Design procedures for variable sampling inspection plans with screening indexed by Taguchi's quality loss under consideration of minimizing average total inspection

The quality of lots has sometimes been evaluated based on attribute properties such as the proportion or number of non-conforming items in the lot. There are two kinds of attribute sampling inspection plans with screening, called lot tolerance percent defective (LTPD) inspection and average outgoing quality limit (AOQL) inspection. These sampling inspection schemes have been designed in order to minimize the average total inspection (ATI), which means the expectation of number of samples to be inspected in the case of acceptance and number of samples inspected in the case that the lot is rejected. Taguchi proposed a concept of quality loss as a measure to evaluate the quality of items based on variable properties instead of attribute qualities such as the proportion of non-conforming items. This variable quality evaluation measure is known as "Taguchi's quality loss." Arizono et al. proposed acceptance sampling plans based on operating characteristics from the viewpoint of ensuring Taguchi's quality loss. Furthermore, Takemoto and Arizono proposed variable sampling inspection plans with screening for the purpose of ensuring the upper limit of maximum expected surplus loss after inspection indexed by Taguchi's quality loss. However, in the variable sampling inspection plans with screening considered in the above articles, minimizing the ATI in the screening sampling scheme has not been considered. In this article, we present design procedures for new variable sampling inspection plans with screening indexed by Taguchi's quality loss for the purpose of minimizing the ATI in conformity with conventional LTPD and AOQL attribute sampling inspection plans with screening.