On designing TEWMA-Tukey control charts for normal and non-normal processes using single and repetitive sampling schemes

Abstract

Technology advancement and the need to produce quality products have necessitated the development of more efficient process monitoring schemes. In the form of a combined triple EWMA (TEWMA) and Tukey control chart (TCC), this study is an analogous attempt to develop a better charting framework for normal and non-normal data using single and repetitive schemes. The proposed and competing charts are compared based on average run length, standard deviation of the run length and median run length criteria under both zero-state and steady-state conditions. The proposed charts display dominance in detecting mean shifts in both directions not only for symmetric distributions but they are also robust for skewed distributions in that they are devoid of the average run length biased problem. The proposed charts beat the Tukey and Shewhart charts under the repetitive scheme. The TEWMA-TCC under the repetitive scheme has emerged as the best chart under all distributional setups. The steady-state findings of the TEWMA-TCC and its repetitive variant are as beneficial as those of the zero-state results, if not more. A practical application for the proposed designs in the surveillance of industrial production index data in the aerospace manufacturing is illustrated. The proposed charts issue more and earlier signals in the Phase-II process compared with their existing counterparts, most notably the two Tukey charts, i.e. the TCC and EWMA-TCC. Hence, quality and industrial engineers can rely on the proposed charts in process monitoring.