考虑负载平衡多目标装箱问题的混合遗传算法研究

Research on Hybrid Genetic Algorithm for Load Balancing Multi-Objective Packing Problem

摘要: 针对集装箱装载问题提出了一种多目标优化模型，旨在优化空间利用率和负载平衡。研究中提出了一种混合遗传算法，将有偏随机秘钥和精英策略的非支配排序遗传算法相结合，改进左下角装箱策略，解决具有负载平衡约束的三维装箱问题。通过对标准BR数据集进行的仿真实验验证混合遗传算法的性能，并与多目标粒子群优化算法进行了比较分析。结果表明，实验结果表明，混合遗传算法在解决强异构空间利用率方面优于多目标粒子群算法并且在力惩罚方面表现差异不大，研究为集装箱装载问题领域提供了新的视角。

Abstract: A multi-objective optimization model is proposed for container loading problems, aiming to optimize space utilization and load balancing. A hybrid genetic algorithm was proposed in the study, which combines a non-dominated sorting genetic algorithm with biased random keys and elite strategies to improve the bottom left corner packing strategy and solve the three-dimensional packing problem with load balancing constraints. The performance of the hybrid genetic algorithm was verified through simulation experiments on the standard BR dataset, and compared and analyzed with the multi-objective particle swarm optimization algorithm. The results indicate that the experimental results show that the hybrid genetic algorithm is slightly better than MOPSO in solving strong heterogeneous space utilization, and its performance in force penalty is not significantly different. This research provides a new perspective for the field of container loading problems.