Russia’s leading steelmaker Magnitogorsk Iron & Steel Works has introduced proprietary information system for predictive analytics in product quality. The digital solution was developed inside the company and uses machine learning to analyze previous test data, enabling specialists to model any technological ideas without the need to spend resources on smelting and rolling steel. The predictive analytics system is a powerful virtual platform that includes a database of technological parameters accumulated over more than 10 years. This includes data on the smelting process, the products manufactured at the hot and cold rolling mills, and the results of the certification process and mechanical testing. The new solution enables automated data processing with the use of machine learning and elements of artificial intelligence, and allows MMK specialists to reduce defects and irregularities, improve the consumer properties of products, clarify standards and adjust technological parameters, put new products into production, increase productivity, reduce costs and much more. The digital modeling platform saves material and human resources while also ensuring that industrial capacities are not used for testing purposes. Specialists are also able to make adjustments to current production technologies to ensure MMK’s products are of superior quality. MMK’s research and development center and experts from MMK-Informservice developed the predictive analytics system using open-source code, which ensures that the IT systems are not dependent on imports and supports the Company’s development of in-house competencies. The system is being considered for the title of “Digital Project of the Year” at MMK’s Digital Olympus corporate competition, with the best technological solutions and developments to be announced on 27 December. The purpose of the award is to promote ideas and initiatives that support MMK’s digital strategy, as well as to identify the best IT projects and support the development of digital transformation leaders among employees.
Russia’s leading steelmaker Magnitogorsk Iron & Steel Works has introduced proprietary information system for predictive analytics in product quality. The digital solution was developed inside the company and uses machine learning to analyze previous test data, enabling specialists to model any technological ideas without the need to spend resources on smelting and rolling steel. The predictive analytics system is a powerful virtual platform that includes a database of technological parameters accumulated over more than 10 years. This includes data on the smelting process, the products manufactured at the hot and cold rolling mills, and the results of the certification process and mechanical testing. The new solution enables automated data processing with the use of machine learning and elements of artificial intelligence, and allows MMK specialists to reduce defects and irregularities, improve the consumer properties of products, clarify standards and adjust technological parameters, put new products into production, increase productivity, reduce costs and much more. The digital modeling platform saves material and human resources while also ensuring that industrial capacities are not used for testing purposes. Specialists are also able to make adjustments to current production technologies to ensure MMK’s products are of superior quality. MMK’s research and development center and experts from MMK-Informservice developed the predictive analytics system using open-source code, which ensures that the IT systems are not dependent on imports and supports the Company’s development of in-house competencies. The system is being considered for the title of “Digital Project of the Year” at MMK’s Digital Olympus corporate competition, with the best technological solutions and developments to be announced on 27 December. The purpose of the award is to promote ideas and initiatives that support MMK’s digital strategy, as well as to identify the best IT projects and support the development of digital transformation leaders among employees.
