Abstract: Unmanned warehouse is an automatic storage system appeared in recent years. The storage allocation strategy is one of the main factors affecting the efficiency of unmanned warehouse system. Based on the scene that there might be only part of the storage locations empty in the replenishment stage of the unmanned warehouse system, the optimization model and algorithm for storage location assignment problem of unmanned warehouse system is studied. A mixed integer programming model of the storage location assignment problem is formulated to maximize the sum of the correlation degrees among items stored on the same shelves. Based on the greedy algorithm and neighborhood search algorithm, a two-stage algorithm is designed for solving the model. In the first stage, a greedy algorithm is used to obtain an initial feasible solution, in the second stage, the neighborhood search heuristic algorithm is designed to optimize the initial feasible solution. An computer generated example is used to verify the optimization effect of neighborhood search algorithm, the results show that the objective value of solution being optimized by neighborhood search algorithm is increase by at least 27%. Several small-scale examples are used to analyze the approximation ratio and running speed of the two stage heuristic algorithm, which verifies the fast and effective of the algorithm. The model and algorithm of this paper not only solve the storage allocation problem with parts of the storage locations being empty on the shelves, but also can be used to solve the problem with all the storage locations being empty on the shelves. So the application of this method is broader, which can be used as the core model and algorithm for designing the unmanned warehouse management information system.

Key words: unmanned warehouse; storage location assignment; mixed integer programming; greedy algorithm; neighborhood search