基于有约束聚类算法的B2C电商物流网络优化研究

发布时间：2018-04-10 19:48

本文选题：物流网络优化 + B2C电商　；参考：《上海交通大学》2015年硕士论文

【摘要】：随着居民消费水平的提升,对质量的关注也越来越成为消费者选择电商的重要参考标准,使得电商之间从“价格战”逐步转向“服务战”。然而“服务战”的竞争归根结底依然是企业占领市场与运营成本间的博弈问题,如何搭建B2C电商全国物流网络,使得在保持高服务标准的前提下,维持较低的运营成本是关系到当前B2C电商企业未来发展的关键性问题。针对B2C电商网络优化问题,本文设计了包含区域配送中心、前端物流中心与城市节点的三级物流网络拓扑结构,并明确了节点功能与辐射范围。文章共完成如下四方面的工作:第一,分析了影响城市电子商务发展水平的主要因素,并从中选取相对重要的两个指标网商指数和网购指数对全国30个省市的订单量进行函数拟合,得到精确度相对较高的城市网购订单量估测模型,并用实际数据加以验证;第二,以城市群理论与引力模型为依托,将城市运输距离与订单量相结合,通过对欧式距离重新定义,以当量距离为约束条件,借助K-中心点聚类算法完成全国物流网络的区域划分,并对划分的结果根据类内差指数进行评估,找出聚类较为合理的方案;第三,首先对划分后的区域进行定性分析,选出符合条件的前端物流中心与前端物流中心备选节点,再根据交通运输、仓库租赁、运营、服务惩罚等成本因素对区域内包含的城市进行评估,选出符合要求的前端物流中心;第四,对上述物流网络优化模型结合实际数据进行了分析,实例中在全国范围内选取了86个地级市,将全国B2C电商物流网络划分为七个区域,并以华东区为例,搭建了完整的三级物流网络结构。本文综合考虑了影响B2C电子商务行业发展的因素与该领域在物流网络搭建过程中的属性,通过对B2C自营物流的电商企业进行多次调研,了解了相关企业的实际运作流程,综合国内外学者对该类问题的研究方法、研究思路与研究成果,设计出B2C电商全国物流网络优化的研究步骤。对该问题的研究分两步进行,在现有网络拓扑结构的基础上,第一步实现全国物流网络的城市聚合,划分出城市群落。第二步结合经济成本等因素选出每个区域内的前端物流中心,完成网络拓扑结构的链接。
[Abstract]:With the improvement of residents' consumption level, the attention to quality has become an important reference standard for consumers to choose e-commerce, which makes the price war shift from "price war" to "service war".However, in the final analysis, the competition of "service war" is still a game between enterprises' occupation of market and operation cost. How to build B2C e-commerce national logistics network, so as to maintain high service standards,Maintaining low operating cost is a key issue related to the future development of B2C e-commerce enterprises.Aiming at the B2C e-commerce network optimization problem, this paper designs a three-level logistics network topology including regional distribution center, front-end logistics center and urban node, and clarifies the node function and radiation range.The following four aspects of work have been completed: first, the main factors affecting the development level of urban electronic commerce are analyzed.And select two relatively important indexes online merchants index and online purchase index to fit the order volume of 30 provinces and cities in China, and get a relatively high accuracy of urban online purchase order estimation model, and verify it with actual data; second,Based on the theory of urban agglomeration and gravity model, this paper combines urban transportation distance with order quantity, by redefining Euclidean distance, taking equivalent distance as constraint condition, and using K-center point clustering algorithm to complete the regional division of national logistics network.And the result of the division is evaluated according to the intra-class difference index to find out the more reasonable clustering scheme. Thirdly, the qualitative analysis of the divided area is carried out, and the qualified front-end logistics center and the front-end logistics center are selected.Then according to transportation, warehouse leasing, operation, service penalty and other cost factors to evaluate the cities included in the region, select the front-end logistics center that meets the requirements; fourth,The optimization model of the above logistics network is analyzed in combination with the actual data. 86 prefecture-level cities are selected in the example, and the B2C e-commerce logistics network is divided into seven regions, and the East China area is taken as an example.A complete three-level logistics network structure is built.This paper synthetically considers the factors that affect the development of B2C e-commerce industry and the attributes of this field in the process of logistics network construction. Through many investigations on the e-commerce enterprises of B2C self-run logistics, the paper finds out the actual operation process of the related enterprises.Based on the research methods, ideas and achievements of the domestic and foreign scholars, the research steps of B2C e-commerce national logistics network optimization are designed.The research on this problem is divided into two steps. On the basis of the existing network topology, the first step is to realize the urban aggregation of the national logistics network and to divide the urban community.The second step is to select the front end logistics center in each region by combining economic cost and other factors to complete the link of network topology.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F724.6;F259.2;F224

【参考文献】