Enhanced Nonparametric And Robust Strategies In Statistical Process Control

To examine whether manufactured products meet all their design procedures,control charts are particularly helpful. Manufactured products of a production process are sampled sequentially and by using a control chart,their quality characteristics are monitored over time.Any deviations of the quality characteristics from their design procedures would be signaled as soon as possible,and the root causes of such deviations would be recognized and properly removed. This is usually achieved by adjusting the controllable input variables (cf.Montgomery (2009)). Control charts for variables are generally based on the assumption that the underlying process follow normal parent distribution. However, this assumption is also seldom fulfill in the reality. Therefore, a nonparametric control chart is needed that can be used to monitor the process parameter more efficiently for normal as well as non-normal processes. This thesis contributes some improve nonparametric and robust control charting strategies to be used as add-in for Statistical Process Control (SPC) toolkit. The proposed charting strategies are designed for process location using the ranked set sampling scheme based on the sign test and Wilcoxon signed rank test. The performance ability of the proposals is assessed in terms of some useful measures including average run length (ARL),median run length (MDRL), standard deviation of run length (SDRL) and some percentile points of run length distribution. These performance measures are examined for normal,non-normal and contaminated normal process environments using simple random sampling and ranked set sampling schemes. We have investigated and compare the performance of different proposed charting strategies using extensive Monte Carlo simulations. We have also included some real data sets in order to highlight the practical application of the proposals covered in this study.The results of nonparametric design structures using ranked set sampling indicate that all the proposed charting structures are more efficient than the existing parametric and nonparametric control charts consider in this study in detecting shifts in the process location.Also, these charting strategies are easy to implement and provide the real pictures about the state of process control. The results of the robustness study of various control charts show that the Mixed EWMA-CUSUM chart is very robust to normal, non-normal and contaminated normal processes than other existing location charts. The CS-EWMA chart is also more robust than S2-EWMA and CUSUM-S2 charts to monitor process variability. The results of this thesis are very helpful to the researchers and practioners in designing the nonparametric control chart under ranked set sampling scheme to monitor process location and a robust control charts to monitor process location and process dispersion.