Economic-statistical design of fully adaptive multivariate control charts under effects of multiple assignable causes

Fully adaptive multivariate control charts are efficient to monitor several quality characteristics simultaneously. Charts with assuming a single assignable cause (AC) have been investigated in many studies. However, due to the usual complexity of production processes, the assumption of single AC is not close to real-world conditions. In this paper, by a proper Markov chain approach, we develop an economic-statistical design of a variable-parameter (VP) multivariate control chart to monitor the process mean subject to multiple assignable causes. Numerical examples based on the Taguchi method have been provided and for some parameters, sensitivity analysis is given. Moreover, a comparison between proposed model and fix-parameter (FP) model is done to evaluate the savings of using the adaptive control chart. The results indicate that the proposed VP control charts outperform FP control charts.