ANALYSIS OF DECISION-MAKING CHALLENGES IN PRODUCTION PROCESSES

This study focuses on challenges in electronic product manufacturing like quality control, cost optimization and multi-stage decisions. It uses a comprehensive math modeling and optimization approach. Advanced techniques such as sampling inspection, decision analysis, dynamic programming and Bayesian inference are integrated to build decision tree and multi-stage dynamic programming models, implemented via Python. For sampling, a hypothesis testing-based scheme is developed. At 95% and 90% confidence levels, minimum sample sizes of 138 and 108 components are set respectively, with error margin within 5%, balancing accuracy and cost. Also, decision tree modeling optimizes key processes like inspection, disassembly and return management. By simulating 16 decision combinations under different conditions and analyzing costs, the optimal cost-effective strategy is found. Overall, it offers enterprises tools and insights for better production decisions.