Variable Sampling Plan Indexed by Taguchi’s Quality Loss Under Emphasized Difference of Mean

Abstract

AbstractAs an alternative to nonconforming product rate used traditionally as an index of quality, Taguchi has proposed quality loss. From the viewpoint of this new index of quality, some variable sampling inspection plans for assuring product quality indexed by quality loss have been proposed so far. The quality loss is defined as the expected loss by derivation from a target value of product quality characteristic. In concrete, the quality loss is composed of both of the square of the difference between the mean in the actual product quality and the target product quality, and the magnitude of variance in the actual product quality. And then, these two components have the same magnitude of effects in the evaluation of quality loss. On the other hand, it is known that for improving product quality at a production site, adjusting the mean in product quality is often easier than lessening the variance in product quality. Hence, we can see that adjusting the mean is more efficient and effective than lessening the variance as a way of reducing the quality loss. From the mentioned above, this research will consider a single sampling plan indexed by quality loss with consideration of emphasizing the loss derived from the mean in the product quality. And, the operating characteristics of the proposed sampling plan are confirmed through some numerical simulation.KeywordsVariable sampling plansTaguchi’s quality lossPatnaik’s approximation