[Review] Industrial IoT Driving Smart Manufacturing

3 main points
✔️ Comprehensive Review of IIoT Driving Manufacturing Transformation
✔️ Proposes a Hierarchical Evolutionary Architecture of IIoT Intelligence with Five Layers
✔️ Identifies seven technologies accelerating manufacturing transformation and identifies their contributions

Industrial Internet of Things Intelligence Empowering Smart Manufacturing: A Literature Review
written by Yujiao Hu, Qingmin Jia, Yuao Yao, Yong Lee, Mengjie Lee, Chenyi Wang, Xiaomao Zhou, Renchao Xie, F. Richard Yu
(Submitted on 2 Dec 2023 (v1), last revised 22 Feb 2024 (this version, v2))
Comments: Published on IoTJ.
Subjects:  Artificial Intelligence (cs.AI); Computers and Society (cs.CY)

code：  

The images used in this article are from the paper, the introductory slides, or were created based on them.

Summary

IIoT Intelligence provides innovative and efficient solutions for various aspects of the manufacturing value chain to light the way for manufacturing transformation. IIoT intelligence provides innovative and efficient solutions to various aspects of the manufacturing value chain, lighting the way for manufacturing transformation. However, existing research often focuses on specific areas of IIoT intelligence, leading researchers and readers to bias their understanding of IIoT intelligence and to believe that one-way research is most important for the development of IIoT intelligence. Therefore, this paper provides a comprehensive overview of IIoT intelligence. First, we analyze in detail the inevitability of change in the manufacturing industry and learn from the successful experiences of Chinese companies in practice. Next, we present a definition of IIoT intelligence and show its value in terms of function, operation, deployment, and application. We then propose a hierarchical development architecture of IIoT intelligence consisting of five tiers. The practical value of technological upgrades for each tier will be illustrated through a real-world example from the Lighthouse Factory. We then identify seven technologies that are accelerating the transformation of manufacturing and highlight their contributions; the ethical and environmental impacts of adopting IIoT intelligence are also analyzed. Finally, open issues and development trends will be explored from four aspects to stimulate future research.

Introduction

Manufacturing is an important part of the global economy, accounting for 16% of global GDP, but faces many challenges, including stagnant productivity, increasing demands for personalized customization, and a shrinking workforce. To overcome these challenges, the manufacturing industry needs digital transformation and upgrading. At the same time, the rapid development of intelligent technology is driving digital transformation for governments and businesses. The government has announced policies such as the "New Robot Strategy" and "Industrial Strategy 2030" to promote industrial revolution and innovation in enterprises.

The Industrial Internet of Things (IIoT) is greatly advancing the digitization of industry by connecting industrial devices via networks to support data collection, exchange, and analysis To define the research value of IIoT, several studies have examined IIoT definitions and technologies and proposed future research To clarify the value of IIoT research, several studies have examined IIoT definitions and technologies and proposed future research directions. This has allowed IIoT to be integrated with advanced technologies such as 5G, time-sensitive networking (TSN), and deep learning, enabling ubiquitous and reliable communications, dynamic environmental adaptation, flexible manufacturing, and other intelligence.

Existing research often focuses on specific areas, which can lead to a biased understanding for researchers and readers. In contrast, this paper provides a comprehensive overview of IIoT intelligence, detailing the value of IIoT in manufacturing transformation, hierarchical development architectures, technologies accelerating manufacturing transformation, and open challenges and future trends. It aims to provide readers with a holistic understanding of IIoT intelligence and insights to drive the digital transformation of manufacturing.

Background

The need for digital transformation in the manufacturing industry is driven by labor shortages, soaring labor costs, and increasing demands for personalized customization. For example, according to data from China's National Bureau of Statistics, the manufacturing workforce will continue to decline between 2014 and 2019, while labor costs are rising significantly (Figure 1). In addition, customers' pursuit of personalized customization and patient experience is increasing the complexity of manufacturing systems. This requires manufacturers to drive intelligent transformation to improve productivity and remain competitive.

Figure 1 - Changes in Manufacturing Labor and Costs in China, 2014-2019

Recently, advanced technologies such as cloud computing, 5G connectivity, industrial data analytics, and IIoT (Industrial Internet of Things) have developed rapidly. These technologies are lighting the way for manufacturing upgrades. Many countries are also actively supporting intelligent transformation of the manufacturing industry through their policies. For example, the U.S. Biden administration issued Executive Order 14110 in October 2023 to guide the safe and reliable development and use of AI. This is a whole-of-government effort to promote the responsible development and deployment of AI through industry regulation and cooperation with international partners.

Figure 2 - How Customization and Service-Oriented Demands Increase Complexity of Manufacturing Systems

The EU also announced the Green Deal Industrial Plan for the Net-Zero Age in February 2023 and the Net-Zero Industry Act in February 2024, aiming to promote clean technology and internationalize technical standards within the EU. This is expected to lead to high-quality development in the field of clean technology by 2050.

In addition to these policies, many companies are also embarking on a path of industrial revolution and innovation. For example, Foxconn Industrial Internet (FII) in China has successfully created an effective production architecture for intelligent transformation. This has enabled the company to increase productivity, improve energy efficiency, increase product quality, and reduce costs.

Figure 3 - The Secret to Intelligent Transformation in Manufacturing

Definition of IIoT intelligence and its effects

Definition of IIoT Intelligence

Industrial Internet of Things (IIoT) intelligence refers to a set of technologies, methods, products, and platforms that enable digital connectivity and awareness, intelligent analysis and cognition, and real-time decision making in the entire manufacturing value chain. The manufacturing value chain includes all segments: research and development (R&D), production, operations and maintenance, marketing, management, and services, etc. The essence of IIoT intelligence is the deep combination of intelligent technologies with industrial scenarios, mechanisms, and knowledge, innovative industrial applications such as digital R&D, efficient and immediate decision making, and rapid line reconfiguration.

Effectiveness of IIoT Intelligence

IIoT Intelligence provides innovative solutions and effectiveness for each segment of the manufacturing value chain, including

1. economic benefits: reduced product R&D costs through digital experimentation

• • The traditional product development process requires a lot of labor, materials, money, and time for repeated physical prototyping and testing; IIoT intelligence enables trial-and-error in a digital environment based on historical data and experience, which can significantly reduce costs. In addition, automated failure prediction technology in a digital environment can detect potential problems in advance and make suggestions for improvement quickly. This improves the efficiency of product design and facilitates the entire development process.

2. technical benefits: increased inspection efficiency through automated vision systems

• • Conventional quality control involves manual inspection by the human eye, which limits the accuracy and speed of inspections; IIoT Intelligence utilizes cameras and sensors to inspect product attributes in real time and automatically detect defects using AI algorithms. This dramatically improves inspection accuracy and reduces defect rates. In addition, it speeds up the entire inspection process, significantly increasing the efficiency of the production line.

3. organizational behavior effects: increased labor efficiency through remote monitoring and control

• • Traditional manufacturing sites require workers to be physically on site, limiting labor efficiency improvements; IIoT intelligence connects and recognizes information about equipment, workers, and intermediate products, enabling remote monitoring and control. This allows for unattended operation and remote monitoring in hazardous environments, improving safety. In addition, workers' schedules can be flexibly adjusted based on real-time progress, greatly improving labor efficiency. This will also create new white-collar employment opportunities, making manufacturing more attractive and helping to solve the problem of labor shortages.

Overall Effectiveness

IIoT intelligence facilitates the rapid development of smart manufacturing from the following perspectives

• Capabilities: IIoT Intelligence connects human, cyber, and physical spaces to accurately and synchronously collect industrial data and build effective data analysis models. For example, in the case of the World Economic Forum's Lighthouse Factory, ACG Capsules in India, which adopted IIoT intelligence, achieved a 39% reduction in batch lead time and a 13% improvement in on-time delivery rate.
• OPERATIONAL: IIoT Intelligence creates a cloud manufacturing business model that improves the bottom line for the enterprise. Cloud manufacturing facilitates effective collaboration by encapsulating and centrally managing distributed manufacturing resources as cloud services. For example, China Resources Building Materials Technology Company leverages the cloud to easily leverage data across the enterprise to improve business performance. And Agilent Technologies, Germany, has increased test station throughput by 13% with AI-powered predictive quality testing.
• Deployment: IIoT Intelligence provides a cloud-edge-terminal architecture to efficiently and economically deploy computing resources. In addition, IIoT Intelligence provides efficient production deployment and scheduling suggestions to improve factory output and energy optimization. For example, ACG Capsules used a new AI-based color matching technology and digital twin to optimize production scheduling and improve on-time-in-full rates by 10-20%.
• Application scenario: IIoT intelligence solves specific problems such as predictive maintenance, product design feedback optimization, and product quality inspection. For example, Johnson & Johnson Consumer Health in India implemented IIoT intelligence to achieve predictive maintenance, improve equipment reliability, and reduce unplanned machine downtime by 50%.

IIoT Intelligence Hierarchical Evolutionary Architecture

To gain a systematic understanding of IIoT intelligence, this paper proposes a development architecture that divides IIoT intelligence into five hierarchies. Each of these tiers has a different mission and interacts to achieve smart manufacturing as a whole.

1.Equipment Layer

• Role: Build the foundation for automated production
• Technology: industrial robotics, smart sensors, cloud/edge/fog computing
• Description: Industrial robots partially replace human labor and enable flexible production. Smart sensors provide highly accurate detection of conditions in the production environment and enhance process control. Cloud/edge/fog computing supports processing and analysis of vast amounts of data.

Networking Layer

• Role: Connecting information flows between people, machines, and objects, eliminating information islands
• Technologies: 5G, TSN (Time-Sensitive Networking), SDN (Software-Defined Networking)
• Details: The network layer provides highly reliable, low-latency communications and efficiently transmits real-time data on the manufacturing floor. This enables factory-wide coordination and optimization.

Software Layer

• Role: Provide digital representation of industrial processes
• Technology: Manufacturing Execution Systems (MES), Enterprise Resource Planning (ERP), Product Lifecycle Management (PLM)
• Details: The software layer improves transparency of industrial processes through data collection and analysis to help optimize management and production.

Modeling Layer

• Role: Model physical processes in digital space and connect virtual and physical spaces
• Technology: Digital Twin, Use Case Maps (UCMs), Message Sequence Charts (MSCs)
• Description: Digital Twin creates virtual models of physical entities, enabling real-time state awareness and optimization.

Analysis and Optimization Layer

• Role: Analyze industrial big data and optimize industrial processes
• Technology: AI algorithms, big data analysis, predictive maintenance
• Details: The Analysis and Optimization layer provides algorithms to solve specific industry problems and support real-time decision making.

Figure 4 - IIoT Intelligence Hierarchical Evolution Architecture

Each layer has its own technology and enhances its own capabilities. For example, technological upgrades in the equipment layer strengthen the foundation for automated production, while improvements in the network layer enable real-time transmission of information. This increases the level of intelligence and digitization throughout the manufacturing industry.

Case study

1. ACG Capsules, Inc.

ACG Capsules is an India-based pharmaceutical capsule manufacturer that has leveraged IIoT intelligence to optimize its manufacturing processes. The following results have been achieved

• Leveraging Digital Twin: ACG Capsules implemented Digital Twin technology to create a digital model of its physical manufacturing line. This enabled the company to monitor and optimize manufacturing conditions in real time, reducing batch lead times by 39% and improving on-time delivery rates by 13%.
• Predictive maintenance: Utilizing sensors and AI algorithms, equipment failures were predicted in advance, reducing unplanned downtime by 50%.
• Custom Color Matching Technology: New color matching technology and digital twinning were used to optimize production schedules and improve on-time-in-full rates by 10-20%.

2. Johnson & Johnson Consumer Health, Inc.

Johnson & Johnson Consumer Health leverages IIoT intelligence to improve production efficiency and quality.

• Remote Monitoring: Through IIoT, equipment and environment in the factory were monitored and controlled remotely in real time to improve labor efficiency and safety.
• Automated quality inspection: An AI-based visual inspection system was introduced to significantly improve the accuracy and speed of inspections. This has reduced product defect rates and improved quality.
• Digital R&D: Utilized digital experimentation and modeling to streamline the product development process and reduce R&D costs.

3 China Resources Building Materials Technology Co.

China Resources Building Materials Technology Company leveraged cloud and IIoT intelligence to drive digitization and efficiency across the enterprise.

• Cloud Manufacturing: Distributed manufacturing resources are encapsulated as cloud services and managed centrally for easy enterprise-wide data utilization and improved business performance.
• Increased energy efficiency: Utilizing IIoT sensors and data analytics, energy use was optimized to achieve cost savings and reduce environmental impact.
• Real-time data analysis: Real-time data analysis in the cloud supported manufacturing process optimization and rapid decision making.

4. Agilent Technologies AG, Germany

Agilent Technologies combines AI and IIoT to improve manufacturing efficiency.

• Predictive Quality Testing: AI-powered predictive quality testing has increased test station throughput by 13%. This has streamlined the quality control process and reduced product defect rates.
• Remote Operation: Using IIoT, equipment can be operated and monitored remotely to improve work efficiency and safety.
• Production line automation: Combining industrial robots and AI technology, we have promoted the automation of production lines to reduce labor costs and improve productivity.

These case studies illustrate the tangible effects IIoT intelligence has on the manufacturing industry and highlight how companies are successfully implementing digital transformation.

Technical Studies

There is a wide range of technical research that supports the development of IIoT intelligence. The following is a list of key technologies and their contributions.

1. 5G Technology

• Features: high bandwidth, low latency, extensive connectivity
• Contribution: Enables rapid transmission of real-time data and simultaneous connection of diverse equipment in the manufacturing industry. This increases the flexibility and responsiveness of production lines.

2. time-sensitive networking (TSN)

• Features: decisive packet transport, low latency, high reliability
• Contributions: In IIoT systems, optimize communication between different devices and better manage network traffic. This improves the efficiency of the entire manufacturing process.

3. cloud/edge computing

• Features: Distributed high-performance computing resources
• Contribution: Streamlines data collection, processing, and analysis to support real-time decision making. The combination of cloud and edge optimizes the deployment of computational resources while minimizing data latency.

4. deep learning

• Features: Advanced data analysis and forecasting capabilities
• Contributions: It is used in many manufacturing processes, including quality inspection, predictive maintenance, and process optimization.Deep learning extracts useful patterns from vast amounts of data to improve the efficiency and accuracy of manufacturing processes.

5. blockchain

• Features: Distributed ledger technology, high transparency and security
• Contribution: Improves traceability and data reliability in supply chain management. Blockchain enables fraud prevention and efficient data sharing.

6. big data analysis

• Features: Extracting insights from large amounts of data
• Contribution: Analyze production data, operational data, and customer data to support strategic decision making. Big data analytics plays an important role in predictive modeling and performance optimization.

7. cyber security

• Features: Data protection, ensuring system reliability
• Contribution: Enhances the security of the entire IIoT system and protects it from cyber-attacks. This ensures secure data exchange and operational continuity.

Figure 5 - Key Technologies Supporting IIoT Intelligence and Their Interrelationships

Figure 7 - Digital Twin

These technologies provide the foundation for accelerating the development of IIoT intelligence. It is important to understand how each of these technologies combine to drive the digital transformation of manufacturing. For example, the combination of 5G and TSN enables real-time, highly reliable communications, while cloud/edge computing and deep learning work together to support advanced data analysis and decision making.

Future Issues and Trends

While there are many benefits to the development of IIoT intelligence, there are also many challenges. Overcoming these challenges and understanding future trends are critical to IIoT success.

1. data privacy and security

• Challenge: IIoT systems collect and analyze vast amounts of data, but there is a risk that this data can be misused. In particular, the threat of cyber-attacks and data breaches is increasing.
• Trends: New cybersecurity technologies for data protection need to be developed and deployed. Encryption technologies and distributed data protection mechanisms are gaining attention.

2. lack of standardization

• Challenge: The lack of standardization of IIoT technologies makes compatibility and interoperability between different systems problematic. This makes system integration difficult and inefficient.
• Trends: Standards are being developed and promoted by international standardization organizations. In addition, common protocols and interfaces are being developed through cooperation among companies.

3. skills gap

• Challenge: Implementing IIoT intelligence requires advanced technical skills, but there is a shortage of people with these skills. In particular, there is a growing demand for data scientists and AI engineers.
• Trend: Professional education programs are being developed through partnerships between educational institutions and industry. Efforts are being made to bridge the skills gap through online learning platforms and in-company training.

4. environmental impact

• Challenge: There is concern about the environmental impact of increased energy consumption associated with the introduction of IIoT technologies. In particular, power consumption associated with data center operations has become an issue.
• Trends: The development of energy-efficient technologies and the use of green energy is on the rise. There is a focus on building data centers using renewable energy sources and research on algorithms to optimize energy efficiency.

Figure 6 - Future Issues and Trends

Future Direction

IIoT intelligence will continue to evolve. The following are some of the directions in which this evolution will take place

• Leverage advanced AI and machine learning: Further developments in AI and machine learning technologies will lead to greater automation and optimization of IIoT systems. This will enable more sophisticated predictive analytics and real-time decision making.
• 5G and Beyond 5G Deployment: As 5G technology becomes more widespread, faster and lower latency communications will become possible. In addition, the development of Beyond 5G technology will further expand the possibilities of IIoT.
• Expanding Edge Comp uting: Widespread use of edge computing improves data processing efficiency and real-time responsiveness. This will significantly improve IIoT system performance.
• Adoption of sustainable technologies: Increased adoption of environmentally sustainable technologies is expected to improve energy efficiency and reduce environmental impact.

Further research and implementation of IIoT intelligence along these directions will lead to further digital transformation of the manufacturing industry.

Conclusion

This paper provided a comprehensive perspective on how IIoT (Industrial Internet of Things) intelligence can support and advance the digital transformation of manufacturing. Below is a summary of the main conclusions of this paper.

1. The Value of IIoT Intelligence
   • IIoT Intelligence provides technologies, methods, products, and platforms to build digital connectivity and awareness, intelligent analytics and cognition, and real-time decision making across the manufacturing value chain. This will result in more efficient manufacturing processes, improved quality, and reduced costs.
2. Hierarchical Development Architecture
   • The IIoT intelligence development architecture is divided into five layers: the equipment layer, the network layer, the software layer, the modeling layer, and the analysis and optimization layer. Each tier has a different role and interacts to achieve smart manufacturing as a whole.
3. Key Technology Contributions
   • Technologies such as 5G, time-sensitive networking (TSN), cloud/edge computing, deep learning, blockchain, big data analytics, and cybersecurity form the foundation of IIoT intelligence and digital transformation of the manufacturing industry. Supporting.
4. case study
   • Companies such as ACG Capsules, Johnson & Johnson Consumer Health, China Resources Building Materials Technology, and Agilent Technologies presented examples of how they are using IIoT intelligence to optimize their actual manufacturing processes to great effect. This underscored the practical value of IIoT intelligence.
5. Future Issues and Trends
6. • There are many challenges to deploying IIoT intelligence, including data privacy and security, lack of standardization, skills gaps, and environmental impacts. To address these challenges, cybersecurity technologies must be developed, standards must be promoted, professional education programs must be expanded, and energy-efficient technologies must be adopted.
   • Future trends are expected to include the use of advanced AI and machine learning, deployment of 5G and Beyond 5G, expansion of edge computing, and adoption of sustainable technologies.

This paper provides a pathway for IIoT intelligence to facilitate the digital transformation of the manufacturing industry and provides guidance for future research and implementation. It will help strengthen the foundation for manufacturing to remain competitive and achieve sustainable growth.

• 
• 
• 

Categories related to this article

• Article
• Survey, Review
• Internet Of Things