Research On The Layout Of Underground Logistics Notes In Chengdu Based On Subway Network

With the continuous improvement of the economic level and the continuous expansion of the city scale,the urban population is increasing day by day,which leads to the increasing traffic flow on the urban roads,thus causing the traffic problem that is difficult to solve fundamentally.When the traffic flow in the city tends to be saturated,due to the constraints of the limited area of the roads on the ground,the roads cannot be expanded without restrictions,and the appearance of the subway passenger transport mode has relieved the urban traffic pressure to a certain extent.Therefore,in terms of freight,it is also considered to rely on the subway.Transportation,moving some of the above-ground freight to the ground.The combination of above-ground and underground freight transportation can effectively alleviate urban traffic problems and improve logistics efficiency.At present,the research on using the subway network to carry out the layout of underground logistics nodes is still in the early stage.In order to meet the needs of logistics,this paper plans the layout of underground logistics nodes based on big data,which provides a certain theoretical basis for future research.Firstly,the research process and mode of the underground logistics system are summarized,and combined with the structural form and characteristics of the logistics network,the underground logistics network structure of this paper is established,including three parts: transfer point,distribution point,and demand point.There is a two-way connection between the distribution point and the demand point.Among them,the demand point is composed of residential areas and office buildings,and the transfer point and distribution point are composed of the Chengdu Metro station.By summarizing the development status of Chengdu Metro,the research object of this paper is divided into core area and external area.Secondly,by summarizing the influencing factors of the underground logistics node layout in the previous literature,it is determined that the influencing factors in this paper include four categories.According to the determined influencing factors,the spatial data of eight characteristic variables including residences,logistics outlets,bus stops,office buildings,banks,hotels,supermarkets and shopping malls are obtained through the Auto Navi map api.Then,the K-means++ algorithm with improved distance measurement and the bisection K-means algorithm with improved distance measurement are used to perform cluster analysis on the demand points and distribution points in the core area to determine the range of distribution points covered by each transit point and the size of the covered demand points.scope.According to the distance from the demand point to the distribution point,determine the range of demand points covered by each distribution point.Finally,a layout map of three-level regional underground logistics nodes including transfer points,distribution points,and demand points is formed.Finally,based on the average distribution distance of subway distribution points in the core area,the core area is gridded,and the number of 8 characteristic variables in the effective grid is counted to form the data set required by the model.Based on the obtained data set,establish a multiple linear regression model,a BP neural network regression model,and a support vector machine regression model,determine the optimal model of each model,and compare the matching degree between the prediction results of the three optimal models and the actual value,and determine the optimal model.planning model.Based on GIS,the data set of the outer area of Chengdu is obtained,and the planning model is used to plan the layout of the subway distribution points in the outer area.