Designing a new adaptive multivariate control chart for simultaneously monitoring mean and variability of process under effects of multiple assignable causes

There have been some advances in multivariate control charts with the ability to monitor both the mean and variability of processes. However, due to the complexity of production processes, the assumption of single assignable causes is not close to the real-life conditions. As a novel contribution, this article proposes a new control chart for monitoring the mean and variability of a multivariate normal process which is under effects of multiple assignable causes. We develop a Markov chain model to compute the average run length and average time to signal (ATS) values. We also make it fully adaptive by varying all control chart parameters. The presented model involves complex non-linear models with a mix of continuous and discrete decision variables, and discontinuous, non-convex solution spaces. Therefore, one of the most suitable metaheuristic search approaches, Genetic algorithm is implemented. Numerical examples based on the Taguchi method are presented and sensitivity analyses are conducted to measure the performance of the proposed chart.