Vehicle Routing Problem With Time Windows Of E-commerce Distribution In Time-dependent Networks

With the rapid development of e-commerce,the distribution of e-commerce orders has also increased day by day.With the continuous improvement of urban motorization,the urban road network is more complicated,which also brings greater challenges to the terminal distribution of e-commerce logistics.In the static vehicle problem,the time cost between nodes is only related to the distance between nodes,which is based on the static constant.In the static vehicle problem,the time cost between nodes is only related to the distance between nodes,and is based on a static constant.In the urban distribution network,different road types and different depart periods will have a corresponding impact on the vehicle speed.The time cost between nodes changes in real time according to the change of time,so the distribution scheme based on the static road network is difficult to meet the reality.In the actual distribution process,the vehicle path planning is not only affected by the road network,but also limited by other factors.Based on the actual background of E-commerce distribution,this paper proposes a vehicle routing problem with the maximum vehicle capacity limit and the time windows limit for the storage center and distribution stations in the time-dependent networks.In response to this problem,this thesiscarried out the following work:(1)Analysis of vehicle routing problem with time windows in time-dependent networks.First,the problems of this study are described in detail,the characteristics of time-varying road network are analyzed in depth,and the method of characterizing the time-varying road network is proposed.(2)Based on the time-dependent road networks with FIFO criterion,a vehicle routing problem model with time window is constructed.Through the construction of the upper limit of vehicle capacity,hard time window constraints and other basic constraints,the optimization objective of the shortest total delivery time is solved.(3)Design of two-stage solution algorithm.In the dynamic vehicle problem,the spacetime distance is introduced for clustering,and then the initial solution within the group is constructed.Finally,the variable neighborhood search algorithm is used to re-optimize the initial solution according to the set neighborhood structure and local search.(4)Taking an e-commerce logistics enterprise in Shenyang city as an example background,on the basis of actual data,reasonably randomly generate an experimental example of this problem.Based on the changing travel time between nodes,the vehicle problem with time windows in the time-varying road network environment is solved.Comparing the optimal solution with the solution set of the static VRPTW problem verifies the necessity of this research problem and the effectiveness of the model.This thesis uses e-commerce distribution as the problem background to model and solve the time-dependent vehicle routing problem with time windows.It not only provides decision support for e-commerce companies to formulate distribution plans,but also provides solutions to similar urban distribution problems.Further enriching the relevant theoretical research on time-varying road networks.