A hypothesis testing procedure on assessing process performance

Abstract

Process incapability index has been introduced to provide quantitative measures on process performance. Investigations of the estimated incapability index based on multiple samples have been proposed and arrested substantial research attention. In this paper, investigation based on multiple samples is considered for normally distributed processes. A reliable inferential procedure based on the Uniformly Minimum Variance Unbiased Estimator (UMVUE) of the incapability index is proposed. Useful critical values required to ensure the process reaching a certain desirable level of the time are also tabulated. A practical example is provided to demonstrate how the proposed procedure may be applied for judging whether the process runs under the desirable quality requirement.