A likelihood-ratio-based EWMA control chart for monitoring variability of multivariate normal processes

Abstract

To effectively monitor small changes of variability of multivariate normal processes, a new control chart is proposed and studied. The proposed control chart is constructed based on taking the exponentially weighted moving averages of the logarithm of the likelihood ratio for testing the hypothesis that two variance-covariance matrices are equal. The applicability of the proposed control chart in detecting changes in process variability is demonstrated through an example. The simulation studies further show that the proposed control chart outperforms the existing procedures in most cases.