Research On Urban Distribution Service Point Location And Route Optimization In E-commerce Environment

The form of competition between e-commerce companies has been changed from thecompetition of business models to the competition of logistics services, how to balance thecontradiction between the speed and cost of logistics and distribution has become the key towin the competition between e-commerce companies. Distribution network is the basis oflogistics distribution, and vehicle routing is the core of the logistics distribution, to balancethe contradiction between the speed and cost of logistics and distribution need to look intothis two issues.This article first surveyed several e-commerce companies, to understand the currentnetwork structure and distribution method of e-commerce urban distribution. Based on it, theexisting problems were analyzed, and the corresponding optimization countermeasures andprinciples were put forward.Then in view of the real life distribution route selection and service nodes locationproblem has two different decision makers, the bi-level programming was studied and used todescribe e-commerce urban distribution optimization problem, and the general distributionnetwork service node location bi-level programming model was given. Based on it,considering the city distribution characteristics and the road restrictions policy, the urbandistribution network location bi-level programming model with road restrictions was putforward.Based on the characteristic of bi-level programming problem, a process of hybridhierarchical genetic algorithm suitable for bi-level programming was put forward, and in viewof the urban distribution network location bi-level programming model with road restrictions,the specific operators were put forward, and so were the set covering problem random greedyalgorithm and multiple model vehicle routing problem with road restrictions partheno geneticalgorithm, which were used for solving the nested subproblems. The convergence of thealgorithm was also proved.Finally a particular city distribution case was chosen as an example. The optimizationproblem was solved on the real road network of Guangzhou city. By comparing the results of the optimization solution and the existing distribution solution, the rationality andpracticability of the model and algorithm were proved.Through the above research, can plan out a reasonable service nodes location layout anddistribution route selection, improve the efficiency of single vehicle, reduce the number ofneeded vehicles, reduce the transportation distance and fuel consumption, realize thecompression of transportation cost in distribution, in turn, reduce business operating costs,improve enterprise’s competitiveness.