Abstract
The acceptance sampling plan is an important management approach for quality assurance which provides reliable decision rules for lot sentencing. Single sampling plan (SSP) has been considered as the most popular sampling plan since its operating procedure is simple to implement. However, the procedure of SSP is also the reason why it demands more unnecessary sample. In order to improve SSP, some scholars developed repetitive group sampling (RGS) plan. Comparing to SSP, RGS plan usually has less required sample size under the most of quality levels. Nevertheless, while the quality level is between acceptable quality level (AQL) and rejectable quality level (RQL), RGS plan probably has to resample many times. In this situation, the average sample number (ASN) may exceed SSP. To overcome the above problem, this paper constructs the new RGS plan with the adjustable mechanism based on process capability index. Through the limitation of total sampling times, the designed plan effectively reduces ASN when the quality level is in the region [AQL, RQL]. Moreover, the operating characteristic function is derived by using the exact sampling distribution in the proposed plan. The plan parameters are calculated by solving optimization model and tabulated under benchmarking quality levels and allowable risks for practical applications. The performance of the plan is examined and compared with the existing sampling plans in terms of operating characteristic curve and ASN curve. Finally, two examples are provided to illustrate the applicability of the proposed plan.