APPLICATION OF DATA MINING TECHNIQUES TO FIND CORRELATION BETWEEN QUALITY DATA AND PROCESS VARIABLES

摘要

The enhancement of product quality poses a constant challenge in the steel industry. This paper is intended to demonstrate that data mining methods can be successfully applied in order to optimise existing processes and plants in terms of the achievable product quality. The starting point for such an optimisation is to make it possible to use data archives from various process stages that contain valuable information covering the entire process chain and thus the history of the product. In addition to the IT challenges that have to be solved, the work steps required to analysis the data and subsequently find correlation are of crucial significance. Both of these will be explained here. It becomes clear that, by applying modern data analysis methods, it is possible to arrive at meaningful information about the dependencies between quality data and process variables.