We will continue to focus on developing the real economy and adhere to the direction of intelligence, greening and integration.

At 3:00 PM on October 22nd, images of defects on a coating repair machine flashed across the control room screen at the 8.5-generation LCD panel factory of Suzhou Huaxing Optoelectronics Technology Co., Ltd. in Jiangsu Province. "This is a defective product on the production line. In the past, technicians had to repeatedly identify and confirm this. Now, no manual intervention is required; our 'digital employees' will handle it," said Zhang Dongliang, Public Affairs Manager of Suzhou Huaxing.

As they spoke, the "digital worker" sprang into action. First, they compared the real-time product image with a database of defect samples, quickly locating a breakpoint in the panel circuit. Then, they issued an instruction, and the equipment in the workshop immediately performed "laser surgery" on the product. In seconds, the breakpoint was repaired.

Today, Suzhou Huaxing has "cloned" thousands of devices and every pipeline in the workshop into the electronic screen of the central control room, and they are managed intelligently by the "digital brain".

"The production of an LCD panel involves more than 200 steps. It is not easy to tailor a 'digital brain' for such a complex production line," said You Wei, the company's director of intelligent manufacturing.

In 2021, to reduce costs, increase efficiency, and enhance competitiveness, the company decided to digitally transform this large-scale production line. They assembled an artificial intelligence project team consisting of dozens of experienced engineers, with You Wei as the lead. "We encountered difficulties from the very beginning," You Wei said. Although he had extensive production management experience, he had little knowledge of digital algorithms. The talent recruited by the project team, while proficient in algorithms, was a complete outsider to panel production processes.

To solve this problem, You Wei delved into digital algorithms. Gradually, the intelligent solutions he proposed became more and more specialized, and everyone immediately understood and collaborated seamlessly.

The transformation process was a journey of ups and downs. You Wei recalls that when the "digital employee" was first put online for testing, its performance was less than satisfactory. For example, when black spots appeared on defective panels, the "digital employee" often couldn't tell whether they were foreign objects or broken wires.

To make the "digital employees" smarter, the project team continued to optimize the algorithm while feeding it large amounts of scenario data. After trying thousands of solutions, the system was finally debugged to the optimal state, able to accurately determine the causes of microscopic defects in various panels and make repair decisions. "Sometimes it's more reliable than the old masters," said You Hao, the company's production director. "Some time ago, the system analyzed the torque, vibration, pressure and other multi-dimensional real-time data of an important equipment, discovered the risk of equipment failure and issued an early warning, allowing us to intervene in advance and avoid losses."

At Suzhou Huaxing, the application of artificial intelligence throughout its R&D, design, production, and equipment maintenance processes is like having a smart brain, sharp eyes, and nimble hands. Compared to before the transformation, the company's yield rate has increased from 95% to 98%, operating costs have decreased by 7.74%, and overall employee productivity has increased by 123.91%. "Before the production line transformation, the production cycle from a glass substrate to an LCD panel took at least seven or eight days. With the 'digital brain,' it can be completed in four days," said Zhang Dongliang.