An Enhancement for the Multivariate EWMA Chart

A quality control chart is one of the tools that is used to solve quality control problems. A control chart is used to identify the causes of a problem. Quality control for most processes involves more than one correlated variable. This is known as multivariate quality control. A multivariate EWMA (MEWMA) control chart is a good chart to detect a small shift in the mean vector of a multivariate process. This paper suggests the use of the maximum likelihood method on the MEWMA chart for the detection of the actual time of the occurrence of a permanent shift in the mean vector. The use of a time step change estimator on a MEWMA chart will benefit quality control engineers in process monitoring in industries.