Nonparametric control charts provide a robust alternative in practice when the form of the underlyingdistribution is unknown. Nonparametric CUSUM (NPCUSUM) charts blend the advantages of aCUSUM with that of a nonparametric chart in detecting small to moderate shifts. In this paper, weexamine efficient design and implementation of Phase II NPCUSUM charts based on exceedance(EX) statistics, called the NPCUSUM-EX chart. We investigate the choice of the order statistic fromthe reference (Phase I) sample that defines the exceedance statistic. We see that choices other thanthe median, such as the 75th percentile, can yield improved performance of the chart in certainsituations. Furthermore, observing certain shortcomings of the average run-length (ARL), we use themedian run-length (MRL) as the performance metric. The NPCUSUM-EX chart is compared with theNPCUSUM-Rank chart proposed by Li et al. (2010) based on the popular Wilcoxon rank-sumstatistic. We also study the choice of the reference value, k, of the CUSUM charts. An illustrationwith real data is provided.
展开▼
