考虑制造业即时物流的拣选订单分配与排序优化研究

Optimization of Order Allocation and Sequencing Considering the On-demand Logistics in Manufacturing

制造业即时物流是指从供应商或仓库拣选原材料,不经中转直接运输至生产线的生产物流,具有订单变化快、配送时间紧、齐套要求高、涉及场景多、单次运量大、运输要求严的特点。订单拣选作为制造业即时物流的上游环节,其效率将直接影响后续配送与生产。因此,研究制造业即时物流的订单分配与排序问题十分必要。本文建立多目标混合整数规划模型,以最小化延迟订单数和最小化料箱出库次数为目标,决策订单在拣选台中的分配和拣选顺序。针对该场景的特点,提出即时优化和连续优化两种订单下发策略,分别设计GSC(greedy spectral clustering)算法和VNS-CI(variable neighborhood search-cheapest insertion)算法进行求解。数值实验表明:GSC算法的运算速度最快,1000个订单的优化时间仅0.83 s,延迟订单数较低,料箱出库次数平均节约7.26%;VNS-CI算法能在GSC算法的基础上进一步提升优化效果,料箱出库次数平均节约16.37%。

The on-demand logistics in manufacturing can be defined as manufacturing logistics in which the required raw materials are picked from suppliers or warehouses and directly transported to the production line.The characteristics of on-demand logistics in manufacturing are rapid order change,tight delivery time,high requirement of assembly kitting,diversity of optimization situations,large quantity,and strict transportation requirements.As the upstream link of on-demand logistics in manufacturing,the efficiency of order picking will directly affect subsequent distribution and production.Therefore,it is necessary to study the order allocation and sequencing problem in ondemand logistics in manufacturing.A multi-objective mixed integer programming model was established to decide the assignment of orders among picking stations and the picking sequence of each picking station.The goal of the model was to minimize the number of tardiness and the number of retrieving of turnover boxes.Two order placement strategies,which were real-time optimization and consistent optimization,were proposed based on the requirements of on-demand logistics in manufacturing.The GSC(greedy spectral clustering)algorithm and the VNS-CI(variable neighborhood search-cheapest insertion)algorithm were designed for the two strategies.Numerical experiments show that the GSC algorithm is the fastest,with an optimization time of 0.83 seconds for 1000 orders,a lower number of delayed orders,and an average savings of 7.26%in the number of retrieving.The VNS-CI algorithm can generate a better solution based on the GSC algorithm and saves 16.37%of the retrieving.