Abstract

Maiti et al.(2010), defined a Generalized Process Capability Index (GPCI)  Cpy that is based on the ratio of proportion of specification conformance to proportion of desired conformance of the process under study and has several appealing features. One of its advantages is that it can be used not only for continuous processes, as in the case with majority of the indices considered in the literature, but also for discrete processes which are frequently considered in statistical process control. This paper aimed at evaluating the index Cpy for Poissin distribution. The well-known maximum likelihood estimator (MLE) is used to estimate the parameter. The bootstrap confidence intervals are considered in this paper consists of various confidence intervals. Three bootstrap confidence intervals -standard, percentile and bias corrected are compared based on average width and coverage probability.

Keywords: Poisson distribution, process capability index, maximum likelihood estimate, bootstrap confidence interval.