A Comparative Analysis of Robust Dispersion Control Charts with Application Related to Health Care Data

One of the most efficient tools of statistical process control is the control chart. The efficiency and effectiveness of control charts depend on its in-control robustness, i.e., how the control chart reacts against the violation of the designed model of the chart. The current study evaluates the in-control robustness properties of a chart that is based on the mixture of the statistics of cumulative sum and exponentially weighted moving averages (CS-EWMA) control charts for monitoring the process dispersion under normal, nonnormal, and contaminated normal environments. The in-control robustness performance of the CS-EWMA chart is compared with some existing control charts. Moreover, the appropriate values of the design coefficients for selected charts are also determined. In-control robustness is evaluated in terms of different properties of run length distribution, such as average run length, standard deviation of the run length, and various percentile points. In addition, a real-life application of all the selected charts based on the colonoscopy procedure is considered for practical implementation. It is found that the CS-EWMA chart has a better in-control robustness performance as compared with its counterparts.