Research on Production Decision Optimization Based on Multiple Intelligent Algorithms

This paper investigates the production decision optimization model based on multi-algorithm, using genetic algorithm, particle swarm optimization algorithm and reinforcement learning algorithm, aiming at comparing the optimal algorithm. This study is aimed at an electronic product manufacturing enterprise, which involves the procurement of spare parts, assembly and finished product testing and other aspects. First, a sampling inspection model is constructed to ensure that the defective rate of spare parts does not exceed the nominal value, and the rejection domains at 95% and 90% confidence levels are calculated. Subsequently, an optimization model that comprehensively considers the inspection cost, defective rate, assembly cost, exchange loss and disassembly cost is established for the decision-making of inspection and non-conforming product handling in the production process. Genetic algorithm, particle swarm optimization algorithm and reinforcement learning algorithm are used for comparative analysis, and the results show that the particle swarm algorithm performs optimally in specific contexts, while reinforcement learning is suitable for long-term optimization strategies. In addition, the uncertainty caused by sampling detection is further dealt with by introducing the robust optimization method. Finally, a mixed integer linear programming (MILP) model is constructed to verify its effectiveness in real production decisions. The research in this paper provides scientific decision support for production and manufacturing enterprises, which is of great significance for improving production efficiency and reducing costs.