CONICET | Buscador de Institutos y Recursos Humanos

Quality control and safety related issues have become more and more important in industrial productionof high added value products and chemical specialities during last years. In this regard, many successfulapplications of multivariate statistical process control (MSPC) for monitoring and diagnosis of batch processeshave been presented. It is a common industrial practice to monitor the batch progress by exploitingthe information contained in a historical database of successful batches using projection techniques suchas principal components analysis (PCA), partial least squares (PLS) and independent component analysis(ICA). In this work, a new MSPC strategy for batch process monitoring is presented. Its distinctive featureis that it works in the space of the original variables. The technique uses only the T2-statistic fordetection and identification purposes. The identification of the set of observations that signal the fault isaccomplished by decomposing the T2-statistic as a uniquesumof each variable contribution. Performancecomparisons among the proposed strategy and the most popular PCA-based approaches are carried outby simulation of polymerization and penicillin cultivation batch processes. Results show that the newapproach can be successfully applied to monitor this kind of processes since it works very well duringboth fault detection and identification stages.