MMK Implements Automated Digital Quality Control System for Galvanised Cold-Rolled Metal
PJSC Magnitogorsk Iron & Steel Works is conducting pilot tests at rolling mill No. 11 for an automated advisory system for the internal certification of rolled metal produced at the continuous hot-dip galvanising unit. This digital solution, which is based on machine learning and predictive analysis, aims to improve the quality of metal products and prevent possible complaints from consumers.
The system is based on data from a Parsytec surface defect detector, which uses a special scanner and software to analyse the conveyor belt and detect and classify possible defects across its entire width. The detector is able to automatically detect more than 40 defects in metal products, such as holes, rolled-in parts, film, dross and others, and provides the operator with an analytical report.
The advisory system uses information from the detector alongside a mathematical model developed from historical data to assign each batch with the status ‘good’ or ‘defective’. In addition, before the shipment to the customer, the system will assess the likelihood of a complaint. In case of any concerns, the product will be sent for further inspection (visual inspection, examination, etc.), and if the product does not meet the customer’s requirements, the shipment will not be made.
MMK is constantly working to improve customer satisfaction. The advisory system, developed by specialists from MMK-Informservice’s Centre for Mathematical Modelling and Advanced Analytics, significantly reduces the probability of delivering products that do not meet the customer’s requirements, which in turn will reduce the number of customer complaints. During the testing period at rolling mill No. 11, the new solution achieved an efficiency of 91%.