Comparison of statistical process monitoring methods: application to the Eastman challenge problem

Multivariate statistical process control (MSPC) has been successfully appli ed to chemical processes. In order to improve the performance of fault dete ction, two kinds of advanced methods, known as moving principal component a nalysis (MPCA) and DISSIM, have been proposed. In MPCA and DISSIM, an abnor mal operation can be detected by monitoring the directions of principal com ponents (PCs) and the degree of dissimilarity between data sets, respective ly. Another important extension of MSPC was made by using multiscale PCA (M S-PCA). In the present work, the characteristics of several monitoring meth ods are investigated. The monitoring performances are compared with using s imulated data obtained from the Tennessee Eastman process. The results show that the advanced methods can outperform the conventional method. Furtherm ore, the advantage of MPCA and DISSIM over conventional MSPC (cMSPC) and th at of the multiscale method are combined, and the new methods known as MS-M PCA and MS-DISSIM are proposed. (C) 2000 Elsevier Science Ltd. All rights r eserved.