Abstract: A hybrid strategy heuristic algorithm（HSHA） is proposed to solve the three-dimensional layout optimization problem of strongly heterogeneous container loading. Firstly, considering the characteristics of the packed goods and the local space, the clustering method is used to transform the strong heterogeneous problem into the weak heterogeneous problem. Then, the strategy of remaining space merging and the strategy of local cargo space posture adjustment are designed, to improve utilization rate of container space. Secondly, on the premise of ensuring the diversity of solutions, combining the convergence advantage of greedy algorithm and the optimization advantage of ant colony algorithm, a hybrid algorithm based on selection probability is designed. Thirdly, in order to reduce the feasible solution space and improve the solution speed, a new state value representation and pruning matrix structure are designed. Finally, the effectiveness of the proposed method is verified by the experiment based on the real container data and public datasets.

Key words: strongly heterogeneous, three-space greedy stacking method, three-space ant colony stacking method, hybrid strategy, simulation

摘要： 针对强异类集装箱三维装载问题提出了一种融合策略启发式算法。考虑装箱货物特征和空间特征，采用聚类方法将强异类问题转化为弱异类问题；设计了剩余空间合并策略以及货物局部空间姿态调整策略，以提高集装箱空间利用率；在保证解的多样性的前提下，结合贪心层叠法的收敛优势和蚁群层叠法的寻优优势，设计了基于选择概率的融合算法；设计了新的状态值表示和剪枝矩阵结构，以有效缩减可行解空间。对实际集装箱数据以及公共数据集进行测试仿真，结果验证了算法的有效性。

关键词: 强异类, 三空间贪心层叠法, 三空间蚁群层叠法, 融合策略, 仿真模拟