Xinhua News Agency, Beijing, September 14th Title: Observation on the development of smart factories in the first year of the “15th Five-Year Plan”
Xinhua News Agency reporters Tang Shining, Li Qianwei and Li Ziwei
The robotic arm rises and falls flexibly, autonomously and efficiently completing processes such as grabbing and assembly; the digital twin system builds a “mirror copy” of the production line, and key data is clear at a glance, enabling dynamic deduction and optimization of the entire production process; “AI quality inspectors” use algorithms to automatically identify appearance defects, dimensional deviations and other issues, achieving millisecond-level high-speed inspection… Intelligent applications have given a new look to the factory floor.
The “15th Five-Year Plan” outline proposes to promote the “intelligent digital transformation to network connection” in the manufacturing industry and implement intelligent manufacturing projects and industrial Internet innovation and development projects. Smart factories are the key to realizing smart manufacturing. The reporter’s investigation found that currently, smart factories in various places are deeply integrating big data, artificial intelligence and other technologies to inject new momentum into the industry to improve quality and efficiency and reshape competitive advantages.
Accelerating the release of economies of scale
Walking into the Shuangliu Smart Factory of Sichuan Tianwei Food Group Co., Ltd., the robots at the smart batching station automatically complete the ingredients according to the process formula, the AGV cars shuttle back and forth to send the materials to the next process, and the robotic arms accurately put the raw materials into the reactor for cooking… Packets of hot pot base ingredients are being “refined” along the automated production line to create a delicious flavor.
Here, the intelligent operating system is like equipping the factory with a “smart brain”, achieving 100% digital management from ingredients, frying to packaging and warehousing, and establishing an accurate traceability system from raw materials to products. In the past, the heat and duration of stir-fried ingredients were all controlled by the “experience of old masters”; now, the entire intelligent production process has made product quality more uniform and stable.
“Relying on intelligent transformation and upgrading, the single-line production capacity of our production lines has increased by more than 2 times, and the per capita efficiency has increased by more than 6 times.” said Yang Yan, public affairs director of Tianwei Food.
This is a vivid microcosm of the vigorous development of smart factories in our country. As an important carrier of the integration and innovation of new generation information technology, advanced manufacturing technology and advanced management concepts, smart factories are increasingly becoming the main battlefield for digital transformation and intelligent upgrading of the manufacturing industry.
The reporter’s investigation along the way revealed that from industrial robots, intelligent testing and processing equipment, and air logistics chains to continuously improve production efficiency, to industrial Internet platforms breaking down information islands, smart factory construction is promoting all-round changes in industrial innovation paradigms, production methods, and management models.
In recent years, my country has carried out in-depth smart factory gradient cultivation actions at four levels – basic-level and advanced-level smart factories focus on guiding enterprises to popularize and apply intelligent manufacturing equipment, industrial software and system solutions, and consolidate basic digital and network capabilities; excellence-level and pilot-level smart factories focus on forward-looking layout, promote the in-depth application of new technologies such as artificial intelligence and digital twins, explore future manufacturing models, and lead the intelligent transformation of industries.
The latest data shows that my country has built more than 56,000 basic-level, more than 9,000 advanced-level, and more than 500 excellence-level smart factories, and has cultivated 15 pilot-level smart factories.
“With scenario-based implementation as the starting point, the scale effect is accelerated under the four-level linkage of smart factories, key industries are expanding to a deeper level, and the integrated innovation of processes, equipment and software continues to accelerate.” said Hao Shimeng, an engineer at the Institute of Integration of Informatization and Industrialization of the China Academy of Information and Communications Technology.
Deeply reconstruct the wisdom “gene”
Intelligent technology is accelerating its evolution, and smart factories are also continuing to be upgraded. The reporter learned that the deep integration of AI has enabled smart factories to continuously improve on their original basis, upgrading from “looking at data” to “managing production lines”, “optimizing design” and “strong ecology”, and are deeply reconstructing the “gene” of intelligence.
Before the steel coils enter the rolling mill, artificial intelligence has already begun to “arrange troops.” On the high-efficiency integrated cold rolling production line of Taigang Stainless Steel Cold Rolling Plant in Baowu, China, automatic unbaling robots, corrugated paper robots and other relay operations remove the straps and protect the steel coils. In the remote intelligent control center, AI analyzes the robot movements and production line conditions in real time. The next batch of steel coils has been grabbed by the unmanned overhead crane and moved quickly.
The monitoring covers the operating status of 26 robots in the entire production line. The data flow and steel coils “run” synchronously. The AI system protects the entire process. Li Shengxiong, head of the tandem cold rolling mill at Taigang Stainless Steel Cold Rolling Plant, told reporters that thanks to the application of intelligent algorithms, the cold rolling industry brain has realized refined management such as automatic identification of on-site problems and intelligent supervision of work behaviors.
Currently, the new generation of artificial intelligence technology represented by large models, agents, and embodied intelligence is developing rapidly and accelerating its integration into the manufacturing system, bringing new opportunities for the manufacturing industry to leap from digital to intelligent.
Reporters conducted in-depth investigations on factory floors in various places and found that AI has become a key variable driving industrial upgrading——
In Shandong, Weichai Power Co., Ltd.’s high-end engine smart factory has created “AI designers” that can independently complete more than 1,800 types of research and development tasks, shortening simulation reports that took eight hours to complete in the past to just a few minutes;
In Hubei, Wuhan BOE Optoelectronics Technology Co., Ltd.’s 10.5-generation LCD panel smart factory, the AI defect management platform supports the introduction of factory-wide precision inspection equipment, integrating edge computing and super-resolution algorithms to achieve accurate identification of micron-level defects;
In Jiangsu, Xuzhou Heavy Machinery Co., Ltd.’s mobile crane smart factory has built a dedicated computing cluster to create five types of industrial intelligence, including product development and production management, to promote intelligent collaboration and efficiency improvement in key business links…
According to statistics, in my country’s leading smart factories, artificial intelligence has penetrated more than 70% of business scenarios, accumulated more than 6,000 vertical field models, and accelerated its in-depth penetration into core production scenarios such as process optimization, equipment management, and human-machine collaboration.
Continuous efforts to expand boundaries
The outline of the “15th Five-Year Plan” plans to promote technological transformation and upgrading and develop intelligent manufacturing, green manufacturing, and service-oriented manufacturing. Where is the breakthrough to promote the development of smart factories?
To promote the “two-way empowerment” of artificial intelligence technology and manufacturing applications, the “Implementation Opinions on the “Artificial Intelligence + Manufacturing” Special Action” released by eight departments including the Ministry of Industry and Information Technology at the beginning of this year sent a clear signal –
By 2027, my country will achieve safe and reliable supply of key core technologies for artificial intelligence, launch 1,000 high-level industrial intelligence agents, and create 100 high-quality data sets in the industrial field.
To make smart factories more “smart”, we need to continue to make efforts in several aspects.
At the decision-making level, improve the independent global scheduling capabilities of large model-driven factories. The relevant person in charge of the Science and Technology Department of the Ministry of Industry and Information Technology stated that it is necessary to promote the deep integration of large-scale model technology into the core links of production and manufacturing, and transform the entire process of R&D design, pilot verification, production and manufacturing, marketing services, and operation management.
At the data layer, we will accelerate the construction of high-quality data sets in the industrial field. Next, we need to take the needs of typical scenarios as a guide, focus on the “small cuts” of key scenarios in key links, and strive to break through the bottlenecks of “acquisition”, “gathering” and “use” of industrial data.
At the execution level, we will further promote intelligent agent-empowered flexible production. At present, a number of explorations are accelerating. For example, Jikrypton Smart Factory uses flexible assembly systems, AGV scheduling, etc. to achieve seamless connection and automatic assembly of car bodies of different models and configurations on the same production line.
“Looking to the long term, factories are no longer just manufacturing workshops, but industrial ‘intelligents’ that continue to grow iteratively.” Li Yi, head of supply chain management of Jikrypton Smart Factory, said that the company will continue to solve the production problem of “thousands of vehicles and thousands of faces” of new energy vehicles and achieve in-depth collaboration between customization and scale.
