Missile Life Extension Maintenance Scheduling Optimization: Multi-Objective Differential Evolution Algorithm Based on Reinforcement Learning

Abstract

Missile life extension maintenance is a key component of equipment storage and life extension engineering, involving complex maintenance task scheduling problems. This paper proposes a multi-objective differential evolution algorithm based on reinforcement learning to solve the multi-objective optimization problem of how to minimize the overall maintenance time and prioritize the completion of key maintenance tasks under limited resources and strict logical constraints. First, the algorithm constructs a multi-strategy framework to transform the goals of minimizing the overall maintenance time and completing the priority tasks into solving the Pareto optimal solution set. Then, this paper integrates reinforcement learning to adaptively adjust the parameters of the differential evolution algorithm to optimize the search efficiency and accuracy. Finally, experiments are conducted on the missile maintenance benchmark dataset to verify the effectiveness and superiority of the proposed algorithm. Simulation results show that compared with the existing algorithms, this algorithm has an improvement advantage in maintenance scheduling effect. In addition, the ablation experiment further confirms the practical effectiveness of the optimization measures in improving the performance of the algorithm.