감마 수명분포에 대한 혼합관측과 무고장기간 합격판정 샘플링 계획의 개발 및 비교

Abstract

In this paper, we develop two replacement-type reliability acceptance sampling plans(RASPs) for the gamma lifetime distribution assuming that the shape parameter is known. The two plans are respectively based upon failure-free period and hybrid life tests. We then compare the plans in terms of expected test time to reach a decision, power, etc. Computational results indicate among others that the failure-free period RASP has a shorter expected completion time than the corresponding hybrid RASP when the true scale parameter is large. Finally, sensitivity analyses reveal that the effects of the uncertainties involved in the assumed shape parameter on the producer and the consumer risks are in favorable directions for both parties for both types of plans.