The Establishment Forecast Model Of Traffic Accidents On Long And Steep Downgrades In Qinghai Province

The rapid development of transportation results in the increase of traffic accident, and imperils the people's life and possession. Continuous long and steep downgrade sections are often black-spots. It is urgent to find a method which can reduce the loss, prevent traffic accident and improve the status of traffic safety. The road accidents forecasting is the important content of the research of traffic safety. The purpose of road accidents forecasting is to analyze the tendency of road accidents under existing road traffic conditions, evaluate the feasibility and practical effectiveness of traffic safety measures reasonably, control the factors affecting road accidents, and reduce the traffic accidents.This topic is based on one technical project of Qinghai Province transport department named study on safety and improve policy on long and steep downgrades of highway in Qinghai province. The establishment forecast model of traffic accidents on long and steep downgrades in Qinghai province is an important part of the project. Through the statistic analysis of various factors which may lead to the accident, it reveals the intrinsically rule and causes of accidents, based on which we can predict the future trend of accidents in order to provide reliable information for effective accident prevention.Firstly, this paper analyses the causes of accidents from several aspects, such as driver, vehicle, road and surroundings based on the investigation of the traffic accidents on long and steep downgrades from Huang Yuan to Dao Tang river highway in Qinghai province. It expatiates on the correlation between the accident and its causes, and lays the foundations for the prediction model. Secondly, some prevalent theories for accident prediction are presented, such as regression analysis, time serial prediction, gray prediction and neural network prediction, etc. And then a traffic accident forecast model on long and steep downgrades is set up with the nonlinear regression.