Statistical approaches to industrial process plant modelling

The objective of this contribution is to describe the role of the increasingly popular multivariate SPC (MSPC) methods in the context of process engineering. This involves a discussion of some of the techniques used, the application areas in which they are being implemented and some of their limitations which are being addressed in recent algorithm modifications. To illustrate some of the topics raised, a case study is briefly described in which a principal component analysis based approach is used for predictive modelling of an overheads condenser reflux drum system