A flexible sampling scheme for variables inspection with loss consideration

Abstract

This article develops a variables sampling scheme for resubmitted lots by incorporating the concept of Taguchi loss function. The probability of lot acceptance is derived based on the exact sampling distribution and two-point condition on operating characteristic curve is used to determine the plan parameters that meet both the producer's and consumer's quality and risk requirements. Moreover, the performance of the proposed variables resubmitted sampling plan is investigated and compared with the classical variables single sampling plan. The results indicate that the developed resubmitted sampling plan can provide the same protection with less inspection when the submitted lot is good enough. Tables of the plan parameters under various conditions are provided and the use of the proposed plan is also illustrated with an example.

An improved measure of quality loss for notching processes

“…Wu 13 investigated the coverage rates of the generalized confidence intervals on the basis of the C pm index with the consideration of measurement errors and performed a sensitivity analysis on the confidence interval performance. Recently, Wu et al 14 incorporated the concept of a Taguchi loss function to develop a variable sampling scheme for resubmitting lots and presented that the new method needs less times of inspection has less inspection under certain conditions. Liao 15 compared the estimators of C pm by the Markov Chain Monte Carlo and bootstrap techniques, and the simulation results indicated that the former performed better in most cases.…”