Batch multivariate SPC monitoring of a sulfite pulp digester

This paper discusses how principal component analysis (PCA) is used to develop multivariate statistical process control (SPC) charts for a batch process. Historical data from one sulfite batch digester was used to build a reference model. Multivariate batch SPC models are built using two levels, the observation level and the batch level. The observation level is used to monitor the evolution of new batches, while the batch level is used to classify whether the batch is acceptable or rejected.