Abstract: Aiming at the highly personalized and heterogeneous three-dimensional packing problem in the field of e-commerce, a combinatorial heuristic algorithm with strong applicability in the field of e-commerce logistics is proposed. According to the characteristics of packing problems, firstly, a spatial matrix representation of three-dimensional packing was proposed, and based on this representation, an optimized packing inspection algorithm was designed. Secondly, the coding and decoding rules of the heuristic algorithm were optimized, and the large-scale neighborhood search algorithm was introduced to further optimize the path search process of the packing sequence. The experimental results show that the algorithm has good performance in the field of e-commerce logistics packing problem. The solution result is extremely close to the theoretical optimal solution, and the solution quality is better than other packing algorithms and commercial software, especially in more complex packing problems. This algorithm can effectively solve the packing problem in the field of e-commerce logistics with less research and higher demand, thus greatly reducing the packing cost and packaging waste of e-commerce logistics enterprises.