Research on Production Process Quality Control Optimization Based on Multi-stage Decision Model

Abstract

In industrial production, quality control optimization while minimizing costs remains a critical challenge. This paper proposes an integrated approach combining sampling hypothesis testing and multi-stage decision models to optimize production processes. Through establishing a comprehensive theoretical framework, we develop a novel sampling inspection scheme that achieves 95% confidence for rejection and 90% confidence for acceptance at a 10% nominal defect rate. Based on dynamic programming principles, a multi-stage decision model is constructed to optimize component supplier selection and production phase decisions. The model is extended to handle complex manufacturing scenarios with m processes and n components, effectively managing 8,192 possible decision paths. Experimental results demonstrate significant improvements, including 15-20% cost reduction and 25-30% defect rate improvement. Monte Carlo simulation validates the model's robustness under uncertain defect rates, providing practical insights for manufacturing quality control optimization.