物流中心订单分拣策略的研究

发布时间：2018-12-28 13:22

【摘要】：物流中心是现代物流活动的核心部门与场所,订单分拣是物流中心最为关键的环节,其成本占到整个物流中心成本的50%以上,以往的研究及实践表明通过仓储分布设计,储位指派,订单分批以及路径规划等策略的有效实施能减少拣选人员行走路程,从而减少分拣时间以改善客户服务水平。本文旨在通过制定合理的订单分批策略以改善人工拣选系统中拣选作业的工作效率。订单分批通过将多个订单合成一个批次或更大的订单以提升拣选设备的利用率并减少工作量,使得分拣过程得以更有效的实施。订单分批问题是NP难问题,因此采用一些高效的算法进行求解是国内外研究的主要方向。本文希望通过研究订单批次分拣策略以求最小化拣选路程,从而节约拣选时间,使得货品的流动周期更短,对订单的响应更迅速。本文在以往研究的基础上提出了三种订单分批算法,分别是降批次启发式算法、基于粒子群的分批算法和基于降批次的遗传分批算法。降批次算法考虑到拣选过程中批次数量对于结果的影响。基于粒子群的分批算法则考虑将粒子群算法用于求解订单分批问题,为了使算法匹配所要求解的问题,对二进制粒子群算法进行了改进。基于降批次的遗传分批算法的提出得益于前人的工作,在遗传分批算法的基础上引入了降批次过程,使得算法有了更好的搜索特性。遗传算法具有快速的收敛和高效的寻优能力,因此在三种算法中求得的结果最好。‘为了便于计算本文还设计了一种穿越策略下的路径计算方法。最后利用matlab仿真软件进行仿真实验,三种算法与经典的启发式算法进行了比较,实验结果显示三种算法都有较经典的启发式算法更好的求解结果,特别是基于群的算法优化性能有大幅提升。
[Abstract]:Logistics center is the core department and place of modern logistics activities, order sorting is the most critical link of logistics center, its cost accounts for more than 50% of the cost of the whole logistics center. The effective implementation of storage assignment, order batching and path planning can reduce the walking distance of the pickers, thus reduce the sorting time and improve the customer service level. The purpose of this paper is to improve the efficiency of sorting in manual sorting system by making reasonable order batch strategy. In order to increase the utilization rate of sorting equipment and reduce the workload, the sorting process can be implemented more effectively by synthesizing multiple orders into one batch or more orders in order to increase the utilization rate of sorting equipment and reduce the workload. Order batch problem is a difficult problem of NP, so it is the main research direction to use some efficient algorithms to solve the problem at home and abroad. This paper hopes to study the order batch sorting strategy in order to minimize the picking path, so as to save the picking time, make the goods flow shorter, and respond more quickly to the order. In this paper, three kinds of order batch algorithms are proposed based on previous studies, namely, reduced batch heuristics, particle swarm optimization and genetic batching based on reduced batches. The batch reduction algorithm takes into account the effect of batch number on the result. Particle Swarm Optimization (PSO) based batch algorithm is considered to solve order batch problem. In order to match the required solution, binary PSO algorithm is improved. Based on the previous work, the genetic batch algorithm based on reduced batch is proposed, and the process of batch reduction is introduced on the basis of genetic batch algorithm, which makes the algorithm have better search characteristics. Genetic algorithm has fast convergence and efficient optimization ability, so the results obtained in the three algorithms are the best. In order to be easy to calculate, this paper also designs a path calculation method based on traversing strategy. Finally, the three algorithms are compared with the classical heuristic algorithm by using the matlab simulation software. The experimental results show that the three algorithms have better results than the classical heuristic algorithm. In particular, the performance of swarm-based algorithm has been greatly improved.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP18

【参考文献】