Y., Li, J., Wang, F.Y.: Parallel governance for decentralized autonomous organizations enabled by blockchain and smart contracts. In: 2021 IEEE 1st International Conference on Digital Twins and Parallel Intelligence (DTPI), pp. 1–4. IEEE (2021) 9. Drakatos, P., Demetriou, E., Koumou, S., Konstantinidis, A., Zeinalipour-Yazti, D.: Triastore: A web 3.0 blockchain datastore for massive iot workloads. In: 2021 22nd IEEE International Conference on Mobile Data Management (MDM), pp. 187–192. IEEE (2021)

46

6 Enabling Technology

10. Firdausiyah, N., Taniguchi, E., Qureshi, A.: Modeling city logistics using adaptive dynamic programming based multi-agent simulation. Transpor. Res. Part E Logist. Transpor. Rev. 125, 74–96 (2019) 11. Frome, A., Corrado, G.S., Shlens, J., Bengio, S., Dean, J., Ranzato, M.A., Mikolov, T.: DeViSE: A Deep Visual-Semantic Embedding Model. In: Advances in Neural Information Processing Systems, vol. 26. Curran Associates, Red Hook (2013) 12. Golub, G.H., Reinsch, C.: Singular value decomposition and least squares solutions. Numer. Math. 14(5), 403–420 (1970). https://doi.org/10.1007/BF02163027 13. Han, S., Wang, X., Zhang, J.J., Cao, D., Wang, F.Y.: Parallel vehicular networks: a CPSS-based approach via multimodal big data in IoV. IEEE Int. Things J. 6(1), 1079–1089 (2019). https:// doi.org/10.1109/JIOT.2018.2867039. 10 citations (Crossref) [2022-06-30] Conference Name: IEEE Internet of Things Journal 14. Hou, J., Ding, W., Liang, X., Zhu, F., Yuan, Y., Wang, F.: A study on decentralized autonomous organizations based intelligent transportation system enabled by blockchain and smart contract. In: 2021 China Automation Congress (CAC), pp. 967–971. IEEE (2021) 15. Hoyer, P.O.: Non-negative matrix factorization with sparseness constraints. J. Mach. Learn. Res. 5, 1457–1469 (2004) 16. Lee, K., Lam, M., Pedarsani, R., Papailiopoulos, D., Ramchandran, K.: Speeding up distributed machine learning using codes. IEEE Trans. Inf. Theory 64(3), 1514–1529 (2017) 17. Lu, Y.: The blockchain: State-of-the-art and research challenges. J. Ind. Inf. Integr. 15, 80–90 (2019) 18. Ngiam, J., Khosla, A., Kim, M., Nam, J., Lee, H., Ng, A.Y.: Multimodal deep learning. In: ICML (2011) 19. Nguyen, D.C., Ding, M., Pham, Q.V., Pathirana, P.N., Le, L.B., Seneviratne, A., Li, J., Niyato, D., Poor, H.V.: Federated learning meets blockchain in edge computing: opportunities and challenges. IEEE Int. Things J. 8(16), 12806–12825 (2021) 20. Ouyang, L., Wang, S., Yuan, Y., Ni, X., Wang, F.: Smart contracts: architecture and research progresses. Acta Automatica Sinica 45(3), 445–457 (2019) 21. Paranjape, R., Sadanand, A.: Multi-Agent Systems for Healthcare Simulation and Modeling: Applications for System Improvement: Applications for System Improvement. IGI Global, Pennsylvania (2009) 22. Pokhrel, S.R., Choi, J.: Federated learning with blockchain for autonomous vehicles: analysis and design challenges. IEEE Trans. Commun. 68(8), 4734–4746 (2020) 23. Siebers, P.O., Aickelin, U.: Introduction to multi-agent simulation. In: Encyclopedia of Decision Making and Decision Support Technologies, pp. 554–564. IGI Global, Pennsylvania (2008) 24. Siebers, P.O., Aickelin, U., Celia, H., Clegg, C.: A multi-agent simulation of retail management practices (2008). Preprint arXiv:0803.1598 25. Srivastava, N., Salakhutdinov, R.R.: Multimodal learning with deep boltzmann machines. In: Advances in Neural Information Processing Systems, vol. 25. Curran Associates, Red Hook (2012) 26. Uhrmacher, A.M., Weyns, D.: Multi-Agent Systems: Simulation and Applications. CRC Press, Boca Raton (2009) 27. Verbraeken, J., Wolting, M., Katzy, J., Kloppenburg, J., Verbelen, T., Rellermeyer, J.S.: A survey on distributed machine learning. ACM Comput. Surveys (CSUR) 53(2), 1–33 (2020) 28. Wang, F.: Parallel philosophy and intelligent technology: dual equations and testing systems for parallel industries and smart societies. Chin. J. Intell. Sci. Technol. 3(3), 245–255 (2021) 29. Wang, F.Y., Wang, Y.: Parallel ecology for intelligent and smart cyber–physical–social systems. IEEE Trans. Comput. Soc. Syst. 7(6), 1318–1323 (2020) 30. Wang, X., Li, L., Yuan, Y., Ye, P., Wang, F.Y.: Acp-based social computing and parallel intelligence: societies 5.0 and beyond. CAAI Trans. Intell. Technol. 1(4), 377–393 (2016) 31. Wang, S., Ding, W., Li, J., Yuan, Y., Ouyang, L., Wang, F.Y.: Decentralized autonomous organizations: concept, model, and applications. IEEE Trans. Comput. Soc. Syst. 6(5), 870– 878 (2019)

References

47

32. Wang, F., Wang, Y., Chen, Y., Tian, Y., Qi, H., Wang, X., Zhang, W., Zhang, J., Yuan, Y.: Federated ecology: From federated data to federated intelligence. Chinese J. Intell. Sci. Technol. 2(4), 305 (2020) 33. Yong, Y., Liwei, O., Xiao, W., Feiyue, W.: Blockchain-based intelligent-ware: a novel paradigm for distributed artificial intelligence research. Front. Data Comput. 3(1), 1–14 (2021) 34. Zhang, J.J., Wang, F.Y., Wang, X., Xiong, G., Zhu, F., Lv, Y., Hou, J., Han, S., Yuan, Y., Lu, Q., et al.: Cyber-physical-social systems: the state of the art and perspectives. IEEE Trans. Comput. Soc. Syst. 5(3), 829–840 (2018) 35. Zheng, Y.: Methodologies for cross-domain data fusion: an overview. IEEE Trans. Big Data 1(1), 16–34 (2015). https://doi.org/10.1109/TBDATA.2015.2465959. 211 citations (Crossref) [2022-07-07] Conference Name: IEEE Transactions on Big Data 36. Zheng, Z., Xie, S., Dai, H., Chen, X., Wang, H.: An overview of blockchain technology: architecture, consensus, and future trends. In: 2017 IEEE International Congress on Big Data (BigData Congress), pp. 557–564. IEEE (2017)

Chapter 7

Research on Parallel Services

7.1 Parallel Transportation Management Systems 7.1.1 Background As the foundation of the modern economy, urban traffic service systems play a critical role in our daily life. According to Zhang et al. [15], 40% of the population spends more than one hour on the road each day. However, our transportation systems face several challenges. First, with the increase of metropolitan population and economy activities, the amount of traffic congestion increases rapidly, which reduces the satisfaction of passengers and increases the transportation cost [12]. In China, 2/3 of the 687 cities suffer from traffic congestion especially during the rush hours [18]. In the United States, traffic congestion cost people a total of $160 billion from 6.9 billion extra hours traveled in 2014 [1, 9]. Second, traffic congestion aggravates air pollution. Transportation contributes to over 90% of the total noise intensity, 60% carbon monoxide, and 50% nitrogen in the metropolitan areas [12]. This reduces the life quality of people. Third, accident risks increase with the development of our transportation systems and the increase of traffic congestion. The number of road traffic deaths approximately reaches 1.35 million in 2016 [8]. In addition, road traffic injuries are now the leading cause of death for children and young adults aged 5–29 years [8]. To solve aforementioned problems, significant effort has been made over the past two decades [16]. Over the last decades, intelligent transportation systems (ITS) have been widely studied and deployed in practice. Computer sciences, communication technology, artificial intelligence (AI), and many other emerging information sciences and engineering areas have formed the core of new ITS technology. However, this increases the complexities of the system dramatically, especially the complexities of system-level traffic control and management. In addition, the interactions between urban transportation systems and other metropolitan systems have added a few “road bump” of current research. Therefore, traditional © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_7

49

50

7 Research on Parallel Services

methodologies cannot solve the problems effectively [11]. This motivates us to use parallel services framework to fill the research gap.

7.1.2 Parallel Transportation Management Systems Following the framework of parallel services, Fig. 7.1 represents the architectures of Parallel transportation Management Systems (PtMS). Typically, the architecture of PtMS consists of five major components: actual transportation systems, artificial transportation systems (ATS), traffic operator and administrator training systems (OTSt), decision evaluation and validation systems (DynaCAS), and traffic sensing, control and management systems (aDAPTS). Parallel with the actual transportation systems, many different artificial transportation systems are developed to learn, monitor, and calculate the performance of different policies toward different scenarios (e.g., disasters and emergency) and propose the optimal operations to the actual transportation systems. Learning and Training The first step in the operation process of PtMS is learning and training. The proposal for OTSt was partially inspired by the applications of agent-based simulations for training operators in many other complex industrial scenarios. Except for the regular traffic scenarios, many emergency/worst-case scenarios are also incorporated into OTSt. Therefore, OTSt is able to cope with most situations in reality, which means the robustness and reliability.

Actual transportation systems

Artificial transportation systems

Traffic operator center

Traffic operator center

Traffic sensing, control and management systems

Decision evaluation and validation systems

Fig. 7.1 System architecture and operation processes of PtMS [11]

Traffic operator and administrator training systems

7.1 Parallel Transportation Management Systems

51

Testing and Evaluation: DynaCAS DynaCAS is developed to evaluate, optimize, and train different traffic management strategies to support the use of advanced traveler information systems, traffic management systems. DynaCAS consists of five major components including data support, experiment design, traffic simulation, decision generation, and performance evaluation. According to Fei-Yue Wang [11], DynaCAS is able to pay special attention to rule-based computational modeling of the social and behavioral aspects of people, vehicles, roads, and environments involved in transportation activities.

Control and Management: aDAPTS Agent-based Distributed and Adaptive Platforms for Transportation Systems (aDAPTS) is developed to support and provide the environments for designing, constructing, managing, and maintaining autonomous agent programs for various traffic tasks and functions [11]. Through communication networks, many agents (such as traffic control systems, road side controllers, and sensing devices) are connected to collect and share the real-time information and make real-time optimal decisions.

7.1.3 Applications The PtMS framework is verified by many successful applications in China such as Taicang [11], Binzhou [19], and Qingdao [4].

•

> PtMS in Taicang City

In Ref. [11], we conduct a field study to compare the advantages of PtMS from 2008 to 2009. Figure 7.2 shows the Experimental Station we used in Taicang city. For more detailed information about the application, we refer to Ref. [11].

Figure 7.3 indicates that our PtMS’s deployment can significantly improve the average vehicle speed driving in the selected area of study. Notably, the daily average vehicle speed increased by 11% (e.g., from 58.5 to 64.3 km/h). In addition, the results show more potentials of PtMS in the non-peak periods that may be because that the demand exceeds the upper bound of capacity. Finally, in Table 7.1, we summarized the overall difference between traffic situations on Tailiu Road before and after the application from five important

52

7 Research on Parallel Services

Fig. 7.2 PtMS experiment station in Taicang city

Fig. 7.3 Average vehicle speeds on Tailiu road: before and after PtMS’s deployment

aspects: average vehicle speed, average vehicle delay, numbers of stops, queue length, and arterial capacity. It is clear from Table 7.1 that the application achieves significant improvement of the traffic effectiveness.

7.1 Parallel Transportation Management Systems Table 7.1 Overall traffic performance improvement on Tailiu road

Evaluation indices Average vehicle speed Average vehicle delay Numbers of stops Queue length Arterial capacity

53 Changes Peak .+9.20% .−21.40% .−14.50% .−12.10% .−21%

Off peak .+15.30% .−31.50% .−28.20% .−17.30% .−26.10%

•

> PtMS in Qingdao City (PtMS-QD)

With the rapid development of the Qingdao, the congestion problem in the central city has become increasingly serious [10]. During rush hour, the roadways are congested, and traffic jams are common at many crossroads and route segments. To avoid the congestion and traffic jams, we propose a PtMS-QD as shown in Fig. 7.4.

PtMS-QD consists of three subsystems: an ATS for Qingdao (ATS-QD), a DynaCAS for Qingdao (DynaCAS-QD), and a PES for Qingdao (PES-QD). Different from classic PtMS, we redesigned PtMS-QD in parallel with existing ITS and physical transportation system. As a result, PtMS does not directly execute the control and management operations in the physical transportation system. Specifically, the PtMS collects real-time data from urban traffic information center and provides parallel decision support to the ITS. Finally, the control and management operations in real transportation system are conducted by the ITS. In PtMS-QD, ATS-QD includes not only the basic traffic simulation system, but also the artificial ITS and 3D visualization platform. The artificial ITS is parallel with the real-world ITS, i.e., it performs evaluations and experimental computations for different strategies, and provides parallel recommendations to the real-world ITS. Therefore, ATS-QD is more complete reconstruction of the transportation reality in Qingdao. The DynaCAS-QD consists of three parts: the traffic management decision evaluation platform, the traffic management decision optimization platform, and the staff learning and training platform. Notably, staff learning and training platform is a virtual environment where new staff can learn the ITS and simulate the management operations. Finally, the PES-QD consists of the decision-support system for administrators, the decision-support system for travelers (DSST), and the parallel adaptation platform. Specifically, DSST is to provide traffic advice for travelers, helping them to make better travel plans.

54

7 Research on Parallel Services

Fig. 7.4 Framework of Qingdao’s parallel traffic services system [4]

7.2 Parallel Healthcare Services 7.2.1 Background In the era of the IOT, health-IoT technologies emerge as new methods to address the challenges faced by the healthcare sector [17]. Notably, online healthcare service is a new medical service model to increase the healthcare services resources supply (i.e., the physicians) and improve the medical service quality and accessibility in rural areas. However, due to the lack of professional service design methods, the development of the online healthcare service platform (i.e., the Internet hospital) faces a few “road bump.” In this section, we will use a case of an Internet hospital

7.2 Parallel Healthcare Services

55

platform design to mainly show how to implement our parallel services framework, and demonstrate its advantages. This case is based on our work with a leading healthcare service provider, especially Parkinson’s Disease (PD)-related services, in China.

7.2.2 Design of Hybrid Services System Following our parallel services framework, the first step is to design the artificial services system that supports the computational experiments and parallel execution. According to service package theory [2], a service system is a typical human– machine system. One of the most important characteristics of service systems is the human interactivity with service systems that leads to that artificial services systems cannot take all human-related factors in the real service systems into account. Hence, it is necessary to incorporate the thoughts of psychology, economics, and philosophy into artificial intelligence. Therefore, designers are irreplaceable in service systems to better understand humans (i.e., human-centered design). This leads to the inevitable birth of the hybrid services systems as shown in the right side of Fig. 7.5 [13].

Fig. 7.5 Framework of parallel healthcare services [13]

56

7 Research on Parallel Services

•

> Human-Centered Design

The motto of the 1933 Chicago International Exposition is “science finds–industry applies–man conforms.” It was explained as “people propose, technology conforms” [7]. Finally, this leads to human-centered design [6]. According to the concept of human-centered design, designers should start from and take human capabilities and then create and use the technology to enhance and expand our abilities.

From Fig. 7.5, our human-centered design consists of customer-centered design and designers in the loop. “Customer-centered design” emphasizes customers’ requirements. “Designer in the loop” emphasizes that the designer should participate in all service design phases. Accordingly, hybrid service systems require more accurate service demand generation, parallel learning, and optimization to achieve the human-centered design.

•

? How Does Designers Work?

The designers play an important role in understanding humans. In the requirement analysis step, designers first interviewed stakeholders (20 people in total) through ladder-style interviews to get customers’ needs. Then, the whole needs set is summarized by empathy maps with different typical scenarios. In Table 7.2, we summarize the customers’ needs set.

Table 7.2 Requirements analysis High-level requirements Better treatment results

More economical treatment process

More convenient treatment process

More cheerful treatment process

Sub-requirements Make a diagnosis as soon as possible Solve psychological problems Understand the causes of health changes Inform relevant personnel in time in the case of emergency Get long-term and continuous treatment from reliable doctors Access to authoritative disease-related information Reduce the cost of treatment Reduce travel time Reduce patients family’s burden of accompanying Purchase medicine conveniently Make appointments conveniently Reduce queuing time in hospital Emotional catharsis and expression Equal and harmonious interpersonal relationship Get respect from family and society

7.2 Parallel Healthcare Services

57

Fig. 7.6 The involvement of designers and artificial service systems in the design process

In Fig. 7.6, we present the involvement of designers and artificial service systems in different design processes. Notably, “Involvement” refers to as the participation degree of the elements in the design process. From Fig. 7.6, it is clear that designers play an important role in the “understand” step. With the design progresses, the involvement of artificial service systems continues to increase.

•

? How Does Artificial Services Systems Work?

In our framework, the artificial services systems collaborate with designers that consist of a human–robot system. First, in the design phase, the artificial service systems analyze the word cloud of an online Parkinson’s Disease (PD) health community (see Fig. 7.7) to verify the results of designers’ analysis and preliminary results (i.e., the needs set). It is clear from the figure that the main concerns of PD patients are drugs, symptoms, surgery, and quality of life, which confirm the designer’s analysis results. We call this step “double confirmations.” Next, the needs set is used in the next steps. In this example, we show the operation architecture of hybrid services systems, especially the strong interactions between designers (humans) and artificial service systems (robots).

58

7 Research on Parallel Services

Fig. 7.7 Word cloud analysis of an online health community [13]

7.2.3 Computational Experiments To validate the effectiveness of the system, the simulation model established by artificial service system using AnyLogic platform, based on the resource of the hospital and the process of the treatment, is proposed as shown in Fig. 7.8a. We study a five communities setting. And there are a second-class hospital and five first-class hospitals in each community. In Fig. 7.8b, we present the daily patients number with and without (w/o) the Internet hospital. It is clear from the figure that the number of patients in the offline hospitals (first- and second-class hospitals) decreases when an Internet hospital is added. In addition, the congestion in offline hospitals is reduced. Hence, the results verify the effectiveness of the Internet hospital services as congestion is an important factor reducing the service satisfaction of customers.

7.3 Parallel Retailing Services

59

Fig. 7.8 Simulation experiments [13]. (a) Health status of citizens. (b) Daily hospital patient number w/o an internet hospital

7.2.4 Parallel Execution of the Internet Hospitals Parallel execution step is to run the hybrid service system in parallel with the real service system. The unique feature of SPSS is that the real-time results of the computational experiments can provide better decision support for the real service system. After the verification in computational experiments step, we implement our designed platform and deliver services to customers. At the same time, the artificial services systems collect and analyze customers’ satisfaction to improve the service. In Fig. 7.9, we present the customers’ attitudes to whether the effect of followup treatment with our platform is no worse than going to the hospital or not. It is clear from the figure that most of the participants support our service. Only 11% customers slightly disagree with us. According to this result, the hybrid services systems will do user test to improve the services.

7.3 Parallel Retailing Services 7.3.1 Background Online retailing has become an essential part of our daily life. In 2021, the online retailing turnover of physical goods exceeded 10 trillion RMB for the first time in China, and this number keeps increasing.1 To increase turnovers and profits, the online retailers pay more attention to the customers’ requirements and the development of online recommendation systems. In this section, we propose a parallel 1 http://www.gov.cn/xinwen/2022-01/28/content_5670892.htm.

60

7 Research on Parallel Services

dis Disa

Fig. 7.9 The customers’ attitudes toward “The effect of follow-up treatment with our platform is no worse than going to the hospital”

Fig. 7.10 The framework of the parallel retailing services [5]

retailing services (PRS) framework to provide better product recommendation as shown in Fig. 7.10.

7.3 Parallel Retailing Services

61

7.3.2 Design of the Artificial Services Systems In the right side of PRS as in Fig. 7.10, the artificial services systems set the environment and initialize the simulation with certain frequency. The design of artificial services systems in PRS requires three feature components: service decision support, service sensing, and the parallel learning and optimization.

•

> Service Sensing

The sensing part represents different approaches to observe the behaviors of customers.

Eye tracking technologies have been applied extensively in marketing and psychology research [3]. We developed an eye tracking experiment with an online B2C website. We categorize the customer into three types: item shopper, category shopper, and browsing shopper. According to our study, we are able to predict the type of the shopper based on their eye focus path with more than 85% accuracy. Online behavior can be tracked and analyzed more easily these days. On the contrary, the real-time service perception for offline service systems is still a challenging issue. Extracting crowd flow and density from video streams is a populate approach. However, the travel path and other behaviors (e.g., facial expression) of a customer might have more interesting implications about service. In our research, we tracked the shopping path of more than 200 customers in a convenience store. By analyzing the zones that they traveled, we discovered that the category layout has important impacts on their shopping behaviors. We are using that impact to model the consumer shopping decision process and optimize the category layout for the convenient store.

•

> Service Decision Support

In this case, service decision support is the customer decision model used to capture the purchase motivation of the customers.

We consider the modeling of users’ motivation as a multi-attribute decisionmaking problem. Through service sensing approaches such as eye tracking technologies and browse-path learning, we can capture the customer decision model’s critical parameters. Based on Zhang et al. [14], we model the customers’ purchase

62

7 Research on Parallel Services

motivation as follows: Mj i = P Sj i × Pi × QSj i + F Ti × I Ni + λ .

+μ

 g∈Ti ,h∈Ti

T AGgh + η





CCik

k∈Hj

(7.1)

T AGlm ,

l∈Ti ,m∈TE

where .Mj i represents the motivation of user j toward product i. Parameters .Pi , .Qi , and .Ri are the price, quality, and average rates product i, respectively. Parameter .P Sj i is the price sensitivity, and .QSj i is the quality sensitivities of customer j toward product i. Parameter .F Tj represents the extent that customer j is influenced by other customers’ rates. Parameters .Ti , .Tj , and .TE are the sets of tag of customer j , product i, and event E, respectively. .CCik is the correlation of the category of products i and k. Finally, .λ, μ, and .η are parameters critically set after several related experiments, respectively. Next, we build the artificial system with consumers, goods, and their tags. And then, any user arrive at the system will trigger the simulation as in Fig. 7.11. The recommendation list is based on the neighboring agents’ purchase motivations, which further depends on the preference over the features of the goods and other impact (e.g., environment, special event).

Fig. 7.11 Recommendation process with artificial B2C system

7.3 Parallel Retailing Services

63

7.3.3 Computational Experiments The artificial services systems develop the agent-based consumer decision model and perform simulations on the software platform of AnyLogic. The results of simulations show that the increasing trend along with the user’s income increases, with minor fluctuations in the middle. To further test the quality of the recommendation, the designer also performs a lab experiment with 22 student participants. The experiment processes as follows. First, the participants submit their purchasing bundle after surfing the mini B2C website. Then, 12 products are recommended among which six are recommended by traditional ways and other six are recommended by the ACP-based approach. Finally, the participants make their choices. We present the result in Table 7.3. We observe that the recommendations provided by the PRS are more attractive than those provided by the traditional “best-selling” approach that demonstrates the advantages of the PRS framework.

7.3.4 Extensions Augmented reality provides a brand new way to enhance service quality, through providing extra information to customer. In our lab experiments, we examined the impact of environmental information augmentation on customers’ purchase motivation. By augmenting related information about the context, the customers are more likely getting aware of the risk and initiate an action (e.g., buying heath products). There are various technical approaches, including mobile app, LED screen, etc., to implement information augmentation. Method is not a problem, the attention of people, which is becoming a scarce resource nowadays, is. Thus, what type of information to be augmented becomes a critical issue. In the framework of parallel service management, the artificial service systems will be deployed to analyze the customers’ need under certain service operation status and environmental context, in parallel with the real service system. In order to evaluate the potential benefit of reinforcing servicescape and create an information display decision structure, we develop a simple operations model as follows. Table 7.3 Lab experiment result [5] Criteria ACP-based mean Interest 4.32 Rate 4.14

Traditional mean 3.56 3.33

Estimate of difference 0.756 0.809

95% confidence interval 0.541 0.586

T value P value 5.78 5.98

. Dynamic Decision-Making Platform (DDMP)

DDMP’s tasks are to evaluate, optimize, and train the logistics management strategies, i.e., inventory strategies, logistics network plan, capacity planning, for managers under most of the possible scenarios (e.g., demand surges or plummets, disruptions in the logistics networks). Three components, decision evaluation, decision optimization, and learning and training, are designed in DDMP. The decision evaluation platform provides offline tests of logistics systems management strategies before they are implemented in real logistics systems. Next, with decision optimization platform, DDMP will output the optimal strategies to PES.

•

> Parallel Execution System (PES)

The PES is designed to link the DDMP and logistics information center. PES will select suitable strategies from DDMP to different stakeholders such as managers, workers, and customers through a parallel application platform. And it will also collect the feedback from information center to support the optimization in DDMP.

In future work, we will first perfect the PLS by observing and analyzing its implementation results over a long period of time.

References 1. Afrin, T., Yodo, N.: A survey of road traffic congestion measures towards a sustainable and resilient transportation system. Sustainability 12(11), 4660 (2020) 2. Fitzsimmons, J.A., Fitzsimmons, M.J., Bordoloi, S.: Service Management: Operations, Strategy, and Information Technology. McGraw-Hill, New York (2008) 3. Hui, S.K., Fader, P.S., Bradlow, E.T.: Path data in marketing: an integrative framework and prospectus for model building. Marketing Sci. 28(2), 320–335 (2009) 4. Kong, Q.J., Li, L., Yan, B., Lin, S., Zhu, F., Xiong, G.: Developing parallel control and management for urban traffic systems. IEEE Intell. Syst. 28(3), 66–69 (2013) 5. Lyu, C., Li, L., Pan, T.: A smart B2C e-commerce system based on ACP approach. IEEE Intell. Syst. 29(4), 102–104 (2014) 6. Norman, D.: The Design of Everyday Things: Revised and Expanded Edition. Basic books, New York (2013) 7. Norman, D.: Things that Make us Smart: Defending Human Attributes in the Age of the Machine. Diversion Books, New York (2014)

68

7 Research on Parallel Services

8. Organization, W.H., et al.: Global status report on road safety 2018: summary. Technical Report, World Health Organization (2018) 9. Pishue, B.: US Traffic Hot Spots: Measuring the Impact of Congestion in the United States (2017) 10. Sun, Q., Sun, Y., Sun, L., Li, Q., Zhao, J., Zhang, Y., He, H.: Research on traffic congestion characteristics of city business circles based on TPI data: the case of qingdao, china. Phys. A Statist. Mech. Appl. 534, 122214 (2019) 11. Wang, F.Y.: Parallel control and management for intelligent transportation systems: concepts, architectures, and applications. IEEE Trans. Intell. Transport. Syst. 11(3), 630–638 (2010) 12. Wang, F.Y., Tang, S.: A framework for artificial transportation systems: from computer simulations to computational experiments. In: Proceedings. 2005 IEEE Intelligent Transportation Systems, 2005., pp. 1130–1134. IEEE (2005) 13. Wang, R., Zhang, Y., Li, L.: Semiparallel service systems in CPSS: theory and application. IEEE Trans. Comput. Soc. Syst. (2022) 14. Zhang, T., Zhang, D.: Agent-based simulation of consumer purchase decision-making and the decoy effect. J. Business Res. 60(8), 912–922 (2007) 15. Zhang, J., Wang, F.Y., Wang, K., Lin, W.H., Xu, X., Chen, C.: Data-driven intelligent transportation systems: a survey. IEEE Trans. Intell. Transport. Syst. 12(4), 1624–1639 (2011) 16. Zhang, G., Zhang, H., Li, L., Dai, C.: Agent-based simulation and optimization of urban transit system. Intell. Transport. Syst. 15, 589–596 (2014) 17. Zhibo, P., Lirong, Z., Junzhe, T., Kao-Walter, S., Dubrova, E., Qiang, C.: Design of a terminal solution for integration of in-home health care devices and services towards the internet-ofthings. Enterprise Inf. Syst. 9(1), 86–116 (2015) 18. Zhiyan, L., Xiaoyan, Y., Rui, Z., Ying, Z.: Urban traffic congestion in China: causes and countermeasures. China Finance Econo. Rev. 4(2), 47–59 (2015) 19. Zhu, F., Lv, Y., Chen, Y., Wang, X., Xiong, G., Wang, F.Y.: Parallel transportation systems: toward iot-enabled smart urban traffic control and management. IEEE Trans. Intell. Transport. Syst. 21(10), 4063–4071 (2019)

Chapter 8

Parallel Services and Digital Twins

8.1 Introduction of Digital Twins The inception of the concept of the digital twins can be traced back to Grieves’ conception of the concept in a course on lifecycle management in 2003. His reference to the concepts of entities, virtual doubles, and data transfer is considered to be the prototype for the inception of the digital twins. In 2010, the concept of digital twins was officially introduced in a NASA report. At the time of conception, digital twins were defined as a highly integrated multi-physics, multiscale probabilistic simulation model for equipment or systems. The digital twins are able to use physical models, sensor data, and historical data to reflect the function, state, and future trends of an entity during its lifecycle [1]. The concept of the digital twins was first applied in the aerospace sector, including airframe and subsystem design simulation, real-time inspection of operational environments and capabilities, failure and fault prediction, and maintenance and health management. Following the introduction of the digital twin concept, a series of studies have been conducted in the academic community on the definition, paradigm, framework, and supporting technologies of the digital twins. There is also a richer theory on the essence and scope of the digital twins. Theories on the digital twins can be divided into two main categories. One category is divided according to design prototypes in virtual space and physical digital twins. The other category is divided according to the function of the digital twins as a design, production, and operation related digital twins [2]. Different academics and industry players have different understandings of the digital twins. There are differences in what they consider to be the content, structure, and function of the digital twins. This has led to difficulties in conducting theoretical research on the digital twins and in implementing the digital twins. Tao Fei et al. proposed a standard system for the digital twins in 2019, containing the underlying terminology, architecture, standards for applicable guidelines, etc. However, the

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_8

69

70

8 Parallel Services and Digital Twins

development of specific standards, the construction and development of platforms, and the implementation and promotion of the industry are still to be completed. Combined with the definition, the digital twin is a complete model for the lifecycle of an entity, as well as a local detailed model for building a subsystem at a certain stage. The digital twins capture and maintain digital information during the lifecycle of a product or system and use the digital information and models for simulation and decision-making. The digital twins are created early in the lifecycle during the design and manufacturing phases. As the product enters operational services, acquired sensor data, operational records, etc. are continuously collected and updated in the digital twin model and iterated. The digital twins are not only synchronized with the physical world in the virtual digital world, but also identify problems and optimize decisions through model simulation and algorithm analysis.

8.2 Parallel Services and Digital Twins Parallel services are not simply the application of the digital twins in the services domain. The concept of parallelism is much broader than the concept of digital twins. Parallel services involve artificial services systems learning from real services systems and decision support for real services systems. Digital twins, on the other hand, only build artificial models of an entity. Or rather, digital twin is one of the means of parallel services.

References 1. Glaessgen, E., Stargel, D.: The digital twin paradigm for future NASA and US Air Force vehicles (2012) 2. Grieves, M., Vickers, J.: Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems. Springer International Publishing (2017)

Chapter 9

Parallel Services Metaverses

9.1 Introduction of Metaverses 9.1.1 The Basic Concept of Metaverses Since October 2021, when Facebook CEO Mark Zuckerberg announced Meta as the company’s new branch [27], the term “Metaverse” has become a hot topic. According to Meta, metaverses will become the next revolutionary form of social connection and the Internet—where people communicate in interconnected digital spaces, interact with the real world, and even make creations that are difficult to achieve in the past [4]. In addition, many companies have plunged into the battlefield of the next generation of huge technological and industrial changes, the swarm of metaverses. In such a context, the question “what is a metaverse?” has been frequently asked. Metaverses could bring huge opportunities and diversity. What is a metaverse, how does it work, how to bring it from imagination to reality and build the future edifice of metaverses, etc. Different answers, explanations, controversies, and attempts have been given by people and companies around these questions. Game industry is considered the most likely domain for the emergence of the embryonic metaverse. As an explorer of metaverses, Roblox Corporation, a global platform bringing millions of people together in digital world through shared experiences, made a splash in 2011 with the launch of the game Minecraft [14]. The company’s 2021 annual report shows that its daily active user base has reached 45.4m and is still expanding rapidly, rising to 52.2 million in Q2 2022, up 21.% year over year [20, 22]. Part of Roblox’s acclaim stems from the fact that its business plan nearly 20 years ago foresaw the rise of metaverses [19]. In an interview with CNBC’s Jim Cramer about “Mad Money,” Roblox founder and CEO David Baszucki said,

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Li, F.-Y. Wang, Parallel Services, SpringerBriefs in Service Science, https://doi.org/10.1007/978-3-031-25333-1_9

71

72

9 Parallel Services Metaverses Our business plan for 17 years ago predicted this new category where people can come together. For the last 16 years, we’ve been innovating on this category, building an amazing community not just of players, but an amazing developer community — 2 million strong that makes everything on our platform a rich economy [19]. Our whole company is really focused on the innovation to drive and shepherd this vision of what some people call the metaverse — or human co-experience — forward [19].

In Minecraft, players are free to explore and build things in a 3D digital world. Roblox is willing to describe the future vision of metaverses as a human coexperience category, offering people a common platform to play, to socialize, and to create together [21]. Roblox offers users more of an interactive platform than a game, focusing on building an immersive world with technology that allows creators to pursue their visions freely and come together in a virtual world, such as attending graduation ceremonies and concerts and trade virtual items. As an industry giant in social media, Facebook, now Meta, puts more emphasis on the interface with the virtual world, that is, immersive virtual reality, and social connections when describing metaverses, leveraging its accumulated scale in the social media industry. In Mark Zuckerberg’s view, game is one of the first use cases for virtual reality, which does not represent metaverses, but should be a module in them [2]. At connect 2021, Zuckerberg indicated that meta should become successor to the mobile Internet, enabling interconnected digital spaces of social connection and presence [13]. Meta invests billions of dollars annually in next-generation augmented and virtual reality devices, including its Quest line of headsets, and new virtual reality social platforms such as Horizon [23]. From the company’s perspective, the metaverse envisioned by Meta encompassing socialization, entertainment, and education is believed to break the existing limitations of people’s interaction with the digital world and enhance the closeness and experience of social connection. This intention could be found in founder’s letter, 2021, by Zuckerberg [42]. The next platform will be even more immersive — an embodied Internet where you’re in the experience, not just looking at it. We call this the metaverse, and it will touch every product we build. Our mission remains the same — it’s still about bringing people together.

In addition, the views and attempts of metaverses by various industries are gradually enriched and diversified, offering us a broader reverie about metaverses. Microsoft’s acquisition of Blizzard for 68.7 billion USD is certainly important news for the gaming industry, and with this deal, Microsoft hopes to accelerate its adaptation and readiness in metaverses [28]. According to Alysa Taylor, the company’s CVP for Industry, Apps, and Data Marketing, they are focused on enabling the leap and bridge that connects the digital and physical worlds [15]. Besides, it is emphasized that Microsoft regards secure technology and trusted environments as the cornerstone of protecting metaverses and pays attention to building secure communities for trading, playing, and working in the digital world [1]. What is more, Walmart began experimenting with Roblox to establish virtual merchandise stores, music festivals, and game series to provide shoppers with new

9.1 Introduction of Metaverses

73

experiences, especially for Generation Z [18]. After the success of the possible initial form of metaverses by Roblox, more games toward metaverses debuted, such as Epic Game’s online video game Fortnite [38], released in 2017. Many Chinese companies, such as Tencent and NetEase, have also indicated that they have begun to prepare for the next generation of metaverses [12, 24]. However, with the rapid popularity of the concept of metaverses around the world comes the questions of “Trick or Treat?” and “Hope or Hype?”. We believe that the search for the origin of the development of metaverses and the trend of value changes behind it facilitate the exploration of these queries.

9.1.2 The Value Proposition Behind Metaverses In this part, we will discuss the origin of metaverses first from our perspective and then try to answer the questions about the trend essence and value connotation behind the great attention and potential of metaverses. The concept of metaverses originated in Cyberspace, which is the combination of spaces, and Cybernetics [29–32]. In 1948, Norbert Wiener presented his understanding and elaboration of cybernetics in his book Cybernetics: Or Control and Communication in the Animal and the Machine [37], which leads to a subsequent series of metaphysical research and imagination on Cyber, including cyberspaces and metaverses. True Names (1981) by American writer Vernor Vinge, and Burning Chrome (1982) [7] and Neuromancer (1984) [8] by American–Canadian writer William Gibson are among the earliest works presenting the concept of cyberspaces, which is essential to the concept of cyberpunk and metaverses later. Then in 1991, Yale computer expert David Gelernter proposed “Mirror World” [6], a vast world built by software programs. The Mirror World brings people a more vivid and in-depth experience by imitating the projected real world, where participants can roam in mirrored hospitals and schools, conduct electronic chats, and carry out activities. Soon after, the science fiction novel Snow Crash (shown in Fig. 9.1) by Stephenson [26], published in 1992, first introduced the cyberpunk literary term “Metaverse” as the portmanteau of meta and universe. It refers to a parallel digital world that is separate from the physical world but always online, where all the presence in the real world is digitally projected in this cloud-based platform. People use virtual characters in the digital world to do anything they can do in the real world, including performing content production and consumption. The depiction of the virtual world in Snow Crash is imagined and shaped in the 2018 film Ready Player One as virtual reality entertainment space called the OASIS (Ontologically Anthropocentric Sensory Immersive Simulation) for people to escape reality. From a philosophical point of view, the “meta” of metaverses represents the “metaphysical” philosophical thought, which means to stimulate different people’s understanding, imagination, and innovation of metaverse. We believe that metaverses will bring about a series of major changes in social life and production,

74

9 Parallel Services Metaverses

Fig. 9.1 Snow crash by Neal Stephenson, published in 1992 [39]

similar to the change from agricultural to industrial society, and that metaverses and parallel intelligence, among other technologies, will be the key technological and industrial support in the ongoing change from industrial to intelligent society.

•

> From “old IT” to “new IT”

We refer to these technologies as “new IT” (Intelligent Technology), and this series of changes is also the change and progress from “old IT” (Industrial Technology) to “previous IT” (Information Technology), and then to “new IT.” As we have introduced earlier, the combination of the three kinds of “IT” will develop the three worlds we face: the physical world, mainly based on “old IT” industrial technology; the psychological world, based on “previous IT” information technology; and the artificial world, based on “new IT” intelligent technology.

9.2 CPSS for Metaverses

75

•

> A New Type of Commodity: Attention and Trust

Today, the concept of metaverses is expanded in many attempts, including multiple areas such as games, socialization, and industry, as described in the previous section. From Cyberspace, Mirror Worlds to Metaverse, Digital Twins, AI and Parallel Systems, the essence of this series of trends lies in transforming Attention and Trust, which could not be commodified in the past, into a new type of commodity that can be produced in bulk and distributed on a large scale, thus achieving a more efficient and intelligent vision of development [30]. While transforming the scope of economic goods, metaverses could also expand the ways to improve social efficiency and accelerate the process of moving from the industrial to the intelligent era.

According to this, in order to describe and analyze metaverses more scientifically, we propose cyber-physical-social system (CPSS) as the nature of metaverses. The philosophical connotations, methods, and techniques of parallel intelligence and systems with CPSS can be applied to understanding and analyzing metaverses.

9.2 CPSS for Metaverses In this part, we formally introduce CPSS as the scientific intrinsic of metaverses, the basic concept, and inspiration of which come from Karl Popper’s three-world philosophy. The cognitive progress and commercial commodity revolution will also be discussed. And correspondingly, the philosophical view, methods, and technologies of parallel services systems can be applied to the description and analysis of parallel services metaverses.

9.2.1 Parallel Intelligence for Metaverses

•

> The Essence of Metaverses

As described in Sect. 9.1.2, we indicate cyber-physical-social system (CPSS) as the abstract and scientific name for metaverses. Based on Karl Popper’s three-world philosophy [32] that our universe consists of three worlds, i.e., physical, mental, and artificial world, presented in physical space and cyberspace, we have conducted a range of researches on parallel systems and ACP method in CPSS [29–32, 36, 41].

76

9 Parallel Services Metaverses

CPSS is indicated as a scientific description of the metaverse also because of two important properties: the emergence of physical space and the convergence of cyberspace [31]. These two important properties of complexity science are unified and accomplished in CPSS for decision-making and initialization. Obviously, a concrete metaverse with these two properties is a specific realization of CPSS [31]. Along with the exciting vision presented by parallel intelligence and metaverses, the technical developments in recent years including digital twins, artificial intelligence, block chain, smart contracts, robotics, etc. could better enable the description and analysis of complex CPSSs and hopefully cross James Lighthill’s “Cognitive Gap” [30]. As we mentioned in Sect. 9.1.2, parallel intelligence and metaverses could make possible the occurrence of two kinds of new commercial commodities, i.e., attention and trust. This possibility is a notable revolution. It is stated by Herbert Simon that because of the limitation of cognitive capacity, the two kinds of commodities could not be mass-produced or mass-circulated. However, metaverse, or CPSS with parallel intelligence and supporting technologies, could be a way to break the limit. Lighthill’s Cognitive Gap [30] According to Lighthill, the recognition ability of AI could be summarized by “ABC.” “A” stands for the advanced automation and physical world. “C” stands for computer-based simulation and analysis of CNS (central nervous systems) and actually the mental world. The cognitive gap between the physical world and mental world is “B,” building robots to connect “A” and “C” and establishing the bridge. Parallel intelligence for analyzing complex CPSS brings us one step closer to building a bridge as “B” of the cognitive gap, and this is one of the emphasized progress of parallel system methods and technologies.

9.2.2 The Essence of Parallel Services Metaverses

•

> The Essence of Parallel Services Metaverses

Parallel Services Metaverses, from our perspective, refer to metaverses that could be viewed as services systems. Correspondingly, the essence of parallel services metaverses is believed to be parallel services systems based on the theories of CPSS.

In the future vision of metaverses, services will provide revolutionary changes compared to the present, including immersive experiences, closer connections, and inspired innovation. Concomitantly, future service systems in metaverses will be high-evolutionary and complex, as we have described in Chap. 2. First, the

9.3 DAOs for Parallel Services Metaverses

77

prediction and analysis of customer’s motivations and behaviors would be difficult, attributed to the variation and complexity of the interaction between customers and environments in networks of metaverses. Second, service systems in metaverses would be dynamic and changing, making it necessary to constantly observe and calculate to adapt to changes. Finally, the scale of future service systems in metaverses would grow rapidly with the increase of digital level, proposing higher requirements on the storage, extraction, and computation of data and information. Meanwhile, the problem of data and privacy security, and how to utilize a large amount of low-quality data, should also be considered. Therefore, we need new frameworks that could better cope with the complexity of service systems in metaverses, that is, parallel services systems. Along with these challenges, service systems in metaverses also bring us opportunities closely associated with parallel services. As a mirror and extension of the real world, the virtual world of metaverses, or the artificial world in parallel services systems, can provide a large scale of customer and environment data, which can be utilized in computational experiments and decision-making. The technology development in AI, block chain, big data, edge and cloud computing also facilitates the progress from data to information, and then to intelligence. Furthermore, with the emphasis of social systems in CPSS, parallel services systems can better describe the social properties in metaverses, including the connections between people and between humans and machines. The framework, methods, and technologies of parallel services systems that we present in Sect. 4.2, Chaps. 5 and 6 can be applied to essentially describing and modeling service systems in metaverses, such as gaming, socialization platform, and healthcare. The cyberspace in service metaverses is modeled as artificial service system in parallel service systems, which learns from the real service system with customer and environment information and supports the decision-making and management process. As introduced in Sect. 5.1, the ACP-based parallel services methodology could be applied to the modeling, experimenting, and execution of parallel services metaverses. More specifically, artificial systems are designed to describe the Cyberspace and build the components and complex relationships in service metaverses; computational experiments are used to cope with the simulation, test, and validation of complex systems and assist in decision-making; finally, parallel execution is applied to the virtual–real interaction operations and the decision execution.

9.3 DAOs for Parallel Services Metaverses 9.3.1 “TRUE DAO” Toward Deep Intelligence The development of parallel services metaverses also relies on the support of new technologies, including DAOs, blockchain, smart contracts, Web 3.0, etc.

78

9 Parallel Services Metaverses

Web 3.0 is viewed as the next stage of Internet’s evolution, promising a new generation of decentralized social networks, which reclaims data, value, and anonymity back to users [9], and would be one of the foundational technologies of the metaverse vision [30]. In order to cope with the characteristics of distributed data storage in metaverses and the requirements for data privacy, federal ecology realizes data federalization through federal control and service federalization through federal management, technically supported by blockchain and federal learning algorithms, and could break the problem of data silos and realize group intelligence [35]. Decentralized autonomous organizations (DAOs) as well as related applications of blockchain and smart contracts play an important role in the advancement of intelligent systems for smart societies [30]. Our approach toward deep intelligence is also a “TRUE DAO.”

•

> TRUE DAO [30]

Laozi, an ancient Chinese philosopher and thinker, described in his book “Tao Te Ching” or “Dao De Jing” that “The Dao produces The One, The One produces The Two, The Two produces The Three, and The Three produces The All.” Our previous research also claims that the philosophy behind DAOs is similar to that of the term “Dao” in Chinese, which refers to the meaning of “Journey” or “Meta”. In parallel services systems or parallel services metaverses supported by DAOs, such a philosophy precisely confirms the parallel philosophy of “small data produces big data, and big data produces deep intelligence,” as shown in Fig. 9.2. “The Dao produces The One” means that our problem comes from nature and systems. “The One produces The Two” corresponds to that we need to start with “small data” and limitations. With “The Two produces The Three,” big data is established according to experience, models, and computational experiments. “The

Fig. 9.2 A TRUE DAO for intelligent systems [30]

9.3 DAOs for Parallel Services Metaverses

79

Three produces The All” represents that we can utilize intelligent methods to distill various targeted “deep intelligence” from big data. The whole process finally leads us to “Meta,” realizing the unity of reality and parallel interaction in metaverses and accelerating the process from the Industrial Age to the Intelligent Age.

In principle, DAOs are a form of organization that operates autonomously without centralized control or outside intervention, where management and operation rules are encoded on blockchain in the form of smart contracts[33]. The concept and prototype of DAOs partly come from “self-organization” phenomenon in nature. The Cyber Movement Organizations (CMOs) on the Web also consist of the virtual, weakly centralized, and autonomous characteristics of DAOs. In addition, the development of decentralized artificial intelligence and other decentralized or distributed technologies provide the technical basis for DAOs. Compared with traditional organizations, the important features of DAOs are as following, which are expected to become a new form of organization with more creativity and operational effectiveness in uncertain, diverse, and complex environments: Feature 1 Feature 2 Feature 3 Feature 4

Distributed and Decentralized Autonomous and Automated Organized and Ordered Intelligence and Tokenization

The five-tier architecture model for DAOs is shown in Fig. 9.3, which is available to be referred in order to analyze the structure of DAOs.

9.3.2 Enabling Technologies for DAOs As the core technology for the implementation of DAOs, blockchain integrates peerto-peer transmission, cryptography, consensus mechanisms, and intrinsic incentives in a decentralized system, which enables secure information sharing, value transfer, and self-management without the intervention of a trusted third party [3, 11]. The decentralization capability and consensus mechanism make it suitable for designing and managing autonomous systems or organizations. Furthermore, smart contracts [16, 34] codify the management rules of DAOs in blockchain to automate operations of the organization without a center of rights to advance or intervene, such as open-source public blockchain and smart contract development platforms. Several open-source public blockchain and smart contract platforms have achieved mature development, such as Ethereum [10]. Yong et al. [40] also point out that while blockchain is a general-purpose decentralized technology architecture for DAOs, smart contracts serve as the interface to embed the technical foundation of blockchain into scenario-specific applications. Furthermore, AI algorithms based on blockchain data could be encapsulated in

80

9 Parallel Services Metaverses

Fig. 9.3 Architectural reference model for DAOs [33]

smart contracts to form intelligent-ware that could make flexible and adaptive configurations for computation and business functions in complex environments. Several implementations of DAOs provide us with representative examples, including the DAO and Aragon, as shown below. The DAO The world’s first project of DAOs is the DAO. The DAO is a decentralized autonomous organization based on open-source code and smart contracts. The management and execution of the DAO were implemented by program code without traditional management structure or board of directors. It was established in May 2016 by a few member of the Ethereum community to act as an investor-directed venture capital fund. The DAO raised more than 150 million dollars by the sale of tokens when launching, making it the largest crowdfunding fundraising campaign of all time. However, it was hacked by a hacker on June 17, 2016, and around 60 million dollars was stolen and transferred to a satellite account. The security of the contracts and automatically executed programs of DAOs is an important problem. (The case is provided by Falkon [5] and Reiff [17].)

References

81

Aragon Aragon is a software platform that allows users to create DAOs on Ether. They provide users with infrastructures to manage clubs, corporations, nonprofits, and other organizations. The organizations created by Aragon are managed by a smart contract system called aragonOS. Aragon is a representative platform of DAOs that can be analyzed using the five-tier architecture model in Fig. 9.3[33]. Aragon itself is in the process of developing into a decentralized autonomous organization. (The case is provided by Staff [25].)

References 1. Bell, C.: The metaverse is coming. here are the cornerstones for securing it. - the official microsoft blog. https://blogs.microsoft.com/blog/2022/03/28/the-metaverse-is-coming-hereare-the-cornerstones-for-securing-it/ (2022). Accessed 22 Oct 2022 2. Carlson, K.: At SXSW, Mark Zuckerberg says metaverse is ’Holy Grail’ of social https://www.statesman.com/story/business/2022/03/16/sxsw-facebooks-markexperience. zuckerberg-says-metaverse-future-internet/7051230001/ (2022). Accessed 22 Oct 2022 3. Chen, Y., Lu, Y., Bulysheva, L., Kataev, M.Y.: Applications of blockchain in industry 4.0: A review. Inf. Syst. Front. 1–15 (2022) 4. Connect 2021: Our vision for the metaverse. https://tech.fb.com/ar-vr/2021/10/connect-2021our-vision-for-the-metaverse/. Accessed 18 Jun 2022 5. Falkon, S.: The story of the dao — its history and consequences — by samuel falkon — the startup — medium. https://medium.com/swlh/the-story-of-the-dao-its-history-andconsequences-71e6a8a551ee (2017). Accessed 22 Oct 2022 6. Gelernter, D.: Mirror Worlds: Or the Day Software Puts the Universe in a Shoebox... How It Will Happen and What It Will Mean. Oxford University Press, Oxford (1993) 7. Gibson, W.: Burning Chrome. Hachette, Paris (2017) 8. Gibson, W.: Neuromancer (1984). In: Crime and Media, pp. 86–94. Routledge, Milton Park (2019) 9. Kshetri, N.: Web 3.0 and the metaverse shaping organizations’ brand and product strategies. IT Professional 24(02), 11–15 (2022) 10. Liu, X., Tang, Z., Li, P., Guo, S., Fan, X., Zhang, J.: A graph learning based approach for identity inference in dapp platform blockchain. IEEE Trans. Emerg. Topics Comput. 10(1), 438–449 (2022). https://doi.rg/10.1109/TETC.2020.3027309 11. Lu, Y.: The blockchain: State-of-the-art and research challenges. J. Ind. Inf. Integr. 15, 80–90 (2019) 12. Ma, S.: Metaverse adds new dimensions to games - tencent. https://www.tencent.com/en-us/ articles/2201300.html (2022). Accessed 22 Oct 2022 13. Meta: Connect 2021: Our vision for the metaverse. https://tech.fb.com/ar-vr/2021/10/connect2021-our-vision-for-the-metaverse/ (2022). Accessed 22 Oct 2022 14. Minecraft - wikipedia. https://en.wikipedia.org/wiki/Minecraft (2022). Accessed 18 Oct 2022 15. Muchmore, M.: What is microsoft’s metaverse strategy? — pcmag. https://www.pcmag.com/ news/what-is-microsofts-metaverse-strategy#:~:text=Microsoft%27s%20take%20on%20the %20metaverse,be%20persistent%2C%20in%20Microsoft%27s%20view. (2022). Accessed 22 Oct 2022

82

9 Parallel Services Metaverses

16. Ouyang, L., Wang, S., Yuan, Y., Ni, X., Wang, F.: Smart contracts: Architecture and research progresses. Acta Autom. Sinica 45(3), 445–457 (2019) 17. Reiff, N.: Decentralized autonomous organization (DAO): Definition, purpose, and example. https://www.investopedia.com/tech/what-dao/#:~:text=A%20decentralized%20autonomous %20organization%20(DAO)%20is%20an%20emerging%20form%20of,best%20interest %20of%20the%20entity. (2022). Accessed 22 Oct 2022 18. Repko, M.: Walmart enters the metaverse with roblox. https://www.cnbc.com/2022/09/26/ walmart-enters-the-metaverse-with-roblox.html?&qsearchterm=metaverse (2022). Accessed 22 Oct 2022 19. Roblox founder and ceo: Our business plan 17 years ago predicted the rise of metaverse. https://www.cnbc.com/2021/11/09/roblox-founder-and-ceo-our-business-plan17-years-ago-predicted-the-rise-of-metaverse.html (2021). Accessed 18 Oct 2022 20. Roblox - roblox reports second quarter 2022 financial results. https://ir.roblox.com/news/ news-details/2022/Roblox-Reports-Second-Quarter-2022-Financial-Results/default.aspx (2022). Accessed 18 Oct 2022 21. Roblox ceo says we’re “in the middle of the metaverse right now” — finbold. https://finbold. com/roblox-ceo-says-were-in-the-middle-of-the-metaverse-right-now/ (2022). Accessed 18 Oct 2022 22. Roblox proxy-and-annual-report web-ready bookmarked - roblox corporation 2021 annual report and 2022 - studocu. https://www.studocu.com/fr/document/ecole-superieure-dessciences-economiques-et-commerciales/financial-accounting/roblox-proxy-and-annualreport-web-ready-bookmarked/30997186 (2022). Accessed 18 Oct 2022 23. Roettgers, J.: Mark zuckerberg on the metaverse, NFTs and the future of VR - protocol. https:// www.protocol.com/entertainment/mark-zuckerberg-interview-metaverse-vr (2022). Accessed 22 Oct 2022 24. Si, M.: Netease leverages gaming prowess to tap into metaverse - chinadaily.com.cn. https:// www.chinadaily.com.cn/a/202209/22/WS632c0e88a310fd2b29e79399.html (2022). Accessed 22 Oct 2022 25. Staff, C.: Aragon network dao and decentralized governance — gemini. https://www.gemini. com/cryptopedia/aragon-crypto-dao-ethereum-decentralized-government#section-aragonsblockchain-technology (2021). Accessed 22 Oct 2022 26. Stephenson, N.: Snow Crash: A Novel. Spectra (2003) 27. The facebook company is now meta — meta. https://about.fb.com/news/2021/10/facebookcompany-is-now-meta/. Accessed 18 Jun 2022 28. Thorbecke, C.: Microsoft to acquire gaming giant activision blizzard for $68.7 billion - abc news. https://abcnews.go.com/Business/microsoft-acquire-gaming-giant-activision-blizzard687-billion/story?id=82324695 (2022). Accessed 22 Oct 2022 29. Wang, F.Y.: The emergence of intelligent enterprises: From CPS to CPSS. IEEE Intell. Syst. 25(4), 85–88 (2010) 30. Wang, F.Y.: Parallel intelligence in metaverses: welcome to hanoi! IEEE Intell. Syst. 37(1), 16–20 (2022) 31. Wang, F.Y., Wang, X., Li, L., Li, L.: Steps toward parallel intelligence. IEEE/CAA J. Autom. Sinica 3(4), 345–348 (2016) 32. Wang, X., Li, L., Yuan, Y., Ye, P., Wang, F.Y.: Acp-based social computing and parallel intelligence: Societies 5.0 and beyond. CAAI Trans. Intell. Technol. 1(4), 377–393 (2016) 33. Wang, S., Ding, W., Li, J., Yuan, Y., Ouyang, L., Wang, F.Y.: Decentralized autonomous organizations: concept, model, and applications. IEEE Trans. Comput. Soc. Syst. 6(5), 870– 878 (2019) 34. Wang, S., Ouyang, L., Yuan, Y., Ni, X., Han, X., Wang, F.Y.: Blockchain-enabled smart contracts: architecture, applications, and future trends. IEEE Trans. Syst. Man Cyber. Syst. 49(11), 2266–2277 (2019) 35. Wang, F., Wang, Y., Chen, Y., Tian, Y., Qi, H., Wang, X., Zhang, W., Zhang, J., Yuan, Y.: Federated ecology: From federated data to federated intelligence. Chinese J. Intell. Sci. Technol. 2(4), 305 (2020)

References

83

36. Wen, D., Yuan, Y., Li, X.R.: Artificial societies, computational experiments, and parallel systems: an investigation on a computational theory for complex socioeconomic systems. IEEE Trans. Serv. Comput. 6(2), 177–185 (2012) 37. Wiener, N.: Cybernetics or Control and Communication in the Animal and the Machine. MIT Press, Cambridge (2019) 38. Wikipedia: Fortnite - wikipedia. https://en.wikipedia.org/wiki/Fortnite (2022). Accessed 22 Oct 2022 39. Wikipedia: Snow crash - wikipedia. https://en.wikipedia.org/wiki/Snow_Crash (2022). Accessed 22 Oct 2022 40. Yong, Y., Liwei, O., Xiao, W., Feiyue, W.: Blockchain-based intelligent-ware: a novel paradigm for distributed artificial intelligence research. Front. Data Comput. 3(1), 1–14 (2021) 41. Zhang, J.J., Wang, F.Y., Wang, X., Xiong, G., Zhu, F., Lv, Y., Hou, J., Han, S., Yuan, Y., Lu, Q., et al.: Cyber-physical-social systems: the state of the art and perspectives. IEEE Trans. Comput. Soc. Syst. 5(3), 829–840 (2018) 42. Zuckerberg, M.: Founder’s letter, 2021. https://about.fb.com/news/2021/10/founders-letter/ (2021). Accessed 22 Oct 2022