An adaptive EWMA mean chart in the presence of outliers

Abstract

ABSTRACTOccasional outliers that might be a natural part of a process may distort the properties of a control chart. In this paper, we show that a recently proposed adaptive EWMA (AE) mean chart is highly sensitive to the outliers. The false alarm rate of the AE chart increases when the proportion and/or magnitude of the outliers increase and vice versa. In order to circumvent this demerit of the AE chart, we propose a truncated normal distribution-based AE (TAE) chart for monitoring the mean of a normal process in the presence of outliers. The zero-state and steady-state average run-length profiles of the proposed chart are estimated using Monte Carlo simulations. Based on detailed run-length comparisons, it is found that the TAE chart may outperform the existing EWMA chart (based on a truncated normal distribution) when detecting various mean shift sizes of an outlier-prone normal process. Illustrative examples are also included in this study to demonstrate the implementation of the existing and proposed charts.KEYWORDS: Control chartadaptive EWMAMonte Carlo simulationprocess meanrun length propertiesstatistical process control AcknowledgementsThe author is thankful to the associate editor and two anonymous reviewers for providing useful comments that led to an improved version of the article.Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationNotes on contributorsAbdul HaqAbdul Haq is an Associate Professor at the Department of Statistics, Quaid-i-Azam University, Islamabad, Pakistan. His research interest is in Statistical Process Monitoring.