Passenger Volume Forecast Of Coastal Railway In The West Of Guangdong Province

Traffic forecast plays a key role in preliminary work of transportation construction project. The result of traffic forecast depends the necessity and resolution and scale and standard of the project, and it's also the foundation that the operation of project were economically appraised.The west railway corridor in the west of guangdong province is the main corridor between guangdong province and the south-west area of china. As the west railway corridor just has a single-line (guangzhou- zhaoqing-maoming)in present ,it doesn't be fitted for the demand of area economic. So, the government has a urgent plan that a new high -standard railway located at the west coastal area of guangdong province should be built as soon as possible.The paper firstly analyze the main use of the west by-coast railway and the present social economical state and its trend ,together with the present state of transportation and its planning in the project-influencing area, then discuss the theory of traffic forecast, summarize the universal rules and the technical points, and finally, use based on bp neural network 4-stage method ,make a forecast on the traffic of the project in the future specific years. The main word includes that the bp neural network is used in the stage of trip generation and the gravity model and growth model are both used in the stage of trip distribution. The logit model is used in the stage of mode split.The paper major innovation points is following :1) the paper uses the line and the flow of passages in the main road-station and the flow of highway to build base road passage od. This method can save plenty of money and time.2) this paper is measured national's total output value and population etc factor and the traffic flows of passengers and has been build the direct contact ,thus carries on the flow of passengers forecasting. The living example certification indicates that the neural networks can pledge the satisfactory mapping precision and gain the forecasting result of ideal. 3) based on the real effect of using the gravity model, the paper shows the shortcomings of the model and indicates the guideline of developing in the stage of mode split.