Comparing alternative approaches for multivariate statistical analysis of batch process data

Batch process data can be arranged in a three-way matrix (batch x variable x time). This paper provides a critical discussion of various aspects of th e treatment of these multiway data. First, several methods that have been p roposed for decomposing three-way data matrices are discussed in the contex t of batch process data analysis and monitoring. These methods are multiway principal component analysis (MPCA)-also called Tucker1-parallel factor an alysis (PARAFAC) and Tucker3. Secondly, different ways of unfolding, mean c entering and scaling the three-way matrix are compared and discussed with r espect to their effects on the analysis of batch data. Finally, the role of the time variable in batch process data is considered and methods suggeste d to predict the per cent completion of batch runs with unequal duration ar e discussed. Copyright (C) 1999 John Wiley & Sons, Ltd.