Research On Optimization Of Multi-center Distribution Route Of City Logistics With Time Windows

With the rapid development of the market economy and the network information technology,the impact of the urban logistics industry on the level of the urban economy is increasingly obvious.According to the National Bureau of Statistics,although the total cost of urban logistics has declined slightly in the past five years,accounting for about 15.4%of the national economy on average,the distribution cost accounts for about 53.6% of the total logistics cost,which is still in a high position.Therefore,optimizing urban logistics scientifically and reasonably and decreasing urban logistics distribution costs can not only help urban enterprises reduce unnecessary cost,but also improve the development level of urban economy.For larger enterprises in large and medium-sized cities,the single-center distribution model can no longer meet diversified market demands.In order to ensure that vehicles can be dispatched in a timely manner in the face of emergencies,and that delivery can meet customer time windows and ensure the quality of goods in the face of scattered orders companies will establish multiple centers for delivery according to the division of urban areas.The article is based on the background of urban logistics,and the specific studies are as follows:First,the article discusses the academic status of domestic and foreign research and related theories on vehicle route problem.we can understand the distribution model of urban logistics and the layout planning of the urban's distribution network.Based on the detailed description of the multi-center distribution routing problem and the time window problem,the article determines that the total distribution cost is the optimization goal,analyzes the cost structure of the optimization target and introduces the necessary conditional constraints,and finally establishes a multi-center distribution vehicle routing optimization with time window.Second,the article considers the applicability and limitations of each optimization algorithm,and then it designs improved saving-genetic hybrid optimization algorithm and proposes a three-stage solution strategy of the algorithm: In the first stage,the paperdetermines the distribution scope of each center and allocates customer groups,which simplify the difficulty of solving the problem;in the second stage,the paper uses an improved saving algorithm to generate a feasible solution for the multi-center distribution vehicle route problem.The algorithm improvement is to add problem constraints when planning the distribution route,and to introduce the 2-opt algorithm idea when adjusting the customer order on the route;In the third stage,the article re-optimizes the feasible solution as the initial population of the genetic algorithm to obtain the final optimal solution of the model,and verifies the validity and feasibility of the model and algorithm through the MATLAB R2016 a simulation software with the help of the relevant delivery data of Nanjing LW Trading Co.,Ltd.Finally,the article compares the company's previous distribution route plan.The optimized plan has reduced the total cost of distribution by 20.54%,and also increased the vehicle utilization rate by 10.4%.The company's economic benefits have been significantly improved.At the same time,through the performance detection and analysis of the algorithm,the improved saving-genetic hybrid optimization algorithm has improved the convergence speed by 13.31% compared with the standard genetic algorithm,which improves the operating efficiency and also enhances the parallel search ability.It is easier to find the optimal solution of the scheme while ensuring the quality of the solution.