Predicting paper making defects on-line using data mining

This paper describes an application that was jointly developed by Caledonia n Paper and Intelligent Applications for the early prediction of paper defe cts from process data, so that corrective action can be applied before the defect becomes too significant. Correlations between process data and past faults were extracted and then programmed into an on-line predictive softwa re model which is able to analyse current process data in real time, lookin g for bad patterns which may lead to defects in the paper. Depending on the degree of severity of defect that the model predicts, and the nature of th e developing problem, the machine operators can take steps to prevent the d efect from becoming so significant as to result in salvage. This article de scribes the way the application was developed and shows how data mining can be successfully applied to the paper industry. (C) 1998 Elsevier Science B .V. All rights reserved.