摘要

Various acceptance sampling plans have been developed for different objectives. A repetitive group sampling (RGS) plan has been shown to be an efficient and easy-to-implement scheme for lot sentencing. However, it does not consider the available information from preceding samples. As a result, it may reduce the sampling efficiency in terms of cost and time. In this study, a modified variables RGS plan is proposed based on the commonly used capability index C-pk for normally distributed processes with two-sided specification limits and to consider the sample results from preceding lots. The plan parameters for various required quality levels and allowable risks are tabulated for practical applications, and the advantages of the proposed plan is compared with existing variables sampling plans in terms of operating characteristic curve and average sample size.