STATISTICAL PROCESS MONITORING WITH PRINCIPAL COMPONENTS

Most industrial processes are characterized by a system of several var iables, ail of which are subject to drifts, disturbances, and assignab le causes of variation. In the chemical and process industries, there are often inertial forces arising from raw material streams, reactors and tanks that introduce serial correlation over time into these varia bles. This autocorrelation can have a profound impact on the effective ness of the statistical monitoring methods used for such processes. Th is paper reviews some of the available methodology for multivariate pr ocess monitoring and shows the effectiveness of principal components i n this context. An application of the principal components approach wi th correlated observation vectors is presented. The effectiveness of t his procedure to indicate process upsets is discussed.