配送中心订单分批拣选并行优化研究

发布时间：2018-10-15 10:23

【摘要】：当今，信息技术、移动互联网的快速发展，电子商务作为新型的商业运作模式完全融入到人们生活的各个方面。而配送中心是电子商务得以实现的核心，也是发展电子商务的一个瓶颈。在配送中心中，拣货作业就是按照订单上的订单信息将货品项从货架上检出、分类、集中、包装、装箱的作业过程。整个过程中，，最耗时且劳动量最高的活动就是拣选作业。尤其是近年来，客户订单逐渐向多样化、小批量的方向发展，对拣货作业运作效率的研究逐渐成为物流供应链研究领域的热点，因此对拣货作业运作过程的优化对提高配送效率有重要意义。本论文研究了拣货订单分批模型、分批拣选并行优化，首先，对一定量的订单按合计量分批、时窗分批、订单量分批、智能型分批模型建立比较，采用智能分批模型利用节约算法进行分批中的订单路径计算，设计遗传算法，在满足约束的条件下寻找局部最优解，使拣货行走时间减少，提高拣货效率。其次，对建立起的分批模型进行进一步的并行改进，对细粒度模型、主从式模型和粗粒度模型三种并行模型研究，选取了主从式模型进行并行改进，将一定数量的订单按种群方式进行初始化并行分批计算，分批订单就是各个种群，各个分批依靠迁移算子传递自己的所有变化信息，使各个批次达到协同最优，通过人工选择系数对每个种群进行最优个体保存，取得较好的收敛度[1]，提高了精度，得到最优解。根据论文中设计的拣货分批策略和并行遗传优化目标，对设计策略及并行遗传算法进行测试用例模拟仿真验证，结果表明为方法选择与应用提供了依据并提高了效率。
[Abstract]:Nowadays, with the rapid development of information technology and mobile Internet, E-commerce, as a new business operation mode, is fully integrated into all aspects of people's life. Distribution center is the core of e-commerce and a bottleneck of e-commerce development. In the distribution center, picking is the process of checking, sorting, centralizing, packing and packing items from the shelves according to the order information on the order. The most time-consuming and laborious activity in the whole process is picking. Especially in recent years, customer orders have been gradually diversified and developed in small quantities. The research on the operational efficiency of picking operations has gradually become a hot spot in the field of logistics supply chain research. Therefore, the optimization of picking operation process is of great significance to improve the efficiency of distribution. In this paper, the batch model of picking orders is studied, and the batch picking parallel optimization is studied. First of all, a comparison is established for a certain number of orders according to the combined measurement batching, time window batch, order quantity batching, and intelligent batch model. The intelligent batch model is used to calculate the order path in batches and the genetic algorithm is designed to find the local optimal solution under the condition of satisfying the constraints so as to reduce the walking time of picking and improve the picking efficiency. Secondly, further parallel improvement is carried out on the batch model, and three parallel models, fine-grained model, master-slave model and coarse-grained model, are studied, and the master-slave model is selected for parallel improvement. A certain number of orders are initialized in parallel batches according to the population mode. Each batch order is each population. Each batch depends on the migration operator to transfer all its own change information, so that each batch can achieve the cooperative optimum. The optimal individual preservation of each population is carried out by artificial selection coefficient, and a better convergence degree [1] is obtained, the accuracy is improved and the optimal solution is obtained. According to the batch picking strategy and the parallel genetic optimization goal designed in this paper, the test case simulation of the design strategy and the parallel genetic algorithm is carried out. The results show that the method selection and application are based on and the efficiency is improved.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP18

【参考文献】