The work aims to propose a prefabricated logistics bin specification optimization model and algorithm to solve the order packaging problem of multi-variety and multi-batch products in warehousing logistics environment. Firstly, the split packing rules were established for product orders to determine all split packaging plan. A logistics bin size optimization model was established with the minimum total cost of packaging materials for all orders as the optimization objective. Then, an improved simulated annealing algorithm was proposed for this model. After using greedy strategy to solve the optimal split packaging plan to reduce the computational complexity of the model, a new solution updating operator was designed to improve the optimization ability of the algorithm, and an adaptive step size strategy was designed to balance the global search and local search ability of the algorithm. The experimental results showed that the algorithm had stronger solving ability than other algorithms; compared with the packaging status of example enterprises, the same order could reduce 17% packaging cost. This method can be used to solve the problem of optimizing logistics bin specifications for series transportation under the application scenario of multiple product types, large size differences and dynamic updates, providing an effective set of packaging optimization ideas and solutions for enterprise logistics and transportation management.