Monitoring the variability of beltline moulding process using wilks’s statistic

Abstract

Due to the increase of the complexity of customer demand on products and services, monitoring process quality is becoming multivariate in nature. In this setting there are two important parameters to be monitored, i.e., the mean vector and the covariance structure which determines the variability of the process. This paper deals with process variability monitoring of beltline moulding process at an automotive industry where the process is in multivariate setting and monitoring process is based on individual observations. Our approach is based on Wilks’s statistic. A real application will be presented and the strength of that statistic, as well as its limitations, will be discussed.