Cyber-Physical-Social System (CPSS) Architecture

Framework and Methodology

Qing Li

, Zhixiong Fang and Mengjin Qu

Department of Automation, Tsinghua University, Beijing 100084, China

Keywords: Cyber-Physical-Social System (CPSS), Architecture Framework, Reference Model.

Abstract: With the development of the Internet of things, mobile Internet, cloud computing and other emerging

information technologies, as well as the development of a new generation of artificial intelligence represented

by big data and machine learning, the cyber physical system (CPS) formed by the integration of intelligent

systems and physical systems is increasingly showing the characteristics of self-evolution and self-growth

like human society. Although human is still the most important and leading part in the technology-society

system under the current technical conditions, a new ternary integrated collaborative model is being formed

between human and intelligent system and physical system, and more complex integration logic is being

brought. This paper studies the architecture framework and methodology of ternary integrated cyber physical

social system (CPSS), which provides a framework and method for analysis and design of this complex system

of systems.

1 INTRODUCTION

With the development and revolution of industrial

technology, information technology and management

technology, the relationship between human and

nature has undergone profound changes.

As shown in Figure 1, the first industrial revolution

was marked by the realization of mechanization.

Human beings began to deal with systematic

industrial systems and assumed the main

responsibility for the mechanical operation and

control of industrial systems. The main interaction is

between human and machine. With the second

industrial revolution, electrification and automation

upgraded the interaction between people and

industrial systems to electromechanical manipulation

and control, but still the physical way. The typical

feature of the third industrial revolution is

informatization. The application of information

technology enables people to manage and control

industrial systems at more levels. Machines are

integrated with computer chips so that they can be

operated in digital way. But this doesn’t mean that

electromechanical manipulation disappeared. It

became the underlying logic. At this stage, the

https://orcid.org/0000-0002-6013-1921

concept of cyber physical system (CPS) was proposed

by the National Science Foundation (Shi, J., Wan, J.,

Yan, H., & Suo, H., 2011), but it is mainly classified

into the category of industrial systems, forming a

binary structure with people. The core of the fourth

industrial revolution is intelligence. The development

of the new generation of artificial intelligence makes

the cyber system independent from the industrial

system, and interacts with the industrial system to

realize the evolution and emergence characteristics

(Zhou, J. et al. 2018). People, cyber and industrial

systems characterized by machines constitute the

ternary structure of CPSS (Wang, F. Y., 2010).

With the development of emerging information

technology represented by computer software and

hardware, network and database technology, and

especially the Internet of things, cyber-physical

system (CPS) has become the basic architecture of

technical system (Zhou, K., Liu, T., & Zhou, L.,

2015), and is promoting the traditional technology-

management system integration architecture of

industrial system to a decentralized network

architecture. However, generally, people still regard

CPS as a whole to interact with human (Griffor, E. R.,

Greer, C., Wollman, D. A., & Burns, M. J., 2017).

206

Li, Q., Fang, Z. and Qu, M.

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology.

DOI: 10.5220/0011559600003329

In Proceedings of the 3rd International Conference on Innovative Intelligent Industrial Production and Logistics (IN4PL 2022), pages 206-217

ISBN: 978-989-758-612-5; ISSN: 2184-9285

Industrial

Technologies

Business/

Management

Information

Technology

Business/Management

Information

Technology

Industrial

Technologies

Performance Improvement

Mechanization

Era

Electrification

Era

Automation and

Information Era

Intelligent Era

Human Machine

Machine

operation

Human Machine

Electromechanical

operation

Control

Human Machine

Electromechanical

operation

control

Management

Automatic

control

human machine

Automa ti c

control

Cyber

Figure 1: The integrated development of industrial technology, management technology and information technology, and the

emergence of CPSS.

Because CPSS is developed on the basis of CPS, at

present, many studies focus on the role of human in

the whole CPSS system, the relationship and

cooperation between people and CPS, and the

possible system changes caused by artificial

intelligence. Romero et al. (2016) analysed the

change of the cooperative relationship between

people as operator and machine, and put forward the

classification method of human agent, artificial agent

and hybrid agent on the basis of intelligent agent

theory. Pacaux-Lemoine et al. (2018) put forward the

ability classification method of human agent and

artificial agent on this basis. The ability of each

intelligent agent is divided into the ability of industry

related knowledge and skills and the ability to

cooperate with other agents, but the ability of human

agent and artificial agent is not completely the same.

Sowe et al. (2016) believed that there are differences

between human and machine in motivation,

predictability and perception, and Gil et al. (2019)

also believed that human and machine do well in

different fields. In terms of the relationship between

people and systems, Calinescu et al. (2019) divided

people's roles into three types: input provider, system

contributor, and the consumer of services provided by

the system. From another perspective, Yao et al.

(2022) divided the roles played by people into three

types: "human in the loop", "human on the loop" and

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

207

"human out of the loop", which respectively represent

people who directly operate the physical system,

people who indirectly control, supervise and analyse

the physical system through the cyber system, and

people who design, plan, make decisions, experience,

use and evaluate the whole system. Bousdekis et al.

(2020) proposed the framework of HCPS on the basis

of industry 4.0, which constructs the digital twin

model of human, physical entities and how they

interact in the information world.

With the development of cloud computing, big data

and machine learning, the new generation of artificial

intelligence is changing this traditional philosophical

paradigm. In quite a number of fields, artificial

intelligence has been able to promote the independent

development and evolution of CPS, and the emerging

characteristics have already taken on some

characteristics of the human world. Zhou et al. (2019)

combed the process of information system from

assisting human to being able to learn, recognize and

make decisions independently from the perspective of

intelligent manufacturing. The new generation of

artificial intelligence makes CPS have the ability to

evolve independently of human beings.

Currently, artificial intelligence cannot completely

replace human intelligence, and human beings are

still the leading factor in the development of the

material world. However, in the interaction with the

material world, human intelligence will be more and

more influenced by the cyber system. The binary

system composed of human (consciousness) and

physics (matter) is developing into a ternary system

in which human (consciousness), artificial

intelligence (cyber) and physics (matter) interact,

which will lead to a series of new problems in system

cognition, control, decision-making and optimization.

How to analyse, design and run such a system needs

the support of architecture, methodology, system

modelling and system evaluation.

On the basis of summarizing the industrial

technology revolution, information technology

revolution and management reform, this paper

constructs the architecture of CPSS by combing the

relevant research results of complex system

architecture and analysing the meta model and

conceptual model. On the basis of the architecture

framework, this paper puts forward the main concerns

for each dimension, and describes the relationship

between these concerns, which provides a method for

the analysis and design of CPSS.

2 CPSS AND ITS META MODEL

There are studies about meta model of CPSS to show

what is it made of. Yilma et al. (2019) added social

components on the basis of CPS meta model from

Lezoche and Panetto (2018) to form CSS and CPSS

to propose the first CPSS meta model. On this basis,

Abera et al. (2020) added the relationship between

social component and physical component to form the

Physical-Social System, and showed that the CPSS

system is composed of CPSS space and CPSS objects.

He et al. (2021) proposed a meta model of cyber,

physical and human, including events, entities and

services, which determines the interaction scenarios

of cyber, physical and human through events, and

realizes the interaction among the three elements

through services.

Based on the previous works, we proposed

another meta model of CPSS, which focuses on

relationships between systems, as shown in Figure 2.

CPSS has three basic units/subsystems:

 Social system is a network of relationships

formed by people based on their subjective

consciousness. Because people have the

intelligence ability of perception, cognition

and decision-making, they can act on their

will to recognize and transform the physical

world. A social system can be made of social

subsystems.

 Physical system usually refers to the

physical world corresponding to human

society, which can be perceived, recognized

and transformed by people. The physical

system is usually related to the material

world, and will be counteracted by human’s

activities. a physical system can be made of

physical subsystems.

 With the development of automation

technology and artificial intelligence

technology, human beings gradually endow

their perception, cognition, control and

decision-making abilities to physical

systems, and then cyber systems become

independent systems. With the development

of information technology, cyber systems

have more and more abilities of perception,

cognition and decision-making, which can

act and react with the physical world

independently, and realize self-evolution

and self-growth without human participation

in this process. With the development of

artificial intelligence technology, the

independence of cyber system becomes

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

208

more and more obvious. A cyber system can

be made of cyber subsystems.

Physical system, social system and cyber system

interact with each other, resulting in three binary

systems:

 Human-machine system is an interactive

system produced by the interaction of social

system and physical system. It embodies the

concept of the original dualistic philosophy

of material consciousness. Human-machine

system is a subject in the field of traditional

industrial engineering. Human factor

engineering is one of the important branches,

which supports the development of

industrial mode of mass production in the

era of traditional mechanization. In smart

manufacturing, the system that integrates the

functions of an operator (or a group of

operators) and a machine (or a group of

machines) is usually called man-machine

system. This term can also be used to

emphasize the dynamic evolutionary system

formed by the interaction and reaction

between people and the external world.

 Cyber-physical system is a complex

engineering system, which connects the

cyber space with the physical world through

a network of interconnected computing

elements such as sensors, actuators and

computing processing units. These systems

are highly automated, intelligent and

collaborative. CPS is the core concept of

German industry 4.0 and a result of the

interconnection of everything formed by the

development of the Internet of things.

According to Lee et al. (2015), CPS

implementation is divided into 5 levels:

Smart Connection Level, Data-to-

Information Conversion Level, Cyber Level,

Cognition Level and Configuration level,

and each level defines relevant

characteristics and attributes.

 Cyber-social system is usually considered as

a social network that connects people

through the Internet. With the higher and

higher intelligence level of cyber system, the

relationship between human and cyber

system is evolving rapidly. The development

of artificial intelligence has brought new

problems and challenges in the interaction

between human intelligence and artificial

intelligence.

Although many scholars have begun to study the

relevant architecture, model and technical issues of

CPSS, there is little research on the positioning and

complex interrelationship of cyber, physical and

human, especially the evolution of the system caused

the interrelationship. There are still a lot of

Cyber-Physical-Social System /

System of Systems

Social Subsystem / System

Cyber Subsystem / System

Physical Subsystem / System

Cyber-Physical System

Man-Machine System

Cyber-Social System

1..*

1..* 1..*

1..*

Figure 2: Meta model of Cyber-Physical-Social System.

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

209

theoretical and technical problems in the interaction

between human intelligence and artificial intelligence.

For example, after AlphaGo defeated Lee SeDol, it is

unclear what role artificial intelligence plays in

improving human Go Game skills; in automatic

driving, there are still a lot of ethical and technical

problems in the relationship between people's driving

willingness and automatic driving algorithm.

The CPSS, which is composed of man-machine

system, CPS and CSS, presents some new

characteristics and new theoretical and technical

problems:

 It leads to the change of the theoretical

framework of philosophy and the

transformation from the dualistic philosophy

of material consciousness to the ternary

philosophy of material, consciousness and

cyber. And the cyber system has the ability

to act and react independently with the

material world.

 The balance of the relationship between

human and the cyber system is deflecting,

and people's dominant position is weakening.

In many scenarios, for most people, they are

passively accepting the command and

scheduling of the intelligent system. Human

society has gradually formed a situation that

a small number of professionals research

and develop artificial intelligence

algorithms, while most people are served

and dominated by artificial intelligence.

 As cyber plays an increasingly independent

role in decision-making, there is no

complete solution to determine its legal

status. For example, there is no consensus on

the legal status of auto drive system in traffic

accidents

 Since there are still a lot of ambiguous

relationships in CPSS, it is a challenge for

the theory and method of system

engineering to ensure the completeness and

safety of system design during the design of

CPSS.

In order to fully understand the main

characteristics of CPSS and provide a framework and

tools for the analysis and design of CPSS, it is

necessary to build the architecture framework of

CPSS.

3 CPSS ARCHITECTURE

FRAMEWORK

CONSTRUCTION

ISO/IEC/IEEE 42010 (2011) reveals the basic logic

of building a complex system architecture description

This paper builds the architecture framework of CPSS

based on the basic idea of the standard, and the

conceptual model is shown in Figure 3.

Cyber-Physical-Social System

Layer

Level

View

Lifecycle

Stakeholder

Concerns

Viewpoints

Model Kinds

1..*

▲ has interests in

1..*

▼has

1..*

▲frames

1..*

Architecture

Architecture Description

Views

Model

▲ expresses

1..*

exhibits

governs

1 1..*

governs

1..*

▲ addresses

Business

Functional

Information

Communication

Integration

Asset

4 - Business Planning & Logistics

3 - Manufacturing Operations &

Control

2 - Monitoring. Supervisory.Control

and Automated Control of the

Production Process

1 - Sensing the Production Process,

Manipulating the Production

Process

0 - The Physcial Production Process

5 - Group of Enterprises

Performance Aspect

Business Logic Aspect

Structural Aspect

Application / Functional / Service

Aspect

Infrastructure Aspect

Security Aspect

▲ ▲

1..*

identifies

▲

identifies▲

1..*

identifies

▲

Figure 3: Concept model of CPSS architecture.

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

210

As shown in Figure 3, the right part is from ISO

42010, which reflects the construction logic of the

architecture framework description. The architecture

object is specified to CPSS. The construction of CPSS

architecture is to start from the stakeholders related to

CPSS. By analysing their concerns about CPSS, form

the viewpoints and views of CPSS, and then build the

description of CPSS architecture framework.

For a typical CPSS such as an intelligent

manufacturing system, based on Reference

Architecture Model for Industrial 4.0 (RAMI 4.0)

(Hankel et al. 2015) and Industrial Internet Reference

Architecture (IIRA) (Lin et al. 2015), the main

viewpoints include layer, view, level and lifecycle.

Layers defined in RAMI 4.0 include business,

functional, information, communication, integration,

and asset. Based on the definition of ISA95 / ISO

62264 (ANSI/ISA-95, 2000), the level of enterprise

and control system integration design includes 0-4

levels. In order to describe the systematic

characteristics of the systems presented by enterprise

groups and networked enterprises, we have added a

fifth level. In addition, based on the idea of Federal

Enterprise Architecture Framework (FEAF) 2.0

(OMB, 2013), the modelling analysis and model

construction of CPSS can be based on the derived

structure from performance, business, structure,

application, infrastructure to security.

For integrated manufacturing systems, some

architectural frameworks have long focused on the

ternary relationship between people, machines /

facilities and information. For example, Purdue

enterprise reference architecture (PERA) (Williams,

1998) includes information architecture,

manufacturing architecture as well as human and

organizational architecture. As shown in the upper

left of Figure 4, these three systems can be extended

to correspond to cyber, physical, and social systems

respectively. The idea of PERA is that every time the

system changes from the current situation to the target,

it needs to change the physical system, information

system and organizational structure. As shown in the

left middle of Figure 4, the manufacturing system

supported by IMPACS (Integrated Manufacturing

Planning and Control System) (Banerjee, 1997) is

also composed of information technology,

manufacturing technology and organization

description.

INFORMA-

TIONAL

DECISIONAL

PHYSICAL

FUNCTIONAL

ANALYSIS

DESIGN

TECHNICAL

DESIGN

DEVELOP-

MENT

View Domain

INFORMATION

TECHNOLOGY

MANUFACTURING

TECHNOLOGY

ORGANIZATION

DESCRIPTION

Life cycle

CONCEP

-TUAL

STRUC-

TUAL

STRUC-

TUAL

REALISA-

TIONAL

Abstraction

level

III

CONCEPT

PHASE

DEFIN ITION

PHASE

DISTR IBUTION

FUNCTIONS

DISTR IBUTION

FUNCTIONS

FUNCTIONAL

DES IGN

PHASE

SPECIFICATION

FUNCTIONAL

DES IGN

FUNCTIONAL

DES IGN

PHASE

SPECIFICATION

FUNCTIONAL

DES IGN

SPECIFICATION

FUNCTIONAL

DES IGN

DETAILED

DES IGN

PHASE

DETAILED

PHYSICAL

DES IGN

DETAILED

DES IGN

PHASE

DETAILED

PHYSICAL

DES IGN

CONSTRUCTION

AND

INSTALLATION

PHASE

(MANIFESTATION)

ENTERPRISE

READY

FOR

OPERA TIONS

CONSTRUCTION

AND

INSTALLATION

PHASE

(MANIFESTATION)

ENTERPRISE

READY

FOR

OPERA TIONS

OPE RATIO NS

AND

MAINTENANCE

PHASE

OBSOLESCENCE

THE ENTERPRISE

OPE RATIO NS

AND

MAINTENANCE

PHASE

OBSOLESCENCE

THE ENTERPRISE

RENOVATION

DISPO SAL

PLANNING

PHASE

RENOVATION

DISPO SAL

DECISION

RENOVATION

DISPO SAL

PLANNING

PHASE

RENOVATION

DISPO SAL

DECISION

ENTERPRISE

DISSO LUTION

PHASE

ASSET DISPOSAL

ENTERP RISE

DISSOLUTION

ENTERPRISE

DISSO LUTION

PHASE

ASSET DISPOSAL

ENTERP RISE

DISSOLUTION

ASSET DISPOSAL

ENTERP RISE

DISSOLUTION

POLI CIES

USE CASES

NETWORKS

REQUIREMENTS

POLI CIES

USE CASES

NETWORKS

POLI CIES

USE CASES

NETWORKS

REQUIREMENTS

EXTENT OF AUTOMATION

1312

161514

191817

222120

252423

26 27

1312

161514

191817

222120

252423

26 27

1312

161514

191817 191817

222120 222120

252423 252423

26 27

IDENTIFICATION

PHASE

(Initial Idea)

IDENTIFICATION

PHASE

(Initial Idea)

(Mission, Vision,

Values)

Performance Reference Model

Business Reference Model

Data Reference Model

Application Reference Model

Infrastructure Reference Model

Security Reference Model

Cyber

Performance

Behavior

Structure

Application

Infrastructure

Performance

Behavior

Structure

Application

Infrastructure

CPSS

HMS

PERA

IMPACS

FEAF

Views

Security

Performance

Behavior

Structure

Application

Infrastructure

Security

Figure 4: CPSS architecture framework.

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

211

As shown in the lower left of Figure 4, FEAF 2.0

constructs the hierarchical structure of system design

and implementation. Performance, business and data

constitute the business architecture of the system

design: the strategic objectives stipulate the business

logic, and the business logic exports the required data

structure. Application, infrastructure and security

constitute the technical architecture of the system.

Because security needs to be considered at all levels

and stages, application and infrastructure constitute

the core of the technical architecture of the system.

Based on the above analysis, the architecture

framework of CPSS is shown on the right side of

Figure 4. The description of the architecture

framework is similar to a Venn diagram, which

highlights the new interactive system formed after the

interaction of cyber system, physical system and

social system. This idea comes from the meta model,

PERA and IMPACS. Based on the changes of

business system and technical system caused by the

interaction of these systems, the CPSS system is

analysed and designed. Different views in one system

have relationships of dependence. And the same view

in different systems are corresponding.

Following, we will analyse the concepts and logic

of CPSS architecture from the perspective of system

modelling:

 Performance view: Performance requirements

are the basis for the construction of the whole

system. Requirements come from many aspects,

including laws and regulations, standards and

specifications, strategic objectives, and the goals

and needs of stakeholders such as customers.

After decomposition, the performance

requirements of CPSS can be divided into two

parts: one is the professional ability of

individuals, and the other is the ability of

individuals to cooperate with other individuals.

In the social system, as an independent

individual, the performance requirements are

reflected in people's knowledge, skills, physical

strength and other abilities, as well as the ability

to exchange information, make joint decisions

and cooperate with others. However, the social

system has a hierarchical structure, and many

people form an organization. At this time, the

performance requirements among people have

become the performance requirements within

the organization, and the communication and

cooperation between organizations have new

performance requirements. This is also the same

in other systems. It should be noted that the

communication and cooperation capabilities

between people and physical objects, physical

objects and cyber objects, people and objects,

need to be considered in the composite system,

such as the ability of people to understand and

use machines and software, the ability of

machines to understand and execute the

instructions issued by people and information

systems, and the ability of information systems

to monitor and analyse the activities of people

and machines. In each subsystem, the

performance requirements of the top-level

subsystems can be obtained through continuous

decomposition. The performance of the

subsystem depends not only on the performance

of each individual, but also on the way of

cooperation between individuals, that is,

workflow and organizational structure.

Therefore, the performance model also puts

forward performance requirements for business

logic and organizational structure.

 Behaviour View：The behaviour view builds a

series of activity processes to meet the

requirements of the performance view in design.

The behaviour view can be classified into

different layers. For each activity process, some

measurement indicators are proposed to judge

whether it meets the performance requirements.

In addition to defining the whole activity

process from triggering, running to ending, the

behaviour layer also needs to define the roles

and components that undertake some activities

and functions in the activity process, and define

the performance requirements of these roles and

components, and then feed back to the

performance model. Because different

organizational departments, different kinds of

physical components and information

components may be involved in the whole

activity process, in order to improve the

performance of the process and meet the

performance requirements, it is often necessary

to put forward certain requirements for the

structural relationship between people and

organizations, the location relationship of

physical components in the physical

environment, and the data flow relationship

between information components.

 Structure View：The structure view builds the

personnel organization structure, physical

structure and cyber structure to support the

mechanism embodied in the behaviour view.

Excellent personnel organization structure

needs to ensure clear division of functions,

responsibilities and authorities of personnel

organization, full play of individual abilities and

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

212

efficient communication and cooperation of

personnel organization. The placement, splicing

and assembly of physical components in the

physical environment should meet the

constraints of the actual spatial structure and the

business logic timing relationship requirements

of the behaviour layer. The network structure

should facilitate the information exchange

between personnel and physical components,

and the data structure should make the data

storage cost low and easy to query.

 Application view: The application view focuses

on the implementation of system activities. The

application view aims at the behaviour view of

the system, describes the specific technical

implementation and participants of the activity

process, and describes the dynamic structure of

physical components, information components

and personnel organization in the process. In the

physical, cyber and social world, application

views are often embodied as production systems

and products, software application services,

business process and workflow management

systems.

 Infrastructure view: The infrastructure view

focuses on the basic support of the system. The

infrastructure view aims at the performance

view, behaviour view, structure view and

application view, and builds the basic support

service capability model of the system to meet

the performance requirements, behaviour

operation requirements, resource and function

structure requirements and application

construction requirements. Infrastructure is

often embodied in the physical, cyber and social

world as energy system, transportation system,

communication network, data resources, and

human resources.

 Security view: Security is the core concern of

CPSS. Security view is closely related to

performance, behaviour, structure, application

and infrastructure. According to the principles

of FEAF 2.0, the modelling and analysis of

security view has become a key field of system

design.

By transforming Figure 4, we get the description of

CPSS architecture framework as shown in Figure 5.

Different from Figure 4, figure 5 highlights two new

fields of CPSS, cyber–social system and cyber–

physical–social system, which meet new theoretical

and technical problems. On the other hand, it also

points out other dimensions in the architecture: layer,

level and lifecycle.

Performance Reference Model

Behavior Reference Model

Structure Reference Model

Application Reference Model

Infrastructure Reference Model

Performance Reference Model

Behavior Reference Model

Structure Reference Model

Application Reference Model

Infrastructure Reference Model

Social / Human System

Performance Reference Model

Behavior Reference Model

Structure Reference Model

Application Reference Model

Infrastructure Reference Model

Human-Machine System

Performance Reference Model

Behavior Reference Model

Structure Reference Model

Application Reference Model

Infrastructure Reference Model

Physical System

Cyber – Physical – Social System

Performance Reference Model

Behavior Reference Model

Structure Reference Model

Application Reference Model

Infrastructure Reference Model

Cyber-Physical System Cyber System

Social / Human – Cyber System

Technical Security

Architecture

Business Security

Architecture

Step Generation

Layer, Level, Lifecycle

Security Reference Model Security Reference Model Security Reference Model

Security Reference Model

Figure 5: CPSS architecture framework description.

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

213

4 STRUCTURAL APPROACH OF

CPSS ANALYSIS AND DESIGN

In the previous section, we discussed the contents of

the 6-tier views in CPSS architecture, and in this

section, we discussed the specific contents that should

be concerned about each subsystem involved, as

shown in Figure 6. Here we discuss physical system,

cyber system and social system, as well as cyber-

physical system and cyber-social system, rather than

man-machine system, because the relationship

between social system and physical system, that is,

the relationship between people and machines, has

been studied for a long time, and will not be repeated

in this paper. In addition, the integration of physical

system and cyber system has become a trend in many

scenarios, we cannot regard the interaction between

human and physical system as the relationship

between them, but the relationship between human

and CPS.

We discuss the relationship between different

systems in the order of levels:

 Performance view: In physical system, we

generally hope that the system can handle a large

number of physical entities and improve the

operation speed and accuracy. The improvement

of performance depends on the emergence and

application of new materials, new energy, new

equipment and new technology. In cyber system,

the improvement of perception ability promotes

the improvement of system autonomy,

automation and decision-making ability.

Because of the emergence of embedded devices

such as sensors, physical system and cyber

system can be combined to form intelligent CPS

units and overall intelligent systems. Intelligent

CPS also promotes the improvement of physical

system performance. In the social system,

people often pursue safety, wealth, freedom and

achievement. In the interaction between people

and cyber system, because of the improvement

of the decision-making ability of cyber system,

people and cyber system make decisions

together to realize human-cyber collaboration.

 Behaviour view: The system supports the

realization of performance through certain

activity processes. The physical system

completes its own tasks through a certain

operation mode and process, and supports the

data analysis, cognitive learning and decision-

making control in the cyber system through the

signal acquisition of the physical system. The

decision-making results are transmitted to the

physical activities through CPS. Social systems

realize their goals and aspirations through a

series of activities such as perception, decision-

making and execution. Human decision-making

and cybernetic system decision-making together

constitute human-cyber collaboration. Human-

cyber collaboration requires mutual

understanding between human and cyber about

behaviour and intentions and a resolution

mechanism when decisions conflict.

 Structure view: Structure is the organization

form of resources or functional components

required by system activities. In physical

systems, the operation of system activities

requires the consumption of energy and

materials, the support of functional facilities and

other infrastructure, and will produce products.

In cyber system, the analysis and storage of data

need the support of physical resources of

information technology infrastructure,

corresponding big data analysis technology and

algorithm model. Information infrastructure and

physical equipment together constitute the

intelligent unit and system component of CPS.

In the social system, groups realize the

distribution of tasks and functions of each

person through a certain organizational structure,

and there are economic, political, cultural and

other social relations between people. Personnel

interact with cyber system through human-

computer interfaces such as brain-computer

interface.

 Application view: Application is the integrated

implementation of system activities.

Mechanical, electronic and energy functional

components together form a physical system to

realize physical activities. In cyber system,

intelligent algorithm is based on model

knowledge and machine learning, which is the

realization of data analysis and cognitive

learning process. Intelligent units are combined

with physical components to obtain intelligent

applications, and intelligent algorithms integrate

all intelligent units to obtain intelligent systems.

In social system, people realize the arrangement

of business activities through process

management. The combination of human-

computer interaction components and

algorithms provides functional components for

the combination of human activities and cyber

activities.

 Infrastructure view: Infrastructures are the basic

support for application running and the

realization of system structure. In the physical

world, the system platform is the location of

physical system deployment, and the physical

prototype is the realization of structural design.

Big data, cloud computing and the Internet of

things

together constitute the industrial Internet.

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

214

Performance

Behavior

Structure

Application

Infrastructure

Security

Cyber-Social SystemPhysical System Cyber-Physical System Cyber System

Social / Human System

Automation,

intelligence

and autonomy

System

intelligence &

autonomy

Safety, wealth,

freedom and

achievement

Realize Distribution

of power and

responsibilities

Network,

information and

data security

Cyber

Security

Software &

Hardware

Security

Realize

human-cyber

collaboration

Perception

and cognition

Human-cyber

collaboration

mechanism

Human-cyber

mutual

understanding

Human-cyber

conflict

resolution

Human-cyber

collaboration

platform

Decision,

control and

optimization

Data acquisition,

fusion and

analysis

Intelligent

decision and

control

Cognition and

learning

Model and

knowledge

Information

technology

infrastructure

Data and Big

Data

Organizational

structure

Working

system

Industrial

Internet

platform

Human-

computer

interface

Social

relations

Perception

New material,

new energy, new

technology and

new equipment

Unit

intelligence

Decision and

control logic

Operation

mode and

process

Energy, material,

function, product,

infrastructure, …

structure

Intelligent

autonomous

system

components

Mechanical,

electronic and

energy functional

components

Intelligent

application

Business process

management and

reconstruction

Human-cyber

collaboration

functional

component

Internet of

Things, Big

Data

Cloud

computing

Intelligent

system

Intelligent

algorithm

Embedded

end, side

platform

System platform,

physical prototype

Industrial

safety

Fieldbus

Power, Speed,

Efficiency,

Accuracy, ...

Operation

security

Personal

health and

safety

Property

safety

Personality

security

Environmental

safety

Energy safety

System

security

Signal

acquisition

Decision and

control

Implementation

and

optimization

Physical

system

Intelligent

unit

Cyber-Human

Collaboration

Security

Brain-

computer

interface

Model and

knowledge

Data and

machine

learning

Operating

instructions

Distributed

control

system

Social /

Human

Security

Figure 6: CPSS analysis and design.

It is a platform for algorithm model to learn and

deploy. It is combined with the physical

platform through embedded devices, fieldbus

and distributed control systems. It is also

organically combined with the working system

and operating specifications of social systems

through the human-computer collaboration

platform.

 Security view: Security concerns the security

factors at all levels of the system. Industrial

safety includes energy safety and environmental

safety, mainly focusing on the normal operation

of system components. Cyber system mainly

focuses on the security of data, network,

software and hardware applications. System

security and operation security focus on the

control of cyber system over physical system.

The social system mainly focuses on human life,

health, property and personality. There are

security problems in the cooperation between

people and cyber about decision conflicts.

5 CONCLUSIONS

This paper studies the cyber-physical-social system

developed with the new generation of artificial

intelligence technology, constructs its reference

architecture framework, and on this basis, puts

forward the structured process of CPSS analysis and

design.

As artificial intelligence increasingly presents

the characteristics of self-learning, self-growth and

self-evolution, and has the ability to explore,

recognize and transform the material world

independently, the ternary system composed of

human, cyber system and physical system presents

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

215

new characteristics. Based on ISO 42010, this paper

constructs the reference architecture of CPSS and

clarifies the internal logic of ternary blending. The

cooperation between human and cyber system will be

the focus of future research, which will be related to

the development of artificial intelligence technology

and the mode of its role in production and life.

On the basis of CPSS architecture, it is necessary

to build a structured process to clarify the

methodology of analysis and design of CPSS ternary

fusion system. This paper focuses on the impact of the

introduction of intelligent technology into industrial

systems on people and manufacturing, and constructs

the analysis process and important concerns.

With the development of intelligent technologies

such as autonomous driving, the architecture,

methodology, modeling and analysis methods,

security system construction and other technologies

of CPSS need to be further developed in engineering

practice.

ACKNOWLEDGEMENTS

This work was supported in part by the science and

technology innovation 2030 - "new generation

artificial intelligence" major project

(2018AAA0101605), and the National Natural

Science Foundation of China (No.61771281,

No.61174168).

REFERENCES

Shi, J., Wan, J., Yan, H., & Suo, H. (2011, November). A

survey of cyber-physical systems. In 2011 international

conference on wireless communications and signal

processing (WCSP) (pp. 1-6). IEEE.

Zhou, J., Li, P., Zhou, Y., Wang, B., Zang, J., & Meng, L.

(2018). Toward new-generation intelligent

manufacturing. Engineering, 4(1), 11-20.

Wang, F. Y. (2010). The emergence of intelligent

enterprises: From CPS to CPSS. IEEE Intelligent

Systems, 25(4), 85-88.

Zhou, K., Liu, T., & Zhou, L. (2015, August). Industry 4.0:

Towards future industrial opportunities and challenges.

In 2015 12th International conference on fuzzy systems

and knowledge discovery (FSKD) (pp. 2147-2152).

IEEE.

Griffor, E. R., Greer, C., Wollman, D. A., & Burns, M. J.

(2017). Framework for cyber-physical systems:

Volume 1, overview.

Romero, D., Bernus, P., Noran, O., Stahre, J., & Fast-

Berglund, Å. (2016, September). The operator 4.0:

Human cyber-physical systems & adaptive automation

towards human-automation symbiosis work systems.

In IFIP international conference on advances in

production management systems (pp. 677-686).

Springer, Cham.

Pacaux-Lemoine, M. P., Berdal, Q., Enjalbert, S., &

Trentesaux, D. (2018, May). Towards human-based

industrial cyber-physical systems. In 2018 IEEE

Industrial Cyber-Physical Systems (ICPS) (pp. 615-

620). IEEE.

Sowe, S. K., Simmon, E., Zettsu, K., De Vaulx, F., &

Bojanova, I. (2016). Cyber-physical-human systems:

Putting people in the loop. IT professional, 18(1), 10-

13.

Gil, M., Albert, M., Fons, J., & Pelechano, V. (2019).

Designing human-in-the-loop autonomous cyber-

physical systems. International journal of human-

computer studies, 130, 21-39.

Calinescu, R., Cámara, J., & Paterson, C. (2019, May).

Socio-cyber-physical systems: Models, opportunities,

open challenges. In 2019 IEEE/ACM 5th International

Workshop on Software Engineering for Smart Cyber-

Physical Systems (SEsCPS) (pp. 2-6). IEEE.

Yao, X., Huang, Y., Huang, Y., Mai, H., Yang, E., & Yu,

H. (2022). Resource Choreography in Cyber-Physical-

Social Systems: Representation, Modeling and

Execution. Computer Integrated Manufacturing

Systems (02),325-338.

Bousdekis, A., Apostolou, D., & Mentzas, G. (2020). A

human cyber physical system framework for operator

4.0–artificial intelligence symbiosis. Manufacturing

letters, 25, 10-15.

Zhou, J., Zhou, Y., Wang, B., & Zang, J. (2019). Human–

cyber–physical systems (HCPSs) in the context of new-

generation intelligent manufacturing. Engineering,

5(4), 624-636.

Yilma, B. A., Panetto, H., & Naudet, Y. (2019, September).

A meta-model of cyber-physical-social system: The

cpss paradigm to support human-machine collaboration

in industry 4.0. In Working Conference on Virtual

Enterprises (pp. 11-20). Springer, Cham.

Lezoche, M., & Panetto, H. (2020). Cyber-Physical

Systems, a new formal paradigm to model redundancy

and resiliency. Enterprise information systems, 14(8),

1150-1171.

Abera, Y. B., Naudet, Y., & Panetto, H. (2020). A new

paradigm and meta-model for cyber-physical-social

systems. IFAC-PapersOnLine, 53(2), 10949-10954.

He, F., Shen, L., & Peng, X. (2021). Resource

Choreography in Cyber-Physical-Social Systems:

Representation, Modeling and Execution. IEEE

Transactions on Services Computing.

Lee, J., Bagheri, B., & Kao, H. A. (2015). A cyber-physical

systems architecture for industry 4.0-based

manufacturing systems. Manufacturing letters, 3, 18-23.

ISO/IEC/IEEE 42010 (2011). ISO/IEC/IEEE 42010:

Systems and Software Engineering – Architecture

Description

Hankel, M., & Rexroth, B. (2015). The reference

architectural model industrie 4.0 (rami 4.0). ZVEI, 2(2),

4-9.

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

216

Lin, S. W., Miller, B., Durand, J., Joshi, R., Didier, P.,

Chigani, A., ... & Witten, B. (2015). Industrial internet

reference architecture. Industrial Internet Consortium

(IIC), Tech. Rep.

ANSI/ISA-95.00.01-2000 (2000). Enterprise-Control

System Integration Part 1: Models and Terminology.

ISA.

OMB (2013). Federal Enterprise Architecture Framework

version 2. The U.S. Office of Management and Budget.

Williams, T. (1998). The Purdue enterprise reference

architecture and methodology (PERA). Handbook of

life cycle engineering: concepts, models, and

technologies, 289.

Banerjee, S.K. (1997). Methodology for integrated

manufacturing planning and control systems design. In:

Artiba, A., Elmaghraby, S.E. (eds) The Planning and

Scheduling of Production Systems. Springer, Boston

Cyber-Physical-Social System (CPSS) Architecture Framework and Methodology

217