目的 文中以减少汽车线束加工立体仓库堆垛机运行时间和距离，提高订单处理效率为目的，建立货位优化的数学模型。方法 根据某汽车线束加工立体仓库的真实情况，在分析历史订单数据的基础上，通过基于定位储存策略的改进遗传算法对货位分配问题进行仿真和求解。结果 仿真结果表明，定位存储策略优于就近存储策略、随机存储策略和分类存储策略，其减少堆垛机运行时间的比例分别达到了18.8%, 16.9%和35.7%。结论 采用定位存储模型能有效改善汽车线束加工立体仓库系统处理生产订单的效率。

The work aims to create a mathematical model for optimization of storage location assignment problem (SLAP) so as to reduce operation time and distance of the stocker in automobile harness warehouse and improve order handling efficiency. According to the actual conditions in an automobile harness warehouse, the work analyzed the historical data and simulated the SLAP by the improved genetic algorithm that was developed based on the dedicated storage model. The simulation results indicated that the dedicated storage allocation system outperformed the closest open location, the random storage and the class-based storage method. It reduced the operation time of stacker by 18.8%, 16.9%和35.7% respectively. In conclusion, the dedicated storage model can effectively improve the efficiency of order handling in automobile harness warehouse.