Research on Production Decision Making Based on Bayesian Optimization and Genetic Algorithm

The purpose of this paper is to optimize the enterprise's purchasing and assembling decisions for key parts in the production process by constructing a production decision model based on Bayesian optimization and genetic algorithm. To address the impact of defective rate, the study first assumes that the sample size is large enough, and applies Bayesian optimization to determine the minimum number of inspections to ensure the effective detection of defective products under different confidence levels. Second, for the multi-stage production decision, a model with profit maximization as the objective is constructed and solved using an algorithm that combines dynamic programming with Monte Carlo tree search. Finally, using genetic algorithm, the model with profit maximization as the objective function is constructed again for solving. Through these studies, this paper provides scientific basis and practical guidance for enterprise decision-making in complex production environment.