Abstract
When the process yield is increased due to the advanced manufacturing system nowadays, logically, the engineer would attempt to examine fewer samples to evaluate the quality level of process performance or products. Therefore, in this paper, an efficient partial inspection method based on the process yield index is developed for lot sentencing. Both average sample number (ASN) and operating characteristic (OC) functions of the proposed method are derived based on Markov chain technique and further construct an optimization model, which minimizes the ASN and constrains two OC functions restricted to two given sets of (AQL, 1-α) and (RQL, β), where AQL and RQL are acceptable and rejecting quality levels, and α and β denote producer's and consumer's risks, respectively. On the other hand, the study also reveals that the discriminatory power and cost reduction of the proposed method is superior by comparing OC curves and ASN with existing methods under the same quality condition. Finally, we prove the applicability of the proposed method by demonstrating a case study taken from a manufacturing company.