Industry 4.0 and the Industrial Internet of Things
Industry 1.0, with its steam-powered mechanized production, is considered the beginning of industrialization.
Industry 2.0: Large-scale production powered by electricity
Industry 3.0 is characterized by industrial controllers, enabling automated industrial production.
Industry 4.0 Intelligent Production and Control
The development of the entire industry is accompanied by increased system complexity and the threat of cyberattacks, but the benefits include reduced costs and less reliance on single suppliers. This is the progress of the entire industrial revolution.
The Internet of Things (IoT) encompasses the Industrial Internet, including people, things, and even broader spaces. Industry 4.0, viewed horizontally, also includes the Industrial Internet, but it covers the ecosystem, emphasizing the interconnectivity of the entire industrial sector. Currently, smart manufacturing and the Industrial Internet are using Industry 4.0 as a starting point, with the entire Industrial IoT data as its core, including data interaction, computation, and processing.
2 Industrial Internet of Things
Information chain of Industrial Internet of Things
IBM and Deloitte proposed a model for industrial data visualization. In the entire industrial production process, data is first acquired through sensors—this data is unprocessed. Secondly, this data is transmitted and standardized for interaction; this standardized data is called information. By aggregating data from different sources and analyzing it to explain the reasons behind them, predictions are made—this is knowledge programmed into information. Finally, this data helps guide industrial equipment to trigger specific behaviors, achieving functional adjustments and optimizations—this is called "intelligence."
Challenges arising from the information chain of the Industrial Internet of Things
In 2014, IBM proposed a framework diagram: in the entire physical system, data is called discrete elements, and after being correlated, it becomes information. Organized information becomes knowledge, and intelligentization becomes wisdom. The challenges arising from the information chain of the Industrial Internet of Things (IIoT) accompany the entire lifecycle.
Challenges
Challenge 1: Crossing the boundaries between OT and IT
IT stands for Information Technology, and OT stands for Manufacturing Technology. In the automotive manufacturing industry, it is evident that OT technology is prevalent. As OT technology develops, it will become apparent that if these cars are to be interconnected, and if the data of automotive companies and factories is to be optimized, then it will need to be integrated with IT technology.
Challenge Two: Information and Data
There are many industrial buses, and collecting and interoperating them is difficult. Currently, the Industrial Internet of Things (IIoT) hopes to solve this challenge through Industrial PASS.
Challenge 3: Knowledge
The information is inaccurate and incomplete. We hope to use intelligent technology to solve the difficulties in constructing and organizing knowledge caused by the inaccuracy and incompleteness of this information.
Challenge Four:
How to handle end-to-end collaboration throughout the entire industry cycle.
Reference architecture for the Industrial Internet of Things
A Reference System for the Internet of Things in Industrial Systems (from the White Paper on Industrial Internet of Things in China): The Industrial Internet of Things encompasses various domains, including user objects, service objects, sensing and control, and the physical world. In the resource allocation domain, various logistics and financial institutions are involved; in the management and operation domain, the government needs to supervise industrial safety, thus leading to the development of various management and operation and maintenance platforms.
Industrial Internet of Things (IIoT) network architecture
The network architecture of the Industrial Internet of Things (IIoT) can be divided into the sensing layer (data acquisition), the network layer (various communication protocols, including the application of industrial Ethernet, OPC, 4G and even 5G technologies), and the application layer (various service support).
Application models of Industrial Internet of Things
Industrial data has given rise to a variety of R&D tools and production processes. Data processing has led to various analytical models, requiring me to use intelligent technologies to perform causal analysis on these models and correlation analysis using data-driven models. This, combined with application services (overall architecture, microservices), addresses state awareness, real-time analysis, subsequent decision-making, and precise execution.
Industrial Internet of Things (IIoT) Technology Trends
The technological trend of the Industrial Internet of Things (IIoT) is terminal intelligence, with various edge computing technologies being proposed, aiming to complete computations on the terminal itself. Network ubiquity, various Ethernet protocols, industrial protocols, and 5G technology are also beginning to be applied in industrial production. Furthermore, there is the trend towards edge computing, network flattening, and service platformization.
Industrial Internet of Things (IoT) security scope
Safety must be considered in the technological development of the Industrial Internet of Things (IIoT). The entire IIoT process must consider two aspects: functional safety and information technology.
Functional safety
The physical unit under consideration poses no unacceptable risks to the outside world.
For example, if a car's braking system fails when the brakes are applied, this poses an unacceptable risk to the outside world and must be avoided.
Information security
The physical units under consideration do not have any unacceptable risks from the outside.
For example, the Stuxnet virus outbreak in Iran a few years ago caused the shutdown of nuclear power plants. For Iran, this is an unacceptable risk from the outside.
Functional safety vs. information security
Functional safety and information security are actually merging. In the automotive field, functional safety is considered a category of information security, ultimately achieving the integration of information security with functional safety.
The convergence of functional safety and information security involves many factors:
• System boundaries are difficult to define effectively
• The interactive environment is difficult to predict effectively.
Under this trend, two principles cannot be violated:
Information security must not compromise the effectiveness of industrial security.
The mechanism by which functional safety takes into account information security
These analyses reveal that all information security vulnerabilities stem from flaws in the system's design. The Heartbleed vulnerability, for example, fails to require a function to match the string content with its length. This allows users to exploit the vulnerability by specifying a very long string and extracting all subsequent parts of it. From a software engineering perspective, this is a functional safety vulnerability.
Therefore, information security ultimately boils down to functional safety. This is because when designing functional safety, the boundaries are difficult to define, the external environment is unknown, and it is impossible to take all environmental factors into account.
Industrial Internet of Things Security Requirements
The security requirements of the Industrial Internet of Things (IIoT) need to be integrated. In the design phase, the functionality needs to be verifiable; in the usage phase, the ability to handle abnormal failures needs to be ensured; in the certification phase, the guarantee needs to be provided through some international standards; and in the information security phase, the entire pre-emptive protection package (including the ability to perceive dangerous situations) needs to be in place, preventable when attacked, and able to be handled after an attack. This is the controllability of security incidents.
Functional Safety: Test and Verification Toolchain
Information security: control + network + application
Information security forms the framework of the entire Industrial Internet, which is a national strategy. The Industrial Internet consists of four layers: network layer, data layer, security layer, and application layer.
There are various types of security layers, including data security, application security, network security, control security, and device security.
Information security practice cases
If the scope of the Industrial Internet of Things (IIoT) is expanded, the safety of products manufactured in industry could even be included within its scope.
Safety Elevator Network
Professor Pu Geguang's team surveyed many elevator manufacturers and found that the entire elevator industry needs technological innovation. Elevator manufacturers are internally working on elevator interconnection, including the diagnosis and integration of elevator equipment.
Safe vehicle networking
The success and security of the Internet of Vehicles (IoV) are inseparable. The entire automotive industry is advancing automotive information security, and information security will form an industry standard in the next two to three years, gradually becoming an international standard. On the vehicle side, this involves various in-vehicle security measures; on the network side, it involves operators; and on the cloud side, it involves companies like Alibaba and Huawei.
Industrial IoT APP
The Industrial Internet, or the Internet in general, has various apps, and how to ensure the security of these apps is something that needs to be explored.
3 Future Trends of Industrial Internet of Things
Future Trends of the Industrial Internet of Things as Seen from Gartner's Emerging Technologies Hype Cycle
Digital twin
Security technology development trends
Digital twin
This approach dynamically presents the past and present behaviors or processes of physical entities in a digital form, enabling secure and consistent modeling of physical and digital spaces.
A digital twin is a fully digital simulation of a complex system, capable of simulating scenarios that occur before and after the design process, during product operation. Therefore, digital twins are more secure, as security factors can be incorporated into them, thus providing guidance for the entire real-world system.
AI
Artificial intelligence can also be deeply applied in the security field. For example, in the process of constructing honeypots, some fake network scenarios are used to make attackers believe that it is a real device. The construction of honeypots requires AI-based attack and defense collaborative training, which can create more intelligent honeypots.
Software-defined
Software-defined technologies also require security integration. In real life, various security strategies are employed, and many people now use software-defined security technologies to automatically build intelligent combinations of various protection technologies.
Blockchain technology
Blockchain is primarily used in finance. However, the most important concepts of blockchain are distributed consensus mechanisms and data governance. For industrial big data, some data is highly sensitive. After an attack, is there a way to know if the data has been tampered with? Therefore, blockchain technology can be used to build a trust system for the entire industrial big data system.
The entire industrial control and functional platform needs to focus on and develop next-generation technologies, thereby driving the industrial ecosystem forward through technological advancements.
