[2024 Mirae Company Forum] "Steel Industry Meets AI to Create New Value"

Lecture by Lee Deokman, Director of POSCO Holdings AI Robot Convergence Research Center

Lee Deok-man, Head of the AI Enterprise Center at POSCO Holdings AI Robot Convergence Research Institute, is giving a lecture at the '2024 Asia Future Business Forum' hosted by Asia Economy on the 22nd at Lotte Hotel, Jung-gu, Seoul. Photo by Hyunmin Kim kimhyun81@

"The so-called ‘4D tasks’?boring, dirty, dangerous, and difficult?are being reduced through the ensemble of humans and AI. Companies are entering an era where AI lowers costs while enhancing quality, efficiency, and safety."

On the 22nd, Lee Deok-man, Head of the AI Enterprise Research Center at POSCO Holdings’ AI Robot Convergence Research Institute, stated this at the ‘2024 Asia Future Enterprise Forum’ held at Lotte Hotel in Sogong-dong, Jung-gu, Seoul, emphasizing that AI is improving corporate productivity and efficiency.

Lee said, "The AI revolution, known as the Fourth Industrial Revolution, is leading us toward disruptive innovation," and stressed, "Digital transformation has become a new game changer in the industrial sector." He added, "The expertise in specialized industries such as steel can create new value when combined with new ICT technologies," explaining, "Internet of Things (IoT) technology is achieving hyperconnectivity in the smart sensing field, and analysis through big data accumulation is becoming more sophisticated. AI is controlling processes more efficiently than before."

Since 2015, POSCO has been promoting the transition to a smart steelworks. Through the smart platform ‘POSFRAME,’ it collects and structures data from all factories. Utilizing IoT, big data, and AI, the system learns data autonomously to provide and control optimal process conditions. POSCO has also focused on strengthening smart blast furnace technology to automate the previously manual management of blast furnaces. Since 2016, it has been advancing the smart factory step-by-step, converting key variables that determine the blast furnace status into data and turning them into big data. Through this, the temperature of molten iron is digitized via IoT, and high-definition cameras can automatically check the condition of raw materials.

Lee stated, "Currently, hundreds of AI models are applied at POSCO’s steelworks," adding, "In the steelmaking process, the temperature of the blast furnace is predicted, and the amount of pulverized coal injection (PCI coal) into the hot blast stove is controlled. Technologies optimizing load distribution have also been introduced in the rolling process." He further explained that a braking control system was built through deep learning and is currently operating at the No. 2 blast furnace in the Pohang Steelworks. The smart steelworks have also enhanced worker safety. POSCO’s developed smart CCTV automatically detects specific characters, shapes, and movements at the steelworks site. When abnormalities are detected from the collected information, it alerts the managers. In particular, it prevents fires by detecting risks in advance using multi-video devices such as thermal imaging.

Lee explained, "Previously, operating a blast furnace required long training and relied solely on the insights of skilled workers," and added, "Now, by accumulating big data on the operating know-how of skilled workers, operation controlled by five predictive models and AI is possible." Through this, POSCO reports a productivity increase of about 5%, a 1% reduction in costs, and a 1% decrease in carbon dioxide emissions compared to before.

The Pohang Steelworks, which worked hard on the smart steelworks transition, was selected as a ‘Lighthouse Factory’ by the World Economic Forum (Davos Forum) in 2019. Lighthouse Factories are factories that will innovatively lead the future of global manufacturing. POSCO was the first in South Korea to be named.

Lee highlighted that collaboration with small and medium-sized enterprises (SMEs) and universities was crucial in applying smart factory technology at POSCO. He said, "Looking back to 2016, the early stage of AI technology application, there was no infrastructure, but universities took charge of AI algorithm development and smart factory idea development, while SMEs focused on IoT sensor and software development," and added, "The division of labor between POSCO’s accumulated expertise in the steel sector and SMEs and universities was a key success factor in establishing the smart factory."

POSCO has also developed and applied smart technologies in the safety field. The ‘Smart Safety Ball’ is a representative example. The Safety Ball is a tool that can detect the presence of harmful gases in confined spaces in advance. It is easy to use by linking with smartphones and can simultaneously measure oxygen, carbon monoxide, and hydrogen sulfide, making it highly useful for preventing safety accidents, according to Lee.