Designing a yield-based skip-lot sampling plan for lot acceptance determination
本研究将跳批抽样与过程良率指数Spk结合，开发了一种适用于高良率生产环境的变量跳批抽样计划，通过优化模型最小化平均样本量，并满足生产方与使用方的质量与风险要求。
Acceptance sampling has been widely used in practical applications for incoming or receiving inspections, and various sampling plans have been developed for different circumstances and purposes. Skip-lot sampling plans (SkSPs) are an effective sampling procedure and may be useful as a reduced inspection system when a submitted lot’s quality is good and stable. Thus, this study attempts to integrate the concept of the SkSP and the process yield index Spk to develop a variables SkSP of type 2 (SkSP-2) with an Spk-based single sampling plan (SSP) as the reference plan. This plan is suitable for today’s high-yield process environments and products, with a normally distributed quality characteristic under bilateral specification limits. The plan parameters are solved by an optimisation model with nonlinear constraints, which minimises the average sample number (ASN) and satisfies the quality and risk requirements specified by the producer and consumer. The performance and efficiency of the proposed SkSP-2 are examined and compared with the variables SSP in terms of the operating characteristic and ASN curves. Additionally, tables of the plan parameters under selected conditions are provided for reference, and an example from the printed circuit board industry is presented to demonstrate the practicality of the proposed plan.