Intelligent manufacturing needs to solidify the foundation of "data" and "interconnection"
Currently, intelligent manufacturing has become the main focus of China's manufacturing power strategy, and a number of "demonstration enterprises" in intelligent manufacturing have emerged domestically.
What is the essence of intelligent manufacturing? The Fourth Industrial Revolution seems to emphasize the integration of the digital and physical worlds. Intelligent manufacturing is the main manifestation of the Fourth Industrial Revolution in the manufacturing field. So, is the integration of digital and physical worlds the essence of intelligent manufacturing? No, integration is just a phenomenon.
After the second and third industrial revolutions, people's understanding and control of deterministic problems in engineering have matured. There is a lot of uncertainty in manufacturing, whether it is related to efficiency, quality, cost, or green, service, etc. In addition, there are many non patterned problems in manufacturing, such as energy-saving issues in a factory or workshop, which cannot be described by mathematical models and do not have fixed patterns. For humans attempting to have a clear understanding and even control of the manufacturing process, non pattern and uncertainty issues are the biggest challenges. How many factors are interrelated in the system and to what extent do they affect each other?
Fortunately, technologies such as big data and artificial intelligence have opened the door for humanity to further understand and navigate the non patterns and uncertainties of the objective world. Therefore, the essence and essence of intelligent manufacturing is to utilize advanced technologies such as digitization, networking, big data, artificial intelligence, etc. to understand and control uncertainty issues in manufacturing systems in order to achieve higher goals. The reason why Japan's early intelligent manufacturing plans did not achieve significant results may be related to the lack of technology (such as big data and artificial intelligence) to address such issues at that time.
There are two main types of uncertainty problems. One is objective uncertainty, such as the uncertainty of quality during processing and the uncertainty exhibited in product performance. The second is subjective uncertainty, also known as cognitive uncertainty, which mainly refers to the uncertainty in people's understanding of deterministic problems in manufacturing systems that have not been digitized. For example, the arrangement of various activities and processes in enterprises is inherently deterministic, but due to the involvement of too many people and the occurrence of different times, without special means, it is chaotic in people's understanding. This is the subjective uncertainty or cognitive uncertainty of human beings. Why is subjective uncertainty also considered as uncertainty in manufacturing systems? Because the manufacturing system should also include relevant personnel.
What is the key to implementing intelligent manufacturing in enterprises? There are naturally many key issues, but here we only draw people's attention to the most fundamental but not yet widely recognized issue - data and connectivity.
Whether it is objective uncertainty or subjective uncertainty, with corresponding data, there is a foundation for understanding uncertainty. Analyzing data from various aspects of the processing process may enable individuals to clearly identify factors that may affect processing quality that they were previously unaware of; Mastering the relevant data of various activities of the enterprise can reduce the uncertainty of understanding, and make new activities more orderly and corresponding decisions more reasonable. As for non modal scenarios, data analysis is even more necessary.
To obtain data, interconnection is necessary. Interconnection first refers to the collection of physical quantities during the manufacturing process (such as installing them on devices)sensor）By analyzing these data, we may gain a deeper understanding of their uncertainty and discover some correlations between seemingly unrelated physical quantities. The interconnection should also include the relevant data connections between various activities determined by people, provided that the relevant activities are digitized.
The concept of interconnectivity cannot be limited to within a company. We need to have an awareness of the 'enterprise ecosystem', which means that the concept of 'system' cannot be limited to within the enterprise. Suppliers, customers, and other entities form the ecosystem of a company, and members of the ecosystem should have certain data interconnectivity and sharing. There is now a concept of 'digital supply chain', which refers to the supply and demand of not only materials but also data between enterprises. A good enterprise ecosystem should include a 'digital ecosystem', or emphasize data interconnection within the enterprise ecosystem.
The awareness of interconnectivity even prompts entrepreneurs and engineers to redefine the boundaries of industries and the functional boundaries of products. Enterprises that produce smart colored lights need to consider the data connection between the lights and entertainment devices, while the automotive industry needs to consider the data interconnection between cars and smart cities.
The era of human-machine intelligence is coming, what will be the future development of intelligent manufacturing? Currently, it is relatively easy to speculate that "knowledge engineering" will play an increasingly important role, and most of engineers' mental work may be replaced by intelligent systems; The boundary between virtual space and real space will become increasingly blurred, and augmented reality (AR) and mixed reality (MR) technologies will be applied in more scenarios. It is difficult to speculate how the role of humans will be positioned and utilized as manufacturing systems become increasingly "intelligent". We can only explore the answer in practice over time.
What is the best path for intelligent manufacturing? This is a question that the government, entrepreneurs, and R&D personnel should pay special attention to and think about. The current hot topics such as "machine replacement" and "millions of industrial enterprises going to the cloud and implementing millions of industrial apps" are all exploring and practicing in this area. What is the final answer? It is still unknown, but one thing is certain: intelligent manufacturing must rely on numerous basic intelligent technologies and tools, such as sensors, the Internet of Things, intelligent numerical control systems, big data analysis tools, intelligent software, etc. Achieving significant breakthroughs and occupying high ground in these fields is a prerequisite for building a strong intelligent manufacturing country. We must always be aware of this.
Article source: OFweek Industrial Control Network
Author: Industrial Control Pioneer
He Yang/Organizing and recommending
