A Prediction Model Of Truck Traffic Demand And Effect Evaluation Method Of Traffic Management Policy

In recent years,with the increasingly severe air pollution,affected by fuel type and power demand,the pollutant emission of freight vehicles become more and more serious,which has become the top priority of traffic pollution control.In order to reduce the pollutant emission of freight vehicles,cities at home and abroad generally adopt a freight vehicle traffic management policy to reduce the impact of pollutant emission of freight vehicles on specific areas,and the policy is used to restrict the freight vehicles in a certain road or small area first,and gradually expanded to urban areas.However,there is a lack of evaluation on the impact of the implementation of the policy from a theoretical perspective.The main reason for this problem is that with the improvement of urban traffic information collection and analysis technology,although some freight vehicles related data can be obtained,to some extent,it has the basis of quantitative analysis,but,due to insufficient information of OD flow of freight vehicles,it is difficult to directly support the prediction and analysis of the traffic demand of freight vehicles in the road network under the freight traffic management policy and the implementation effect of different policy scenarios and it is possible to cause impact on the spatial and temporal distribution pattern of urban freight transport.Therefore,how to predict the traffic demand of freight vehicle on road network under different policy scenarios is a key issue to analyze the freight vehicle traffic management policies,and it is also a difficult problem urgently needs to be solved.There are three core issues that need to be addressed in order to analyze or predict the policy effect of the freight vehicle traffic management policy:(1)how to conduct multi-source fusion analysis on big data of road network freight vehicle operation and traffic operation,and provide data foundation for policy effect evaluation and cargo traffic demand distribution;(2)how to predict the travel route selection of freight vehicles under the influence of policies and provide a basis for freight vehicle traffic demand forecast;(3)how to grasp the distribution of freight demand and support the road network freight vehicles traffic demand and pollutant emission control under different policy scenarios..Based on the road network freight vehicles operation data and traffic operation data,this paper focuses on the analysis of the temporal and spatial characteristics of freight vehicles operation,the prediction of route selection,and the prediction research on OD demand of freight vehicles based on multi-source big data,andanalyzes the freight vehicles traffic management policy scenarios in Beijing.Firstly,this paper systematically analyses the sources and characteristics of related data indicators of freight traffic volume in various traffic data acquisition methods in existing cities.Based on the positioning data and traditional data of freight vehicles,the analysis method of freight traffic volume shall be studied,and a prediction method of freight traffic volume based on KNN algorithm shall be proposed for non-data sections.Based on freight vehicle location data and traffic survey data,the OD features of available trucks can be extracted by data fusion,and the spatial and temporal distribution characteristics of trucks can also be analyzed.Secondly,the paper researches on the route selection forecasting of freight vehicles and analyzes the influence mechanism and influencing factors of the traffic management policy on the selection of freight vehicle travel route,establishes the generalized cost function of route selection in consideration of goods type,vehicle model and road network constraints,and the freight vehicles travel route selection model based on non-aggregate analysis,designes the calculation method and analyze the sensitivity.Thirdly,the prediction model of freight vehicles OD demand is built.On the basis of building the complex road network topology model,considering the factors such as the proportion of the vehicle model and the driving time under the control measures,the calculation method of the multi-model driving time of the complex road network is proposed,and the dynamic OD reverse deductive model of multi-vehicle of the complex road network is established,predicting the OD demand of freight vehicles,and then the freight vehicle traffic flow distribution model is established,and the heuristic algorithm is designed.Finally,based on the analysis of the current situation of freight transportation and the implementation of traffic management policies in Beijing,the evaluation indicator system of the effect of freight traffic management policies shall be put forward,and the scenes of freight traffic management policies are designed.On the one hand,for the existing scenes of freight traffic management policies,the forecast results of the model are compared with the actual results,and the corresponding analysis conclusions shall be provided as the reference for the next step of policy modification and improvement.On the other hand,different policy scenes shall be designed to predict the core indicators of policy evaluation and analyze the conclusions.The research results of this paper can provide a theoretical basis for the formulation of urban traffic pollution reduction policies at the level of freight vehicles operation.