Defects rule mining for continuous cold rolling mill production lines of Mobarakeh steel company

In today’s competitive world, a bewildering amount of data is being generated. By utilizing appropriate techniques, the knowledge hidden within the data can be uncovered and used as a tool for quality improvements in a business. Data mining presents a collection of efficient tools for knowledge elicitation in data warehouses. Mobarakeh Steel Company, as one of major industries in Iran, has a significant impact on market and economy both in terms of extent and scale. From this point of view, it has advantages over similar companies in the region. One of the most important production lines in the company is the galvanized steel coils production line with a capacity of 20,000 tons per year. In this research, data mining techniques are applied on a database of galvanized steel coils defects in order to extract the governing rules. Having the 10 chemical analysis parameters of the products during 2012 to 2015, the associated rules are obtained. The data used in this study are obtained from the database provided from the quality control department in the Mobarakeh Steel Company. The rules are shown to be effective in modelling the knowledge within the data. In addition, it is shown that not all the defects in the products are rooted in the metallurgic properties recorded in the database. This calls for a deeper investigation for further study the effects of other factors such as human factors and equipment.

کلیدواژه ها:

Continuous production line ، Cold roll mill ، Mobarakeh Steel Company ، Associate rule mining