Non-normality and the variable parameters X¯ control charts

Recent studies has demonstrated the adaptive (i.e., variable sample sizes, sampling intervals, and/or action limit coefficients) X¯ charts are quicker than standard Shewhart (SS) X¯ control chart in detecting process mean shifts. The usual assumption for designing a control chart is that the data or measurements are normally distributed. However, this assumption may not be true for some processes. In this paper, the performance of the adaptive X¯ chart is evaluated under non-normality. Based on the study, it is shown that the variable parameters (VP) X¯ chart is more effective than the other adaptive control charts in detecting small process mean shifts under non-normality. More importantly, the risk of false alarm of the VP X¯ chart can be substantially decreased.