Estimating and testing process yield with imprecise data

摘要

The index SPK provides an exact measure of process yield for normally distributed processes, and has been widely used in manufacturing industry for measuring process performance. Most studies on estimating and testing process yield are based on crisp estimates involving precise output process measurements. However, it is not uncommon for measurements of product quality to be lack precision. This study designs a realistic approach for assessing process yield that considers a certain degree of imprecision on the sample data. By adopting an extended version of the approach of Buckley, the membership function of fuzzy estimator of SPK index is constructed. With normal approximation to the distribution of the estimated SPK, two useful criteria for fuzzy hypothesis testing, critical value and fuzzy p-value, are developed to assess process yield based on SPK. Finally, an application example is presented to demonstrate the application of the proposed approach.