Research on the Optimization Model of Enterprise Production Decision-Making Based on Sampling Test and Expected Cost Analysis

This paper focuses on the decision-making problem in the production process of enterprises, and researches and analyzes the strategies at each stage of the production process. In order to solve the problem of estimating the defective rate of enterprises in the inspection of spare parts, a two-sided inspection method based on hypothesis testing is designed to minimize the number of inspections as the goal, and different inspection schemes are constructed with a confidence level of 95% and 90%, respectively. It can not only effectively control the inspection cost, but also accurately predict the defective rate of spare parts. By applying the binomial distribution hypothesis test and combining different confidence levels, the estimated defective rate of each spare part is obtained. A decision model covering four key decision points is constructed by considering various aspects. Combined with the dynamic planning model, the decision-making of each link is gradually optimized to significantly reduce the inspection cost and dismantling cost involved in the production process. The decision tree model analyzes the cost and defective rate performance in different scenarios, and the optimal strategy for 6 different production environments can be derived to ensure the maximization of economic benefits. In the face of more complex production scenarios, the optimization of multi-process and multi-part combinations is considered. Through comprehensive consideration, a complex decision model covering 16 decision points and 65,536 possible decision scenarios is constructed. In this paper, Dynamic Bayesian Network (DBN) and Markov Decision Process (MDP) are introduced to dynamically update the defective rate estimation in each process, and Monte Carlo simulation method is used to predict the defective rate changes in different scenarios. The optimization of the strategies in combination with genetic algorithms enables the dynamic adjustment of the strategies in complex multi-stage production, and ultimately leads to the optimal strategy that can minimize the total cost in a complex multi-process environment.