عنوان مقاله [English]
In order to monitor univariate auto-correlated processes, many kinds of control charts have been proposed in the literature. However, for multivariate auto-correlated processes, despite of their many applications, control charts have been seldom proposed. In this article, based on a method to reduce auto-correlation in the observed data, a control chart, called Multivariate Grouped Batch Means (MGBM), is proposed to monitor the mean vector of multivariate auto-correlated processes. The parameters of this chart, which is a model-free chart that does not rely on the modeling structure of the data at hand, are optimized based upon a vector auto-regressive of order-one process. Moreover, the performance of the proposed control chart in terms of in-control and out-of-control average run lengths are investigated by a simulation study of a 2-variate process. The result of the simulation study is encouraging.