Enhanced CUSUM control charts for monitoring Coefficient of Variation: A case study in Textile industry

The recent blooming developments of Artificial Intelligence (AI), Internet of Things (IoT), and Data Science (DS) have put Smart Manufacturing (SM) into a new context. This leads to more attractions on control charts as one of the useful tools that contribute to the success in SM by anomaly detection (AD) approach. Coefficient of variation (CV) is a recent popular statistic that is used in the quality control of SM. In this paper, we propose investigating the performance of Cumulative sum (CUSUM) control charts monitoring CV with a fast initial response (FIR) strategy. The chart parameters are also optimized according to the random shift size in a given interval with the proposed Nelder-Mead optimization algorithm. The numerical results show that the performance of FIR CUSUM-γ2 charts are greater than the initial CUSUM-γ2 ones. An example in monitoring yarn quality at the spinning mill with the design of FIR CUSUM-γ2 charts is also proposed. These findings are useful for practitioners as well as managers and researchers. The proposed design of FIR CUSUM-γ2 charts could be applied in other processes of various domains such as finance, business, industrial processes, etc.