EWMA chart with curtailment for monitoring fraction nonconforming

Abstract

The binomial Exponentially Weighted Moving Average (EWMA) control chart has been widely used as an effective tool for detecting small and moderate shifts in fraction nonconforming (p). This article presents a binomial EWMA chart employing a curtailment technique (abridged as Curt_EWMA chart) for monitoring p. The p shift is assumed to follow a Rayleigh distribution. It is found that the curtailment technique can improve the overall detection speed of the EWMA chart while maintaining the false alarm rate at a specified level. An algorithm is presented to find the optimal design of the Curt_EWMA chart. The results of the comparative study reveal that the Curt_EWMA chart outperforms the EWMA chart without curtailment by 41%, in terms of the average number of defectives, under different settings.

Keywords:

• Control chart
• EWMA chart
• curtailment
• Rayleigh distribution
• average number of defectives

Acknowledgements

The authors are very thankful to the editor and reviewers for the constructive and comprehensive comments. Acknowledging gratefully, we have incorporated these insightful comments in the revised version. The editorial assistance of Wayne Solomon and Yash Patel from University of New Brunswick, Canada is greatly appreciated.