Design of a new adaptive MEWMA chart to monitor the mean

The adaptive MEWMA (AMEWMA) charts have been broadly perceived as an effective process monitoring instrument due to the efficient detection of the mean shifts of diverse sizes. This study proposes an AMEWMA control chart to observe the irregular variations in the mean vector of the process which follows the multivariate normal distribution. The idea is to assess the mean shift by using an unbiased estimator and after that adapt the smoothing constant of plotting EWMA statistic through a continuous function. As an evaluation tool, the run-length profiles are computed by using the Monte Carlo simulation method. The proposed AMEWMA chart is proved to be efficient in shift detection as compared to the MEWMA and existing AMEWMA charts. The application of the proposed chart is shown by utilizing a real-life dataset from a case study of process capability for turning aluminum pins where six quality characteristics are to be observed.