Abstract: A shuttle-based storage and retrieval system (SBS/RS) integrates the functions of storage and order picking,had higher operational efficiency.But the system performance may be affected by the scheduling strategy and various parameter settings between devices.Through analyzing the SBS/RS's workflow and computing the service time of equipments,a mixed integer programming model aiming at minimizing the total outbound time was established.As for model solving,the improved ant colony algorithm with tabu search and genetic algorithm were proposed to solve the model separately.At the same time,the heuristic algorithms were combined with Gurobi tools to obtain exact solutions quickly.By analyzing the bottleneck of each improved algorithm,an adaptability scheme between the task scale and the optimal solution was established,which provided strategy for the efficient solution of the task scheduling problem,reduced the order operation time and improved the operation efficiency.

Key words: shuttle-based storage and retrieval system;task scheduling;tabu search , algorithm;genetic algorithm;improved ant colony , algorithm;algorithm adaptation

摘要： 多层穿梭车自动存取系统集存储和拣选功能于一体,利用高层货架实现货物密集存储,穿梭车、提升机等多设备并行作业。多层穿梭车自动存取系统作业效率较高,但是设备调度方案与配置参数等因素均会影响系统性能。通过剖析多层穿梭车自动存取系统的工作流程和设备服务时间,研究穿梭车和提升机并行工作的约束规则,建立了以出库时间最小为优化目标的混合整数规划模型。在模型求解方面上,分别使用禁忌搜索算法、遗传算法改进了蚁群算法,并提出一种Gurobi与启发式算法结合的求解新思路,经过实验验证,求解精度和求解效率有较大提升。最后,通过计算不同的任务规模实例分析了3种改进算法的求解性能,建立了一套出库任务规模与精确求解的适配方案,减少了系统订单作业时间,提高了系统作业效率。

关键词: 多层穿梭车自动存取系统, 任务调度, 禁忌搜索算法, 遗传算法, 改进蚁群算法, 算法适配