On monitoring the multivariate process variability

Abstract

To effectively monitor general changes of multivariate process variability, a new control chart is proposed and studied. The proposed control chart is constructed based on taking the exponentially weighted moving average of the logarithm of the likelihood ratio for testing the hypothesis that two variance- covariance matrices are equal. The applicability and performance of the proposed control chart in detecting changes in multivariate process variability are demonstrated through examples and simulations. (This is a joint work with Arthur B. Yeh and Yu-Fang Wu. )