Industrial Internet of Things

Technologies and Research Directions

Anand Sharma, Sunil Kumar Jangir, Manish Kumar, Dilip Kumar Choubey, Tarun Shrivastava, S. Balamurugan, Anand Sharma, Sunil Kumar Jangir, Manish Kumar, Dilip Kumar Choubey, Tarun Shrivastava, S. Balamurugan

• Publisher: CRC Press
• Year: 2022
• Print ISBN: 9780367702076
• eBook ISBN: 9781000545296
• Edition: 1
• Topic: Technology & Engineering
• Subtopic: Industrial Engineering
• Index: Technology & Engineering

This book focuses on the key technologies, challenges, and research directions of the Industrial Internet of Things (IIoT). It provides a basis for discussing open principles, methods, and research problems, and provides a systematic overview of the state-of-the-art research efforts, directions, and potential challenges associated with IIoT.

Industrial Internet of Things: Technologies and Research Directions covers how industry automation is projected to be the largest and fastest-growing segment of the market. It explores the collaborative development of high-performance telecommunications, military, industrial, and general-purpose embedded computing applications, and offers a systematic overview of the state-of-the-art research efforts and new potential directions.

Researchers, academicians, and professionals working in this inter-disciplinary area will be interested in this book.

1Artificial Intelligence and Machine Learning for the Industrial Internet of Things (IIoT)

Fanoon Raheem and Nihla Iqbal

South Eastern University of Sri Lanka Oluvil, Sri Lanka

DOI: 10.1201/9781003145004-1

CONTENTS

• 1.1 Introduction
  • 1.1.1 The Concept of IoT
  • 1.1.2 The Concept of IIoT
  • 1.1.3 Industry 4.0 and Data
  • 1.1.4 Artificial Intelligence and Machine Learning
• 1.2 Tactile Industrial Internet of Things (Tactile IIoT)
• 1.3 Artificial Intelligence and Machine Learning
  • 1.3.1 Artificial Intelligence
    • 1.3.1.1 Goal of Artificial Intelligence
    • 1.3.1.2 Categories of Artificial Intelligence
  • 1.3.2 Machine Learning
• 1.4 Artificial Intelligence and Machine Learning for Industrial Internet of Things
  • 1.4.1 Machine Learning Algorithms for IIoT
• 1.5 Challenges of Implementing Artificial Intelligence and Machine Learning in IIoT
• 1.6 Summary
• References

1.1 INTRODUCTION

Businesses around the world are increasingly exploiting the Internet of Things (IoT) to develop innovative creation and service networks that are opening up new marketing strategy and establishing new revenue streams. The temporal variation is rapidly transforming the way paradigm through how industries organize business activities and reach potential customers. Though the IoT connectivity ensures advanced business processes, for industries to understand the full value of empowering IoT, IoT and artificial intelligence (AI) technology must be integrated. This results in supporting smart machines that allow intellectual behavior to be replicated and well-informed decision-making practices. Continuing developments in AI would have a substantial effect throughout the near future impacting the employment, expert knowledge, and HR approaches in almost every sector of industry, stressing the fact that businesses should have the necessary experience for AI enablement. The addition of AI into IoT networks is now becoming a prerequisite to succeed in today's IoT-based artificial environments. Yet companies need to work efficiently to determine how to make it to the importance of AI and IoT integration (Chitkara, Rao, and Yaung 2017) that will result in a perfect value chain analysis to achieve the necessary core competencies along with competitive advantages over the rival industries.

1.1.1 THE CONCEPT OF IOT

The association of different physical devices and objects around the globe through the Internet is referenced as IoT. It is the physical world network that allows certain artifacts to gather and share data, including computers, instruments, cars, buildings, and other things embedded with processors, circuits, applications, detectors, and communication technologies. The IoT makes it possible to remotely track and monitor artifacts through existing network infrastructures. The concept of IoT shows it as the individually recognizable associated objects with radio frequency identification (RFID) technology. The definite concept of IoT, however, is still in the phase of creation and is subject to the viewpoints drawn. Evolving global distribution network with norms that has the ability to self-configuring technologies and networking protocols has traditionally been known as IoT (Gokhale, Bhat, and Bhat 2018). Therefore, the IoT can be seen as the IT industry's next revolution after the Internet, which can be applied anytime, wherever on-demand, to detect, track, and control anything.

The advent of the IoT is going to have a strong influence on many dimensions of the distribution of resources, such as production, sales, transportation, usage, and recycling, as well as states’, companies’, and individuals’ actions (Khan 2019). IoT is now an evolving industrial knowledge infrastructure focused on the Internet. The importance of IoT is felt in standardizing business processes. Throughout this emerging phase, IoT and its associated support frameworks need to be quickly built and configured to satisfy requirements of the industry (Wang et al. 2015).

1.1.2 THE CONCEPT OF IIOT

The Industrial Internet of Things (IIoT) incorporates a wide range of machines linked by software for communications. The corresponding programs, and even the wireless device that compose them, will track, capture, share, interpret, and respond to information instantly to modify their actions or their environment smartly, without external influence. The architecture of the IoT framework must ensure IoT operations that connect virtual objects and physical environments. IoT framework design requires several variables, such as networking, connectivity, and procedures. Attention should be paid to the extensibility, scalability, and operability between devices. Thus, the understanding of IoT architecture is important in adoption and deployment of IIoT that can act as a framework in achieving business advancements. For IoT systems, there is a whole range of architectures provided; however, these are general and therefore not directly relevant to IIoT implementations. Figure 1.1 shows an architecture specifically relevant to industrial applications. From this architecture, larger organizations can meet the integration of information and organizational capabilities (Boyes et al. 2018).

FIGURE 1.1 A four-stage IoT solutions architecture. (Boyes et al. 2018.)

In simple terms IIoT can be elaborated as the one that puts a range of domains like traditional automation and machine-to-machine (M2M) communication, big data and machine learning (ML), and cyber-physical systems together to collect, analyze, and use data from industrial properties and devices, frequently in real time to improve operations. Industries embrace IoT as a means of energy optimization, optimizing wear of consumables, preventative analysis, capacity planning, standard costing, and significant trade control, among others. In keeping with the IoT model, emerging IoT technological possibilities such as new industrial sensors and sensing solutions, new factory wireless protocols, new industries based on IoT platforms, new synergistic applications for intelligent diagnoses, smart factories, smart goods, and intelligent logistics are being built for industries.

1.1.3 INDUSTRY 4.0 AND DATA

The fourth industrial revolution, described by Klaus Schwab, the founder and executive chairman of the World Economic Forum, explains the world as where people switch from digital fields to analog realities using smart devices to empower or even control their lives (Xu, David, and Kim 2018).

Thus, the Industry 4.0 introduced the latest emerging technology to make steam, equipment in factories, labor unions, and even robotics in business activities connectable to the digital and physical environment. Physical and automated interconnections contributed to the formation of volumes, culminating in new approaches. As a part of this, data has become an unavoidable aspect where digital information needs to be collected, refreshed, obtained, and distributed in various respects. Mass storage and cognitive capacity will change industry by integrating data. As already stated, the data changes the industry functionalities and demands new regulatory approaches, as given in Figure 1.2.

FIGURE 1.2 The evolution of Industry 4.0. (GSMA Intelligence 2017.)

Centered on IoT concepts, such as a machine, it is not smart until it is connected to another network of computers. Communicators operate together in ambient devices in IoT are smart, and the system becomes smarter with multiple cycles or iterations (Akugizibwe 2020), It is recognized that the IIoT explicitly shows and acknowledges that nearly everything in the industrial world, like production tracking, materials handling, energy management, quality control, predictive maintenance, workers safety, and smart lifecycle, must be related to everything else in order to satisfy the criteria of simulating the smart machines to showcase the intelligent behavior. Therefore, AI is required to make these things “intelligent” or “smart.”

1.1.4 ARTIFICIAL INTELLIGENC...

Table of contents

• Cover Page
• Half-Title Page
• Series Page
• Title Page
• Copyright Page
• Contents
• Preface
• About the Authors
• Contributors
• Chapter 1 Artificial Intelligence and Machine Learning for the Industrial Internet of Things (IIoT)
• Chapter 2 Role of Internet of Things (IoT) in Electronic Waste Management
• Chapter 3 Creating a Reliable IIoT Framework to Prioritize Workplace Safety in Industries Involving Hazardous Processes
• Chapter 4 Parkinson Disease Prediction and Drug Personalization Using Machine Learning Techniques
• Chapter 5 IoT and Deep Learning-Based Prophecy of COVID-19
• Chapter 6 Machine Learning Applications and Challenges to Protect Privacy in the Internet of Things
• Chapter 7 IoT-Enabled Heart Disease Prediction Using Machine Learning
• Chapter 8 Internet of Everything, the Future of Globalization: A Comprehensive Study
• Chapter 9 A Review of Human–Robot Interaction for Automated Guided Vehicles Using Robot Operating Systems
• Chapter 10 Analysis of Cascading Behavior in Social Networks and IoT
• Chapter 11 Performance Evaluation of Machine Learning Classifiers for Memory Assessment Using EEG Signal
• Chapter 12 Robotic Operating System and Human–Robot Interaction for Automated Guided Vehicles (AGVs): An Application of Internet of Things in Industries
• Chapter 13 A Review on IoT Architectures, Protocols, Security, and Applications
• Chapter 14 Performance Analysis of Distributed Mobility Protocol for Multi-Hop IoT Networks
• Chapter 15 Comparative Analysis of Emotional State Classification Using Different Machine Learning Techniques
• Chapter 16 A Survey on Antennas for IIoT Application
• Index