The Optimization Design Method of Multivariate Control Chart with Adaptive Sample Size

Abstract

ARL (Average Run Length) is used as a tool to measure the performance of control chart. But it isn’t very accurate. In this paper, APL (Average Product Length) is used as a criterion of multivariate control chart performance assessment and a multivariate chart with adaptive sample size is proposed. A Markov chain method is proposed to calculate the APL of multivariate chart with adaptive sample size and then the optimization design method of this chart is discussed. By comparing with traditional fixed sample size design method, we can find that this method have higher efficiency.