Optimization of Reliability Testing and Multi-stage Production Quality Control Based on Binomial Sequential Testing

Abstract

In modern manufacturing, enterprises are in urgent need of an optimized inspection solution that can guarantee product quality while reducing the sample size and cost for inspection, to address the challenges posed by high costs and strict reliability requirements. This study commences from multiple scenarios of finished product processing and component assembly procedures, based on the decision-making issues during the enterprise production process. A binomial sequential inspection model is established to obtain a suitable sampling inspection plan under binomial sequential inspection: When the confidence level is 95%, the maximum sample size is 6,355, and only 489 components must be inspected. If the defect rate significantly exceeds 10%, this batch of components will be rejected. When the confidence level is 90%, the maximum sample size is 5,293, and only 260 components must be inspected. If the defect rate does not exceed 10%, this batch of components will be accepted. Additionally, models such as the multi-stage production quality control optimization model based on dynamic programming and linear programming, and the multi-process decision optimization model have been established for systematic decision-making.