Aiming at the characteristics of multi-equipment parallel operation and the requirement of high-efficiency operation in four-way-shuttle based storage and retrieval systems, the work aims to study the optimization of storage allocation. A combined optimization model of storage allocation was established, which took into account the factors of storage efficiency, shelf stability, work balance and goods correlation, and then an improved artificial bee colony algorithm was designed to solve the model. The concept of multi-objective Pareto dominated solution was used to generate the initial solution, and the thought of genetic algorithm and variable neighborhood search algorithm were combined to optimize the solution, which improved the performance of the algorithm. A sensitivity test of algorithm parameters was carried out to determine the parameters. The effectiveness of the proposed algorithm was verified by simulation experiments under three different storage scales (large, medium and small) and nine goods stored scales. The solution of three algorithms (the improved artificial bee colony algorithm, the genetic algorithm, and the ant colony algorithm) signified that the larger the storage scale was, the more obvious the superiority of the improved artificial bee colony algorithm was. Compared with the genetic algorithm and the ant colony algorithm, the performance of the improved artificial bee colony algorithm under largestorage scale was improved by 13.14% and 19.32%, respectively, and its performance under differentgoods stored scales was improved by 6.38%-26.76%. In conclusion, the established optimization model is closer to the actual working state of warehouse, and the proposed algorithm can effectively allocate the storage space and improve the efficiency of the warehouse operation.