移动货架仓库中搬运机器人优化调度模型与算法

Model and Algorithm for Optimizing the Schedules of Handling Robots in Mobile-rack Warehouses

基于搬运机器人的电商仓库近年来在很多大型电商物流中心得到了广泛应用。面对大规模电商订单和多样化商品需求,如何调度好机器人完成栋货任务,发挥出该类仓库面向电商订单的分栋优势,是每个仓库决策者函待解决的关键问题。针对该问题,综合考虑订单栋货任务分配、货架服务顺序、栋货台工作安排及机器人调度复杂环节,本文以栋货时间最小化为目标建立了包含多抹货台的混合整数规划模型,证明了问题的NP-Hard特性,设计了基于模型的数学启发式算法,以及基于模拟退火和特殊反馈机制的两阶段自适应大邻域搜索算法。通过企业真实数据验证了模型和算法的科学性和有效性。

E-commerce warehouses based on handling robots have been widely used in many large e-commerce logistics centers.In the face of large-scale e-commerce orders and diversified product demands,how to schedule robots to com-plete the picking task in mobile-rack warehouses and give full play to the picking efficiency advantages of such warehou-ses for e-commerce orders is the key problem that every warehouse decision-maker needs to solve.To handle the prob-lem,the complex parts,such as order picking task allocation,service sequence of racks,work assignments of picking stations,and robot scheduling,are comprehensively considered.A mixed integer programming model containing multiple picking stations with the aim of minimizing the picking time is constructed.The NP-hardness of the problem is proved.A model-based mathematical heuristic and a two-stage adaptive large neighborhood search algorithm based on simulated annealing and special feedback mechanism are developed.The scientificity and effectiveness of the model and algorithm are verified on the real data of an e-commerce logistics enterprise.