Research on Production Decision-Making of Assembly Enterprises Based on Dynamic Bayesian Sampling and Multi-Stage Optimization

Abstract

This paper focuses on the production decision-making problems of electronic product assembly enterprises, and comprehensively uses probability theory, statistics and optimization theory to construct a series of innovative models and algorithms. For the sampling detection of spare parts, hypothesis testing and Bayesian step-by-step adjustment model based on normal distribution improvement are proposed to effectively balance the detection cost and decision-making reliability. In terms of production decision optimization, dynamic programming, backtracking algorithm and generalized multivariate scheme planning model are used to accurately analyze different production scenarios and determine the optimal strategy at each stage. Through multi-model and algorithm collaboration, it helps enterprises optimize costs and improve benefits in complex production environments, and provides efficient and practical solutions for production decision-making in the assembly industry, which has significant application value.