Adaptive CUSUM Control Chart With Variable Sample Size for Monitoring Under Measurement Error

ABSTRACT Measurement error (ME) is a significant factor that affects the performance of Statistical Process Control charts, causing a decrease in their sensitivity to process variations and resulting in inaccurate monitoring. This paper suggests the improvement of the IACUSUM control chart, which takes into account the Variable Sample Size (VSS) method along with a covariate model and multiple measurements to overcome the difficulties associated with ME. The proposed chart is thoroughly evaluated using metrics as Average Run Length (ARL) and Standard Deviation of Run Length (SDRL) both through Monte Carlo simulations and on real‐world data. The results indicate that ME is harmful to sensitivity of the usual SPC charts resulting in delay in detecting shifts. The integrated application of the VSS technique, along with the covariate and repeated measurement techniques, is definitely helping in the display of the chart, enhancing sensitivity, strength, and fast speed in order to identify process changes. This paper has shown how flexible the suggested IACUSUM control chart can be since it is a feasible resolution to the issue of monitoring industrial processes in the context where ME is a regular occurrence.