Immediate Distribution Route Selection Method Based On Traffic Flow Forecasting

Instant Logistics provides an instant delivery service for goods within the same city without intermediate warehousing and door-to-door.The core of instant logistics is instantaneity,that is,to meet the user's requirements of extremely fast and punctual distribution.Under complex and changeable traffic conditions,how to deliver express quickly,timely and punctually,so as to shorten the delivery time for a certain number of customers,is the main problem to be solved in real-time logistics.In the process of route selection of instant distribution,uncertain traffic conditions will lead to poor timeliness and inefficient distribution,which will lead to unreasonable or even infeasible overall distribution scheme.It is an effective way to solve this problem by forecasting the shortterm traffic flow state of the road network and integrating it into the dynamic route selection.Aiming at minimizing the travel time of instant distribution,this paper constructs a dynamic route selection model which integrates traffic flow prediction,and designs a modified sliding window updating algorithm to solve the model.The main research work is as follows:Firstly,the research background of real-time logistics and the research status of traffic flow prediction and dynamic path selection are summarized.The basic characteristics and parameters of traffic flow data are analyzed.Several methods of traffic flow data preprocessing and the rules of relevant formats are introduced to improve the quality of traffic flow data,so as to facilitate the development of subsequent traffic flow prediction.Next,In the process of traffic flow forecasting,the time and space characteristics of traffic flow data can not be obtained simultaneously,which leads to the defect of forecasting accuracy.Firstly,this paper constructs a time-space graph model of traffic flow data,uses graph convolution cyclic neural network algorithm to predict the road network,and constructs a state transition equation to describe the evolution process of traffic flow spatial state.Traffic flow forecasting provides data support for dynamic route selection of real-time logistics distribution.Finally,an example is given to verify the effectiveness of the algorithm.Finally,an end-point delivery point of a courier company Chongqing branch was selected as the research object.By testing the system,the original distribution model of the distribution point is compared and analyzed to verify the performance of the system.Then,based on traffic flow forecasting,we can get the traffic status of the road network in the future.With the shortest travel time as the objective function,we construct a dynamic path selection model with the weight of prediction time.In the process of solving the problem,aiming at the unreasonable problem of updating route mechanism and overall planning,a modified sliding window updating mechanism strategy is proposed to solve the problem.By updating the sliding window,all the path points are traversed and the journey ends.Finally,through comparative experiment and simulation,a case study of the dynamic path selection method which integrates predictive information is carried out to verify the scientificity and effectiveness of the path selection method.