MULTIVARIATE SPC CHARTS FOR MONITORING BATCH PROCESSES

The problem of using time-varying trajectory data measured on many pro cess variables over the finite duration of a batch process is consider ed. Multiway principal-component analysis is used to compress the info rmation contained in the data trajectories into low-dimensional spaces that describe the operation of past batches. This approach facilitate s the analysis of operational and quality-control problems in past bat ches and allows for the development of multivariate statistical proces s control charts for on-line monitoring of the progress of new batches . Control limits for the proposed charts are developed using informati on from the historical reference distribution of past successful batch es. The method is applied to data collected from an industrial batch p olymerization reactor.