Abstract:
The statistical process control (SPC) is a set of statistical tools and techniques used to monitor and control the production process. Under SPC, one of the prompt tools is control charts. A graphical presentation of the change in a process over time is called a control chart. A control chart is one of the important industrial tools used for monitoring the stability of manufacturing processes. Measurement error is a usually met distortion factor in real world applications that impacts the outcome of a process. The performance of the control charts can be affected in the presence of measurement error (M.E), which also leads to erroneous conclusions regarding the average run lengths. Measurement error plays an important role in the quality control process and adversely affects the shift detection ability of control charts. In the present study, we proposed memory-type exponentially weighted moving average and mixed exponentially weighted moving average-cumulative sum by using auxiliary information (A.I) to monitor the process mean and the effect of M.E on memory-type control charts is examined by using (i) covariate method (ii) multiple measurements (iii) linearly increasing variance method. However, the M.E has a adversely effect on the performance of the control charts in terms of increasing the run length properties such as average run length and standard deviation of run length. Multiple measurements method is considered as a remedy to minimize the effect of measurement error available in the literature. We also studied the A.I based exponentially weighted moving coefficient-of-variation control chart with simple random sampling. The sensitivity of a control chart can be increased by improving the sampling method such as ranked set sampling (RSS). Further, the EWMA based coefficient-of-variation and simultaneous monitoring of process mean and CV control charts are proposed under RSS. Moreover, the effect of M.E is evaluated on simultaneous monitoring for mean and CV control chart.
