Artificial intelligence helps analyze product quality and reduce the number of potential complaints from customers.
Mining and metallurgical group Metinvest uses artificial intelligence (AI) at the Zaporizhstal plant to control production quality. The company’s press service reported this on Tuesday, January 7.
It is indicated that artificial intelligence recognizes defects in the production of slabs – a metallurgical semi-finished product of which the combined sheets are produced in Zaporizhstal. The working name of the solution is ForgeCheck. Projected savings for the year – $250,000.
The head of the R&D center of Metinvest Digital, Maxim Balanyuk, said that the company has been testing artificial intelligence since 2021: at the plant in Mariupol, the technology worked to determine the quality of scrap metal, analyze the quality of sintering of agglomerate, monitor vehicle flow and audit the movement of roll carriers around the enterprise.
The company began using AI to control the quality of products in Zaporizhstal in June 2024. The solution helps workers quickly identify defects when cutting slabs. If a defect is found, it is highlighted in red on the employee’s monitor, Balanyuk said. If there is a defect, but it is not classified according to the system, it is yellow, and if there is no defect, it is green. The accuracy of the model is 75-85%. But, taking into account the similarity of objects, the model can reach 97%.
The main savings for Metinvest is the reduction in the percentage of complaints (claims – ed.) from customers.
“We also save on electricity because the heated metal does not freeze due to the delay of the slab in the cutting stage,” said Balanyuk.
He did not disclose the amount of investment in this AI solution. But, according to market participants, development from scratch can cost up to $80,000 if open-source technologies are used.
Currently, Metinvest Group’s portfolio includes 12 metallurgical enterprises, some of which are in the occupied territories. Therefore, scaling AI solutions for slab quality control is difficult, as businesses produce different types of products, Balaniuk said.
“The priority for next year is to create and launch the second version of the AI tool,” he added.
According to his calculations, this will help reduce the percentage of complaints by 60%.
“The indicator will depend on the volume of production – if they increase, the savings in financial terms will also increase,” said Balanyuk.
According to him, Metinvest is potentially ready to sell the development to competitors, but first it is necessary to complete the testing of the second version. Proposals are then developed for potential partners working in this area.
Earlier it became known that Metinvest started using artificial intelligence in its businesses to improve safety and control product quality. It is claimed that the system works at the level of a high-class expert.
Source: korrespondent
I’m Liza Grey, an experienced news writer and author at the Buna Times. I specialize in writing about economic issues, with a focus on uncovering stories that have a positive impact on society. With over seven years of experience in the news industry, I am highly knowledgeable about current events and the ways in which they affect our daily lives.