Research On Urban End Distribution Approach Under Electronic Commerce

With the rapid development of e-commence, urban end distribution has become a bottle neck problem for e-commence to improve its overall efficiency. "Last mile problem", playing a key role in urban end distribution, has constrained the further development of e-commence. In recent years, more and more researchers have paid attention on joint distribution, which provides a new methodology to solve the above problem.The aim of joint distribution is to maintain a smooth implement of distribution plan. Complex urban transportation network and numerous urban end distribution points are of great significance for the fulfillment of the distribution plan, while also it increases the difficulties of finding the optimized one.In this thesis, an urban distribution network is built based on GIS to imitate the urban transport network. And a concept of distribution time cost is introduced in 1-M-1 model. To solve this problem, a two-stage optimization method based on geographic information system and improved cooperative particle swarm optimization(GIS-CPSO) is proposed. This method takes full advantage of powerful analysis of ArcGIS and strong global search of improved CPSO with a GS-LS-NS cooperative learning mechanism.Several experiments are conducted to show the better performance and faster convergence speed of GIS-CPSO compared with either single PSO or ArcGIS network analysis function.It can be concluded that distribution plan based on urban transport network can deliver great values to logistic agencies. And the GS-LS-NS searching mechanism has good performance on improving the efficiency of PSO, which can also be used in other intelligent search algorithm.