Answers are grounded in this paper and cite their sources.
Manufacturing sectors are continuously improving through the integration of smart manufacturing technologies aimed at enhancing production efficiency and delivering products of exceptional quality. Existing acceptance sampling plans including single, double and group sampling often require larger sample sizes and provide an inadequate balance between producer and consumer risks in high-throughput smart manufacturing environments. This study addresses these limitations by proposing a new attribute acceptance sampling plan designated as the Special Type Double Group Sampling Plan (STDGSP), designed to overcome these gaps within a smart manufacturing setting through truncated life testing based on the Exponential-Poisson (EP) distribution. The proposed plan derives the minimum group size required for inspection while ensuring controlled producer and consumer risks across a range of plan parameters. Key findings show the STDGSP achieves up to a 41% reduction in the Average Sample Number (ASN) compared with benchmark plans, and attains producer/consumer risk levels of approximately α = 0.03 and β = 0.08 versus typical competing-plan ranges of 0.07–0.15. Empirical validation is reported using two published failure datasets that were tested for EP fit (Kolmogorov–Smirnov, Shapiro–Wilk) and compared using AIC/BIC. To demonstrate practical application, a drone manufacturing case study is presented as a numerical illustration based on these published failure datasets. These results provide quality engineers with a statistically rigorous and highly resource-efficient tool for decision-making often minimizing inspection efforts suited to smart manufacturing environments.
No takes yet. Share an insight, caveat, or question.
Velappan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: