Developing a sampling plan by variables inspection for controlling lot fraction of defectives

Acceptance sampling has been widely used tool for determining whether the submitted lot should be accepted or rejected. However, it cannot avoid two kinds of risks, accepting undesired poor product lots and rejecting good product lots. Such risks are even more significant as the rapid advancement of the manufacturing technology and stringent customers demand is enforced. A yield index S_pk has been developed to provide an exact measure on process yield or fraction nonconforming for normally distributed processes with two-sided specification limits. Therefore, the aim of this paper is to develop a variables sampling plan for evaluating the lot or process fraction nonconforming based on the yield index. The probability of lot acceptance is derived based on the sampling distribution and two-point condition on OC curve is used to determine the plan parameters. Tables of the plan parameters and step-by-step procedure are provided for the practitioner to make decision on lot sentencing especially for situations of products with very low fraction of nonconformities.