China Market Insider – Series "China's AI Factories" Part 3 How AI Factories Are Transforming the Manufacturing Industry: Example CATL

Related Vendors

Asia Waypoint

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

SIMTOS (Seoul International Machine Tool Show)

China uses AI for tangible improvements in industrial production. In part three of our series, we illustrate this with the example of CATL.

It is already the fourth AI factory in China that CATL is currently building in Zhongzhou. Artificial intelligence plays a crucial role in the production of batteries here. Just like the first three, which have already been recognized as "Lighthouse Factories" by the World Economic Forum.

Around 6,000 workers are currently working around the clock on the construction of the factory near Luoyang in Henan Province, reports the portal "Pengpai Xinwen." The first and second construction phases are already operational. The last two have already celebrated their topping-out ceremony and are currently being equipped with machinery.

Recently, it was humanoid robots that brought the CATL factory in Zhongzhou into the headlines. The particularly flexible bipedal robots of the "Xiaomo" type, powered by AI, take on complex tasks in battery production, including EOL tests (End of Line) and DCR measurements (Direct Current Resistance). This is a world first.

However, while humanoids often dominate the headlines, their new workplace is at least equally spectacular. In the new CATL factory, quality inspection for new batteries has been so greatly improved using AI, high-precision 3D cameras, and edge computing that entire traditional process steps have become obsolete.

Not much has been reported about the Zhongzhou factory so far. However, those already interested in the details of how production is carried out there can take a look at CATL's first three AI factories. For instance, the factory in Liyang in Jiangsu Province, which was included in the Lighthouse Factories of the World Economic Forum (WEF) in mid-December.

"During the welding of sealing pins, a defect detection rate of 100 percent is achieved, eliminating the need for helium leak testing," writes CATL on its website about production in Liyang. The defect rate has been moved from the "per million" range to the "per billion" range, reported the WEF.

Another example where AI and big data are used is the capacity testing and calibration of batteries on the production line. CATL reports that with its new "intelligent capacity estimation system," both the duration and energy consumption of this production process can be reduced.

A third example of AI application is virtual simulations and verifications during equipment retrofitting. Since the technology of lithium-ion batteries is developing "very rapidly," the quick replacement of machine tools is an important factor in ensuring customer satisfaction, writes CATL.

Previously, this could take several weeks. Now, the time has been significantly reduced with the help of AI, digital twins, and 3D printing. The design of new machines has been accelerated to such an extent that "commissioning and debugging cycles have been shortened by 40 percent," according to CATL.

There are further examples of how AI is used to optimize processes in CATL's modern battery factories. With the help of machine learning, the process control of several production steps as well as energy management are improved.

Together with other technological innovations such as big data and additive manufacturing, "production output was increased by 320 percent and manufacturing costs were reduced by 33 percent" at the CATL factory in Liyang, writes the WEF in its justification for awarding the title "Lighthouse Factory."

All these innovations are once again being utilized in the 31 buildings of the new factory in Zhongzhou. What is emerging here on 700,000 square meters, equivalent to the size of 98 football fields, is furthermore the center of a new "Science & Technology Cluster" with dozens of companies from upstream and downstream supply chains, research laboratories, and training facilities. Thus, the term AI factory almost falls short. It is more like an "AI cluster" for modern battery production.