Big Data Analytics in Smart Manufacturing: Principles and Practices

Description

The significant objective of this edited book is to bridge the gap between smart manufacturing and big data by exploring the challenges and limitations. Companies employ big data technology in the manufacturing field to acquire data about the products. Manufacturing companies could gain a deep business insight by tracking customer details, monitoring fuel consumption, detecting product defects, and supply chain management. Moreover, the convergence of smart manufacturing and big data analytics currently suffers due to data privacy concern, short of qualified personnel, inadequate investment, long-term storage management of high-quality data. The technological advancement makes the data storage more accessible, cheaper and the convergence of these technologies seems to be more promising in the recent era. This book identified the innovative challenges in the industrial domains by integrating heterogeneous data sources such as structured data, semi-structures data, geo-spatial data, textual information, multimedia data, social networking data, etc. It promotes data-driven business modelling processes by adopting big data technologies in the manufacturing industry. Big data analytics is emerging as a promising discipline in the manufacturing industry to build the rigid industrial data platforms. Moreover, big data facilitates process automation in the complete lifecycle of product design and tracking. This book is an essential guide and reference since it synthesizes interdisciplinary theoretical concepts, definitions, and models, involved in smart manufacturing domain. It also provides real-world scenarios and applications, making it accessible to a wider interdisciplinary audience. Features The readers will get an overview about the smart manufacturing system which enables optimized manufacturing processes and benefits the users by increasing overall profit.The researchers will get insight about how the big data technology leverages in finding new associations, factors and patterns through data stream observations in real time smart manufacturing systems.The industrialist can get an overview about the detection of defects in design, rapid response to market, innovative products to meet the customer requirement which can benefit their per capita income in better way.Discusses technical viewpoints, concepts, theories, and underlying assumptions that are used in smart manufacturing.Information delivered in a user-friendly manner for students, researchers, industrial experts, and business innovators, as well as for professionals and practitioners. Cover Half Title Title Page Copyright Page Contents Preface Editors Contributors 1. Machine Learning Techniques and Big Data Analytics for Smart Manufacturing 1.1 An Overview of Smart Manufacturing 1.1.1 Upsides and Downsides of Smart Manufacturing 1.2 Machine Learning in Smart Manufacturing 1.2.1 Supervised Machine Learning in Smart Manufacturing 1.2.2 Unsupervised Machine Learning in Smart Manufacturing 1.3 Big Data Analysis in Smart Manufacturing 1.3.1 Infrastructure 1.3.2 Architecture 1.4 Comparative Study of Smart Manufacturing 1.5 Applications Used in Smart Manufacturing 1.5.1 Distinct Examination for Item Quality Assessment 1.5.2 Symptomatic Investigation for Shortcoming Appraisal 1.5.3 Prescient Examination for Deformity Anticipation 1.6 Challenges of Machine Learning in Smart Manufacturing 1.7 Advantage of Machine Learning in Smart Manufacturing 1.7.1 Deep Learning Model for Smart Manufacturing 1.7.2 Smart Manufacturing of Industrial IoT Robotics 1.7.3 Smart Factory Production 1.7.4 Data Clustering-Based ML 1.7.5 Imbalanced Data and Comparative Analysis in Smart Manufacturing 1.7.6 Human to Machine Applications for Smart Industry 1.8 Future of Smart Manufacturing 1.8.1 Smart 3D Printing Techniques Using AI and Cloud 1.8.2 Blockchain Secured Industry 4.0 1.8.3 Smart Transportation System 1.8.3.1 Safety and Security in Autonomous Vehicles 1.8.4 Augmented Reality in AI-Based Education System 1.9 Conclusion References 2. Data-Driven Paradigm for Smart Manufacturing in the Context of Big Data Analytics 2.1 Introduction 2.2 Historical Background 2.3 Smart Manufacturing 2.4 The DT 2.5 Big Data 2.6 Data-Driven Paradigm 2.7 Conclusion References 3. Data-Driven Models in Machine Learning: An Enabler of Smart Manufacturing 3.1 Introduction 3.2 3D Printing Process 3.2.1 3D Printing - Advantages 3.2.2 3D Printing - Disadvantages 3.2.3 3D Printing - Beneficiary Industries 3.2.4 3D Printing Techniques 3.2.4.1 Powder Bed Fusion 3.2.4.2 Selective Laser Sintering and Melting 3.2.4.3 Electron Beam Melting 3.2.4.4 Photo-Polymerization 3.2.4.5 Stereolithography 3.2.4.6 Digital Light Processing 3.2.4.7 Inkjet: Binder Jetting 3.2.4.8 Inkjet: Material Jetting 3.2.4.9 Material Extrusion 3.2.4.10 Selective Deposition Lamination (SDL) 3.3 Need for Parametric Analysis and Optimization in 3D Printing 3.4 ML Technique - Overview 3.4.1 Reasons for Adoption of ML in 21st Century 3.4.2 Popular Techniques of ML Applied in AM 3.4.2.1 Linear Regression 3.4.2.2 Artificial Neural Networks 3.4.3 Applications of ANN in 3D Printing 3.5 ML in Additive Manufacturing Industry - State of Art 3.6 Case Studies for the Experimental Data 3.6.1 Case Study I 3.6.2 Case Study II 3.7 Comparison of ML Analysis to Statistical Analysis Tools 3.8 Challenges Associated for Ml Applications to 3D Printing 3.8.1 Big Data Challenges 3.8.2 Scope of Issue Addressal/Advanced Techniques 3.8.2.1 Data Augmentation 3.8.2.2 Transfer Learning 3.8.3 Few-Shot Learning 3.9 Conclusions References 4. Local Time Invariant Learning from Industrial Big Data for Predictive Maintenance in Smart Manufacturing 4.1 Portfolio of Predictive Maintenance and Condition Monitoring 4.1.1 Characteristics of Industry 4.0 4.1.2 Industry 4.0: Revolution or Evolution? 4.2 Condition Monitoring and Predictive Maintenance 4.2.1 Taxonomy of Maintenance Activities in Industries 4.2.1.1 Preventive Maintenance 4.2.1.2 Predictive Maintenance 4.3 Role of Predictive Maintenance in Smart Manufacturing 4.4 Niche of Big Data in Smart Manufacturing 4.4.1 Significance of RUL in Mechanical Machineries 4.5 Local Time Invariant Learning Through BGRU 4.5.1 Gated Recurrent Unit 4.5.2 Bidirectional GRU 4.6 RUL Prediction Through BGRU from Mechanical Big Data 4.7 Exploration of the Experimental Results 4.8 Conclusion References 5. Integration of Industrial IoT and Big Data Analytics for Smart Manufacturing Industries: Perspectives and Challenges 5.1 Introduction 5.1.1 Industry Automation System 5.1.2 Industrial Automation Types 5.1.2.1 Fixed Automation System 5.1.2.2 Programmable Automation System 5.1.2.3 Soft Automation System 5.1.2.4 Integrated Automation System 5.2 Industry 4.0 Revolution 5.2.1 International Standards of Industry 4.0 5.3 IoT Components and Its Protocols 5.3.1 Things 5.3.2 Gateways 5.3.3 Cloud Gateway 5.3.4 Data Lake 5.3.5 Data Analytics 5.3.6 Machine Learning 5.3.7 Control Applications 5.3.8 User Applications 5.4 M2M Communication in Smart Manufacturing 5.5 IoT in Smart Manufacturing 5.5.1 Advanced Analysis 5.5.2 Inventory Monitoring 5.5.3 Remote Process Monitoring 5.5.4 Abnormality Reporting 5.6 Big Data Analytics in Smart Manufacturing 5.6.1 Self-Service Systems 5.6.1.1 Elimination of bottlenecks 5.6.1.2 Predictive Maintenance 5.6.1.3 Automation Production Management 5.6.1.4 Predictive Demand 5.7 Convergence of IIoT and Big Data Analytics 5.8 Smart Manufacturing in Industries 5.8.1 Building Blocks of Smart Manufacturing 5.8.1.1 Flat 5.8.1.2 Data-Driven 5.8.1.3 Sustainable 5.8.1.4 Agile 5.8.1.5 Innovative 5.8.1.6 Current 5.8.1.7 Profitable 5.8.2 IIoT Implementation 5.9 Smart Manufacturing in MSMEs 5.9.1 Smart Manufacturing in Large-Scale Industry 5.9.2 Intelligent Robots for Smart Manufacturing 5.9.2.1 Industrial Robots 5.9.2.2 Collaborative Robots 5.10 Challenges in Integrating Industrial IoT and Big Data Analytics 5.10.1 Privacy 5.10.2 Cyber Security 5.10.3 Scalability 5.10.4 Connectivity and Communication 5.10.5 Efficiency 5.11 Research Scope in IIoT 5.11.1 Energy Management 5.11.2 Heterogeneous QoS 5.11.3 Resource Management 5.11.4 Data Offloading Decision 5.12 Conclusion References 6. Multimodal Architecture for Emotion Prediction in Videos Using Ensemble Learning 6.1 Introduction 6.2 Related Work 6.3 Dataset Acquisition 6.3.1 Dataset 6.3.2 Data Pre-Processing 6.4 System Design 6.4.1 System Pipeline 6.4.2 Convolutional Neural Network 6.4.2.1 Input Layer 6.4.2.2 Convolutional Layer 6.4.2.3 Dense Layer 6.4.2.4 Output Layer 6.4.3 Audio Feature Extraction 6.4.4 Support Vector Machine 6.4.5 Multi-Layer Perceptron 6.4.6 Ensemble Learning 6.5 System Implementation 6.5.1 Emotion Prediction from Videos: CNN Model Training 6.5.2 Emotion Prediction from Audio: SVM-MLP Training 6.5.3 Combining the Video and Audio Using Ensemble Learning 6.6 Result and Analysis 6.6.1 Testing the CNN Model 6.6.2 Testing the SVM and MLP Model 6.6.3 Testing the Ensemble Model 6.7 Conclusion References 7. Deep PHM: IoT-Based Deep Learning Approach on Prediction of Prognostics and Health Management of an Aircraft Engine 7.1 Introduction 7.2 Overview of Prognostics and Health Management 7.3 Steps Involved in PHM 7.3.1 Data Acquisition 7.3.2 Data Pre-processing 7.3.3 Detection 7.3.4 Diagnostics 7.3.5 Prognostics 7.3.6 Decision-Making 7.3.7 Human-Machine Interface 7.4 PHM in Aerospace Industry 7.4.1 Sensors Used in the Gas Turbofan Engine 7.4.1.1 Temperature Sensors 7.4.1.2 Total Air Gas Temperature Sensor 7.4.1.3 Exhaust Gas Temperature Sensor 7.4.1.4 Vibration Sensors 7.4.1.5 Speed Sensors 7.4.1.6 Fuel Sensors for Flow 7.4.1.7 Altimeter Sensors 7.5 Dataset Description 7.5.1 Long Short-Term Memory 7.5.2 Experimental Analysis on C-MAPPS 7.5.2.1 Performance Metric Selection 7.5.2.2 Result Analysis 7.6 Conclusion References 8. A Comprehensive Study on Accelerating Smart Manufacturers Using Ubiquitous Robotic Technology 8.1 Introduction 8.2 Related Works 8.3 Smart Manufacturing Systems 8.3.1 Why Do We Need Ubiquitous Robotics? 8.4 Ubiquitous Robotics 8.4.1 System Design 8.4.2 Part-Based Hardware Measure 8.4.3 Concept for Ubiquitous Industrial Robot Work Cell 8.5 Ubiquitous Computing 8.5.1 Advantages of Ubiquitous Computing 8.6 Conclusion 8.7 Future Scope References 9. Machine Learning Techniques and Big Data Tools in Design and Manufacturing 9.1 Introduction 9.2 Literature Survey 9.2.1 Analytics in Climate Big Data 9.2.2 Problem and Challenges 9.3 Contribution 9.4 Proposed Method 9.5 Big Data Analysis 9.5.1 Benefits of Big Data Analytics 9.5.2 Data Understanding 9.5.3 Data Preparation 9.6 Feature Selection 9.6.1 FA-Based Feature Selection 9.7 Exploratory Analysis 9.8 Classification 9.8.1 NB 9.8.2 Multiple Regression-Logistic Regression 9.8.3 XGBoost Classifier 9.9 Evaluation 9.9.1 Methods and Modeling 9.10 Result and Discussion 9.10.1 Performance Analysis 9.11 Conclusion References 10. Principle Comprehension of IoT and Smart Manufacturing System 10.1 Introduction to IoT 10.1.1 History and Evolution of IoT 10.2 IoT Platforms and Operating System 10.2.1 IoT Platforms 10.2.1.1 Hardware 10.2.1.2 Connectivity 10.2.1.3 Best Communication Option 10.2.1.3.1 IoT Connectivity Options 10.2.2 Operating System 10.2.2.1 Ubuntu Core 10.2.2.2 Contiki 10.2.2.3 Android Things 10.2.2.4 RIOT 10.3 Security in IoT Protocols and Technologies 10.3.1 IoT Network Protocols 10.3.1.1 Internet Protocol Network 10.3.1.1.1 HTTP (Hyper Text Transfer Protocol) 10.3.1.1.2 LoRaWAN (Longe-Range Wide Area Network) 10.3.1.2 Non-Internet Protocols Network 10.3.2 Security 10.4 Applications of IoT 10.4.1 Media 10.4.2 Transportation 10.4.3 Manufacturing 10.4.4 Building and Home Automation 10.4.5 Energy Management 10.4.6 Infrastructure Monitoring 10.4.7 Medical and Healthcare System 10.4.8 Environment Monitoring 10.5 IoT Application Fields 10.6 IoT Enabling Technologies 10.6.1 Cloud Computing 10.6.2 Big Data 10.6.3 Wireless Sensors Network 10.6.4 Communication Protocols 10.6.4.1 MQTT 10.6.4.2 CoAP 10.6.4.3 AMQP 10.6.5 Embedded System 10.7 IoT for Smart Manufacturing 10.7.1 IIoT Impact in Manufacturing 10.7.2 Supply Chain 10.7.3 Remote and Third-Party Operations 10.7.4 Remote Production Control 10.7.5 Predictive Maintenance 10.8 IoT Industrial Use Cases 10.8.1 ABB - Smart Robotics 10.8.2 Connected Vehicle 10.8.3 Asset Tracking 10.8.4 Smart Metering 10.9 Conclusion References Index