AI Factory: Theories, Applications and Case Studies

Description

This book provides insights into how to approach and utilise data science tools, technologies, and methodologies related to artificial intelligence (AI) in industrial contexts. It explains the essence of distributed computing and AI technologies and their interconnections. It includes descriptions of various technology and methodology approaches and their purpose and benefits when developing AI solutions in industrial contexts. In addition, this book summarises experiences from AI technology deployment projects from several industrial sectors. Features Presents a compendium of methodologies and technologies in industrial AI and digitalisation. Illustrates the sensor-to-actuation approach showing the complete cycle, which defines and differentiates AI and digitalisation. Covers a broad range of academic and industrial issues within the field of asset management. Discusses the impact of Industry 4.0 in other sectors. Includes a dedicated chapter on real-time case studies. This book is aimed at researchers and professionals in industrial and software engineering, network security, AI and machine learning (ML), engineering managers, operational and maintenance specialists, asset managers, and digital and AI manufacturing specialists. Cover Half Title Series Page Title Page Copyright Page Table of Contents About the Authors Foreword Preface Acknowledgements Prologue Chapter 1 Introduction 1.1 AI Factory 1.1.1 Artificial Intelligence 1.1.2 Industrial Automation 1.1.3 Engineering Knowledge 1.1.4 System Understanding 1.1.5 Context Adaptation 1.1.6 Connectivity 1.1.7 Information Logistics 1.1.8 In Summary 1.2 Artificial Intelligence-Empowered Analytics 1.3 AI Revolution in Industry 1.3.1 The AI Revolution is Happening Now 1.3.1.1 Healthcare 1.3.1.2 Education 1.3.1.3 Banking, Financial Services and Insurance (BFSI) 1.3.1.4 Retail and e-Commerce 1.3.1.5 Gaming and Entertainment 1.3.2 The Road to Superintelligence 1.4 AI Winter, AI Spring 1.4.1 The First AI Spring 1.4.2 The First AI Winter 1.4.3 The Second AI Spring 1.4.4 The Second AI Winter 1.4.5 Another Spring: The DL Revolution 1.4.6 Predictions of Another AI Winter 1.4.7 Surviving the Next AI Winter: From Myths to Realities 1.5 The Value of AI 1.5.1 Challenges of AI 1.6 Power of AI vs. Human Intelligence 1.7 Technologies Powering AI: ML and DL 1.8 Perception and Cognition References Chapter 2 Digital Twins 2.1 Basic Concept of Digital Twin 2.2 History of DT 2.3 What is Digital Twin? Its Intrinsic Characteristics 2.3.1 Why Use Digital Twin? 2.3.2 How Does Digital Twin Work? 2.3.2.1 Digital Twin and Simulation 2.3.2.2 Digital Twin and Cyber-Physical Systems 2.4 The Evolution of Digital Twin 2.5 Data Twin and the Physical World 2.6 Data Twin and the Digital World 2.7 Useful Terms and Classifications 2.7.1 Prototypes, Instances, and Aggregates 2.7.2 Digital Twin Classes 2.7.3 Digital Twin Categories 2.8 Level of Integration 2.8.1 Digital Model 2.8.2 Digital Shadow 2.8.3 Digital Twin 2.9 Main Characteristics of Digital Twin 2.10 Modelling Digital Twins 2.10.1 Systems Modelling Language (SysML) 2.10.2 Simulation as the Basis of Digital Twin Technology 2.10.3 The Connection Between MES-Systems and Digital Twins 2.10.4 Application Tools 2.11 Smart Manufacturing: An Example of Digital Twin Development and Operation 2.11.1 Development of the Smart Factory Cell 2.11.2 Operation of the Smart Factory Cell 2.12 Some Applications of Digital Twins 2.13 Uses of Digital Twin Technology 2.13.1 Current State of the Art 2.13.1.1 Components of DT 2.13.1.2 Properties of a DT 2.13.1.3 How DT Differs From Existing Technologies 2.13.1.4 A Brief Overview of Similar Concepts that Preceded DT 2.13.1.5 Added Value of Digital Twins 2.13.2 Specific Applications of Digital Twins in Maintenance 2.14 How are Digital Twins Used in Maintenance? 2.15 Digital Twins and Predictive Maintenance 2.16 A Digital Twin Maintenance Use Case: Point Machine for Train Switches 2.17 Planning the Digital Twin 2.18 Digital Twin During Operation Phase 2.19 Hybrid Analysis and Fleet Data 2.20 Steps to Ensure Widespread Implementation of Digital Twin 2.21 Digital Twin and its Impact on Industry 4.0 References Chapter 3 Hypes and Trends in Industry 3.1 Asset Management 3.1.1 Challenges to Asset Management 3.1.2 Intelligent Asset Management 3.1.3 Taxonomy of AAM 3.2 Tracking and Tracing in Asset Management 3.2.1 What Can be Tracked and Traced? 3.2.2 Challenges of Tracking and Tracing 3.2.3 Benefits of Tracking and Tracing 3.3 Green Industry (Sustainable) 3.3.1 Sustainability Green Industry 4.0 3.3.1.1 Sustainable Green Industry Model 3.4 Industry 4.0 3.4.1 What is Industry 4.0? 3.4.2 Talking About a Revolution: What is New in Industry 4.0? 3.4.3 On the Path to Industry 4.0: What Needs to be Done? 3.4.4 Key Paradigms of Industry 4.0 3.4.5 Four Components of Networked Production 3.4.6 Connected Technologies 3.4.7 Nine Pillars of Technological Advancement 3.4.7.1 Big Data and Analytics 3.4.7.2 Autonomous Robots 3.4.7.3 Simulation 3.4.7.4 Horizontal and Vertical System Integration 3.4.7.5 Industrial Internet of Things (IIoT) 3.4.7.6 Cybersecurity 3.4.7.7 The Cloud 3.4.7.8 Additive Manufacturing 3.4.7.9 Augmented Reality 3.4.8 Other Industry 4.0 Components 3.4.8.1 Cyber-Physical Systems (CPS) 3.4.8.2 Internet of Things (IoT) 3.4.8.3 Internet of Services 3.4.8.4 Smart Factory 3.4.9 The Impact of Industry 4.0 3.4.9.1 Quantifying the Impact 3.4.9.2 Producers: Transforming Production Processes and Systems 3.4.9.3 Manufacturing-System Suppliers: Meeting New Demands and Defining New Standards 3.4.10 How Will Industry 4.0 Impact Equipment? 3.5 Digitalisation and Digitisation 3.6 Data, Models, and Algorithm 3.7 Transformative Technologies 3.7.1 Artificial Intelligence (AI) 3.7.2 The Internet of Things (IoT) 3.7.3 Blockchain 3.7.4 Some Implications 3.8 Artificial Intelligence vs Industrial Artificial Intelligence 3.8.1 Key Elements in Industrial AI: ABCDE 3.8.2 Industrial AI Ecosystem 3.8.2.1 Data Technology 3.8.2.2 Analytics Technology 3.8.2.3 Platform Technology 3.8.2.4 Operations Technology 3.9 Autonomy and Automation 3.9.1 Autonomy and Asset Management 3.9.2 Drones and Robots 3.9.2.1 Deploying Robots 3.9.3 Strong Automation Base Layer 3.9.4 Autonomy in Industry Today 3.9.4.1 Challenges of Autonomy 3.10 Digital Transformation 3.10.1 Defining Digital Transformation 3.10.2 Digital Transformation – The Future of Predictive Maintenance 3.10.2.1 Applying Digital Transformation in Maintenance References Chapter 4 Data Analytics 4.1 Data-Driven and Model-Driven Approaches 4.1.1 Data Mining and Knowledge Discovery 4.2 Types of Analytics 4.2.1 Descriptive Analytics 4.2.1.1 What is Descriptive Analytics? 4.2.1.2 How Does Descriptive Analytics Work? 4.2.1.3 How is Descriptive Analytics Used? 4.2.1.4 What Can Descriptive Analytics Tell Us? 4.2.1.5 Steps in Descriptive Analytics 4.2.1.6 Benefits and Drawbacks of Descriptive Analytics 4.2.2 Diagnostic Analytics 4.2.2.1 Hypothesis Testing 4.2.2.2 Correlation vs. Causation 4.2.2.3 Diagnostic Regression Analysis 4.2.2.4 How Do You Get Started with Diagnostic Analytics? 4.2.3 Maintenance Predictive Analytics 4.2.3.1 What is Predictive Analytics? 4.2.3.2 How Does Predictive Analytics Work? 4.2.3.3 What Can Predictive Analytics Tell Us? 4.2.3.4 What Are the Advantages and Disadvantages of Predictive Analysis? 4.2.3.5 Predictive Analytics Techniques 4.2.3.6 How Can a Predictive Analytics Process Be Developed? 4.2.3.7 Predictive Maintenance Embraces Analytics 4.2.3.8 Metrics for Predictive Maintenance Analytics 4.2.3.9 Technologies Used for Predictive Maintenance Analytics 4.2.3.10 Predictive Maintenance and Data Analytics 4.2.3.11 Predictive Asset Maintenance Analytics 4.2.4 Prescriptive Analytics 4.2.4.1 What is Prescriptive Analytics? 4.2.4.2 How Does Prescriptive Analytics Work? 4.2.4.3 What Can Prescriptive Analytics Tell Us? 4.2.4.4 What Are the Advantages and Disadvantages of Prescriptive Analytics? 4.2.4.5 Getting Started in Prescriptive Analysis 4.2.4.6 Maintenance Prescriptive Analytics: A Cure for Downtime 4.2.4.7 Prescription 4.2.4.8 Scale Out 4.2.4.9 The Need For Prescriptive Analytics in Maintenance: A Case Study 4.3 Big Data Analytics Methods 4.3.1 Defining Big Data Analytics 4.3.2 Defining Big Data Via the Three Vs 4.3.2.1 Data Volume as a Defining Attribute of Big Data 4.3.2.2 Data Type Variety as a Defining Attribute of Big Data 4.3.2.3 Data Feed Velocity as a Defining Attribute of Big Data 4.3.3 Text Analytics 4.3.4 Audio Analytics 4.3.5 Video Analytics 4.3.6 Social Media Analytics 4.4 Maintenance Strategies with Big Data Analytics 4.5 Data-Driven and Model-Driven Approaches 4.5.1 Data Mining and Knowledge Discovery 4.6 Maintenance Descriptive Analytics 4.7 Maintenance Diagnostic Analytics 4.8 Maintenance Predictive Analytics 4.9 Maintenance Prescriptive Analytics 4.10 Big Data Analytics Methods 4.10.1 Text Analytics 4.10.2 Audio Analytics 4.10.3 Video Analytics 4.10.4 Social Media Analytics 4.11 Big Data Management and Governance 4.12 Big Data Access and Analysis 4.13 Big Data Visualisation 4.14 Big Data Ingestion 4.15 Big Data Cluster Management 4.16 Big Data Distributions 4.17 Data Governance 4.18 Data Access 4.19 Data Analysis 4.20 Bid Data File System 4.20.1 Quantcast File System 4.20.2 Hadoop Distributed File System 4.20.3 Cassandra File System (CFS) 4.20.4 GlusterFS 4.20.5 Lustre 4.20.6 Parallel Virtual File System 4.20.7 Orange File System (OrangeFS) 4.20.8 BeeGFS 4.20.9 MapR-FS 4.20.9.1 Kudu References Chapter 5 Data-Driven Decision-Making 5.1 Data for Decision-Making 5.1.1 Data-Driven Decision-Making 5.1.2 The Process of Data-Driven Decision-Making 5.1.3 The Context of Data-Driven Decision-Making 5.1.4 The Importance of Data-Driven Decision-Making 5.1.5 Common Challenges of Data-Driven Decision-Making 5.1.5.1 A Lack of Infrastructure and Tools 5.1.5.2 Poor Quality Data 5.1.5.3 Siloed Data 5.1.5.4 A Lack of Buy-In 5.1.5.5 Not Knowing How to Use Data 5.1.5.6 Being Unable to Identify Actionable Data 5.1.5.7 Too Much Focus on Data 5.1.6 Data-Driven Decision-Making for Industry 4.0 Maintenance Applications 5.1.6.1 Augmented Reality 5.1.6.2 Internet of Things 5.1.6.3 System Integration 5.1.6.4 Cloud Computing 5.1.6.5 Big Data Analytics 5.1.6.6 Cyber Security 5.1.6.7 Additive Manufacturing 5.1.6.8 Autonomous Robots 5.1.6.9 Simulation 5.1.7 Data-Driven Decision-Making Versus Intuition 5.2 Data Quality 5.2.1 eMaintenance and Data Quality 5.2.1.1 Problems in Data Quality 5.2.1.2 Data Quality in The Maintenance Phases 5.2.2 Data Quality Problems 5.3 Data Augmentation 5.3.1 Importance of Data Augmentation in Machine Learning 5.3.2 Advanced Models for Data Augmentation 5.3.3 Image Recognition and Natural Language Processing 5.3.3.1 Image Classification and Segmentation 5.3.3.2 Natural Language Processing 5.3.4 Benefits of Data Augmentation 5.3.5 Challenges of Data Augmentation 5.3.6 Data Augmentation Methods 5.3.6.1 Traditional Transformations 5.3.6.2 Generative Adversarial Networks 5.3.6.3 Texture Transfer 5.3.6.4 Convolutional Neural Networks 5.3.7 Data Augmentation for Data Management 5.3.7.1 Data Augmentation for Data Preparation 5.3.7.2 Data Augmentation for Data Integration 5.3.8 Advanced Data Augmentation 5.3.8.1 Interpolation-Based Data Augmentation 5.3.8.2 Generation-Based Data Augmentation 5.3.8.3 Learned-Data Augmentation 5.3.9 Data Augmentation with Other Learning Paradigms 5.3.9.1 Semi-Supervised and Active Learning 5.3.9.2 Weak Supervision 5.3.9.3 Pre-Training for Relational Data 5.4 Information Logistics 5.4.1 Information Logistics and eMaintenance 5.4.2 Information Logistics and Information Flow in an Era of Industry 4.0 5.4.3 Information Life Cycle 5.4.4 eMaintenance – Information Logistics for Maintenance Support 5.5 Data-Driven Challenges References Chapter 6 Fundamental in Artificial Intelligence 6.1 What is Decision-Making? 6.1.1 Importance of Decision-Making 6.1.2 Features or Characteristics of Decision-Making 6.1.3 Principles of Decision-Making 6.2 General Decision-Making Process 6.3 Problem-Solving Process in Industrial Contexts 6.3.1 Six-Step Problem-Solving Model 6.3.1.1 Step One: Identify the Problem 6.3.1.2 Step Two: Determine the Root Cause(s) of the Problem 6.3.1.3 Step Three: Develop Alternative Solutions 6.3.1.4 Step Four: Select a Solution 6.3.1.5 Step Five: Implement the Solution 6.3.1.6 Step Six: Evaluate the Outcome 6.4 System Thinking and Computer Science 6.5 Decision Support Systems 6.6 Data in a Decision-Making Process 6.7 Knowledge Discovery 6.7.1 Approaches to KDP Modelling 6.7.2 The Leading KDP Models 6.8 Business Intelligence 6.8.1 Business Intelligence on a Practical Level 6.8.2 What Does Business Intelligence Do? 6.8.3 Differences Between Artificial Intelligence and Business Intelligence 6.9 Database and Knowledge Base in Decision Support Systems 6.9.1 Differences Between a Database and a Knowledge Base 6.10 Inference Mechanisms in Artificial Intelligence 6.10.1 Deductive Reasoning 6.10.2 Inductive Reasoning 6.10.3 Adductive Reasoning 6.10.4 Case-Based Reasoning 6.10.5 Monotonic Reasoning and Non-Monotonic Reasoning 6.11 Knowledge Interpretation: The Role of Inference 6.11.1 Inference Engine Architecture 6.11.2 Inference Engine Implementations 6.12 From Data to Wisdom 6.12.1 Data, Information, Knowledge, and Wisdom 6.13 AI and Software Engineering 6.13.1 Artificial Intelligence and Software Engineering 6.13.1.1 Aspects of AI 6.13.1.2 Aspects of SE 6.13.2 The Role of Artificial Intelligence in Software Engineering 6.13.3 When Does Artificial Intelligence for Software Engineering Work Well? 6.13.4 Relationship Between Approaches to Artificial Intelligence for Software Engineering 6.13.5 Intersections Between Artificial Intelligence and SE 6.13.5.1 Agent-Oriented Software Engineering 6.13.5.2 Knowledge-Based SE 6.13.5.3 Computational Intelligence and Knowledge Discovery 6.13.5.4 Ambient Intelligence References Chapter 7 Systems Thinking and Systems Engineering 7.1 Definition of System 7.1.1 Characteristics of a System 7.1.1.1 System Structure 7.1.1.2 System Stakeholders 7.1.1.3 System Attributes 7.1.1.4 System Boundaries 7.1.1.5 System Needs 7.1.1.6 System Constraints 7.1.2 Systems Engineering 7.1.2.1 Holistic View 7.1.2.2 Interdisciplinary Field 7.1.2.3 Managing Complexity 7.1.2.4 Systems Engineering Processes 7.2 Systems-of-Systems 7.2.1 Manufacturing Supply Chains 7.2.2 Embedded Automotive Systems 7.2.3 Smart Grids 7.3 System of Interest 7.3.1 System of Interest Architectural Elements 7.3.1.1 Personnel System Element 7.3.1.2 Equipment System Element 7.3.1.3 Mission Resources System Element 7.3.1.4 Procedural Data System Element 7.3.1.5 System Reponse Element 7.3.1.6 Facilities System Element 7.4 Enabling Systems 7.5 System Lifecycle 7.5.1 Stages of a System Lifecycle 7.5.1.1 Conception 7.5.1.2 Design and Development 7.5.1.3 Production 7.5.1.4 Utilisation 7.5.1.5 Maintenance and Support 7.5.1.6 Retirement 7.5.1.7 Applications of the Six Lifecycle Stages 7.5.2 Defining Lifecycle Models 7.5.2.1 A Linear Lifecycle Model 7.5.2.2 An Iterative Lifecycle Model 7.5.2.3 An Incremental Lifecycle Model 7.6 Hierarchies 7.7 System Item Structure 7.7.1 Types of Systems 7.7.1.1 Open and Closed Systems 7.7.1.2 Deterministic and Probabilistic Systems 7.7.1.3 Physical and Abstract Systems 7.7.1.4 Man-Made Information Systems 7.7.2 Information Systems in the Organisational Context 7.7.2.1 Formal Information Systems 7.7.3 Informal Information Systems 7.7.4 Computer-Based Information Systems References Chapter 8 Software Engineering 8.1 Software Engineering Overview 8.2 From Programming Languages to Software Architecture 8.2.1 High-Level Programming Languages 8.2.2 Abstract Data Types 8.2.3 Software Architecture 8.3 System Software 8.3.1 Functions of System Software 8.3.2 Application Software 8.3.3 Engineering/Scientific Software 8.3.4 Embedded Software 8.3.5 Web Applications 8.3.6 Artificial Intelligence Software 8.4 Software Evolution 8.4.1 Software Evolution Laws 8.4.1.1 Static-Type (S-Type) 8.4.1.2 Practical-Type (P-Type) 8.4.1.3 Embedded-Type (E-Type) 8.5 Paradigms of Software Engineering 8.5.1 Traditional Software Engineering Paradigms 8.5.1.1 Classical Lifecycle Development Paradigms 8.5.1.2 Incremental Development Paradigms 8.5.1.3 Evolutionary Paradigms 8.5.2 Advanced Software Engineering Paradigms 8.5.2.1 Agile Development Paradigm 8.5.2.2 Aspect-Oriented Development Paradigm 8.5.2.3 Cleanroom Development Paradigm 8.5.2.4 Component-Based Development Paradigm 8.6 Software Architecture Models 8.6.1 Layered Architecture 8.6.2 Event-Driven Architecture 8.6.3 Microkernel Architecture 8.6.4 Microservices Architecture 8.6.5 Space-Based Architecture 8.7 Software Systems and Software Engineering Processes 8.7.1 Software System Components 8.7.2 Properties of Software 8.7.3 The Importance of Software Engineering 8.7.4 The Software Engineering Process 8.7.4.1 The Layers of Software Engineering 8.7.4.2 A Generic Framework of the Software Process 8.8 Component-Based Software Engineering 8.8.1 Construction-Based Software Engineering Processes 8.8.2 Characteristics of Component-Based Software Engineering 8.8.3 Evolution of Component-Based Software Engineering 8.8.3.1 Preparations 8.8.3.2 Definitions 8.8.3.3 Progression 8.8.3.4 Expansion 8.8.4 Componentisation 8.9 Software Maintenance Overview 8.9.1 Types of Maintenance 8.9.2 Cost of Maintenance 8.9.3 Maintenance Activities 8.9.4 Software Re-Engineering 8.9.4.1 Reverse Engineering 8.9.4.2 Programme Restructuring 8.9.4.3 Forward Engineering 8.9.5 Component Reusability 8.9.5.1 Reuse Process 8.10 Applications of AI in Classical Software Engineering 8.10.1 AI in The Software Engineering Lifecycle 8.10.1.1 AI in Software Project Planning 8.10.1.2 AI at the Stage of Problem Analysis 8.10.1.3 AI at the Stage of Software Design 8.10.1.4 AI at the Stage of Software Implementation 8.10.1.5 AI at the Stage of Software Testing and Integration 8.10.1.6 AI at the Stage of Software Maintenance References Chapter 9 Distributed Computing 9.1 Cloud Computing 9.1.1 Advantages of Cloud Computing 9.1.2 Challenges of Cloud Computing 9.1.3 Relationship of Cloud Computing with Other Technologies 9.1.4 Cloud Computing Service Models 9.1.4.1 Software as a Service 9.1.4.2 Benefits of Software as a Service 9.1.4.3 Platform as a Service 9.1.4.4 Infrastructure as a Service 9.2 Cloud Computing Types 9.2.1 Advantages of Hybrid Clouds 9.2.2 Cloud Computing Architecture 9.3 Fog Computing 9.3.1 Benefits of Fog Computing 9.3.2 Disadvantages of Fog Computing 9.3.3 The Fog Paradigm 9.4 Edge Computing 9.4.1 Advantages of Edge Cloud Computing 9.4.2 How Edge Computing Works 9.4.3 Edge Computing and Internet of Things 9.5 A Comparative Look at Cloud, Fog, and Edge Computing 9.6 Data Storage 9.6.1 Relational Databases 9.6.2 Non-Relational Databases 9.6.3 Other Database Structures 9.6.4 Message Brokers 9.7 Information Management 9.7.1 Centralised Databases 9.7.2 Decentralised Databases 9.7.3 Web Databases 9.7.4 Cloud Database 9.7.5 Data Lakes 9.7.6 Object Storage 9.8 Data Fusion and Integration 9.8.1 Problems with Data Fusion 9.9 Data Quality 9.10 Communication 9.10.1 Machine Learning in Communications 9.10.1.1 Routing in Communication Networks 9.10.1.2 Wireless Communications 9.10.1.3 Security, Privacy, and Communications 9.10.1.4 Smart Services, Smart Infrastructure, and Internet of Things 9.10.1.5 Image and Video Communications 9.11 Cognitive Computing 9.11.1 Theoretical Foundations for Cognitive Computing 9.11.1.1 Cognitive Informatics for Cognitive Computing 9.11.1.2 Neural Informatics for Cognitive Computing 9.11.1.3 Denotational Mathematics for Cognitive Computing 9.11.2 Models of Cognitive Computing 9.11.2.1 Abstract Intelligence Model of Cognitive Computing 9.11.2.2 Computational Intelligence Model of Cognitive Computing 9.11.2.3 Behavioural Model of Cognitive Computing 9.11.3 Applications of Cognitive Computing 9.11.3.1 Autonomous Agent Systems 9.11.3.2 Cognitive Search Engines 9.12 Distributed Ledger 9.12.1 Distributed Ledger and Blockchain 9.12.1.1 How Blockchain Works 9.12.1.2 Types of Blockchain 9.12.1.3 Advantages of Blockchain 9.12.1.4 Uses of Blockchain 9.12.1.5 Limitations of Blockchain 9.12.1.6 Blockchain and Smart Contracts 9.12.1.7 Blockchain Frameworks in Use 9.12.1.8 Bitcoin 9.12.1.9 Ethereum 9.12.1.10 R3 Corda 9.12.1.11 Hyperledger Fabric 9.12.1.12 Comparison of Distributed Ledger Technologies 9.13 Information Security 9.14 Cybersecurity 9.14.1 Challenges and Responses 9.14.2 A Cybersecurity Framework 9.14.3 Access Control Security 9.14.4 Information Transmission Security 9.14.5 Data Storage Security References Chapter 10 Case Studies 10.1 Case Study 1 – AI Factory for Railway 10.1.1 AI Factory for Railway 10.1.1.1 Technology Platform (AI Factory) 10.1.1.2 Digital Governance Platform (eGovernance) 10.1.1.3 Communication Platform 10.1.1.4 Coordinating Platform 10.2 Case Study 2 – AI Factory for Mining 10.2.1 AIF/M Concept 10.2.2 Expected Results and Effects 10.2.3 Four Pillars of AIF/M 10.2.3.1 Technology Platform (AI Factory) 10.2.3.2 Digital Governance Platform (eGovernance) 10.2.3.3 Communication Platform 10.2.3.4 Coordinating Platform 10.3 Case Study 3 – AI Factory – Augmented Realtiy and Virtual Reality Services in the Railway 10.3.1 Data Fusion and Integration 10.3.2 Data Modelling and Analysis 10.3.2.1 Point Cloud Pre-Processing 10.3.2.2 Classification 10.3.2.3 Labelling 10.3.2.4 Object Extraction 10.3.2.5 Model Creation 10.3.3 Context Sensing and Adaptation 10.4 Case Study 5 – AI Factory – Cybersecurity Services in the Railway 10.4.1 Railway System Stakeholders 10.4.1.1 Infrastructure Manager 10.4.1.2 Maintainer 10.4.1.3 Passenger Operator 10.4.1.4 Interaction with Other Operators 10.4.1.5 Dependencies with External Stakeholders 10.4.1.6 Interaction in Public Areas 10.4.1.7 Interactions in Operational Environments References Chapter 11 AI Factory: A Roadmap for AI Transformation 11.1 What is the AI Factory? 11.1.1 Infrastructure of the AI Factory 11.1.2 Becoming an AI Company 11.2 Mastering AI in Manufacturing: The Three Levels of Competency 11.2.1 AI Competency 11.2.1.1 The Apprentice 11.2.1.2 The Journeyman 11.2.1.3 The Master 11.3 Is AI Ready to Run a Factory? 11.3.1 Cyber-Physical Systems 11.3.1.1 Manufacturing and Cyber-Physical Systems 11.3.2 Artificial Intelligence 11.3.2.1 Industrial AI 11.3.2.2 Machine Learning 11.3.2.3 Why is the Adoption of AI Accelerating? 11.3.2.4 Digital Twins in Industry 11.4 The Data-Driven Approach and AI 11.4.1 Pros of AI and a Data-Driven Approach 11.4.2 Cons of AI and a Data-Driven Approach 11.4.3 Implementing the Data-Driven Approach 11.4.3.1 Governance 11.4.3.2 Business 11.4.3.3 Technology 11.5 Data-Driven Approach in Production: Six Implementation Stages and Failure Factors 11.5.1 Six Implementation Steps 11.5.2 Factors of Failure 11.6 Data-Driven Policymaking 11.6.1 Innovations in Data-Driven Policymaking 11.6.1.1 Use of New Data Sources in Policymaking 11.6.1.2 Co-Creation of Policy 11.6.1.3 Government Policymaking in the Transport Sector 11.7 Sustainability: The Triple Bottom Line 11.7.1 Economic Dimension 11.7.2 Social Dimension 11.7.3 Environmental Dimension 11.7.4 Other Sustainability Dimensions 11.7.5 The Triple Bottom Line Implementation 11.7.6 Measuring Sustainability References Chapter 12 In Industrial AI We Believe 12.1 Industrial AI 12.1.1 Industrial AI Versus Other AI Applications 12.1.2 Industrial AI and Levels of Autonomy 12.1.3 Data and Training 12.1.4 Using Trained Industrial AI 12.1.5 Conceptual Framework for Industrial AI 12.1.6 Categories of Industrial AI 12.1.6.1 Product Applications 12.1.6.2 Process Applications 12.1.6.3 Insight Applications 12.1.7 Why Industrial AI? 12.1.8 Challenges and Opportunities 12.1.8.1 Data Availability 12.1.8.2 Data Quality 12.1.8.3 Cybersecurity and Privacy 12.1.9 Industrial AI Application Case Examples 12.1.9.1 Monitoring 12.1.9.2 Optimisation 12.1.9.3 Control 12.1.10 Monitoring, Optimisation, and Control as an AI Maturity Model 12.2 Industrial AI in Action 12.2.1 The Future of Industry: The Self-Optimising Plant 12.3 Applying Industrial AI 12.3.1 Industrial AI Requirements 12.3.2 Industrial AI Solutions Landscape 12.3.2.1 Point Solutions 12.3.2.2 Pre-Trained AI Models and Services 12.3.2.3 Development Platforms 12.3.2.4 Developer Libraries 12.3.2.5 Statistical Packages 12.4 The IMS Architecture for Industrial AI 12.4.1 Data Technology 12.4.2 Analytic Technologies 12.4.3 Platform Technologies 12.4.3.1 Operations Technologies 12.5 Visible and Invisible Issues 12.6 Building the Future with AI 12.6.1 AI in Industry 4.0 12.6.1.1 Predictive Analytics 12.6.1.2 Predictive Maintenance 12.6.2 Industrial Robotics 12.6.2.1 Computer Vision 12.6.2.2 Inventory Management 12.7 We Believe in Industrial AI References Index