MULTIVARIATE STATISTICAL PROCESS CONTROL; EXAMPLE FROM THE CHEMICAL PROCESS INDUSTRY

摘要

This paper describes an application of multivariate statistical methods with the aim to improve the production of titanium dioxide at Kronos Titan AS. Multivariate statistical methods were used to make a PLS model of one process stage. This model was then used to predict the product quality as a function of the chemical variables analysed in the laboratory. New process data were plotted using multivariate charts. An example shows trends and deviations from the normal operating region. Probable reasons for the deviations are indicated.