Ortalamada kaymalar olduğumda parametrik olmayan CUSUM ve EWMA kontrol kartlarının performanslarının karşılaştırılması

Generally, Shewhart control charts, which require normality hypothesis, are used in monitoring process mean. However, Shewhart control charts may not show adequate performance when the distribution of the process in question does not suit the normal distribution and when there are small shifts in the process mean. For this reason, in cases in which the process distribution is not known, it is more beneficial to use nonparametric control charts that do not require any hypothesis about the distribution. In addition, if there are shifts less than 1.5 sigma, which can be defined as small in the process, preferring the CUSUM (cumulative sum) and EWMA (exponentially weighted moving average) charts, developed as alternatives to Shewhart control charts, would yield more accurate results. In this study, the nonparametric CUSUM control chart and the nonparametric EWMA control chart, designed with the change-point model and the Mann-Whitney Statistic, were introduced and the simulation study was conducted using the R statistical programming language. In this simulation study, data from four different distributions were generated and the average run length (ARL) values for both control charts were calculated. As a result of the calculated ARL values, both control charts were compared in terms of performance and it was observed that the CUSUM chart performed better than the EWMA chart for all distributions applied under the determined conditions.