Application of Improved Quantitative Theory for Microscopic Prediction of Traffic Accidents

As the random feature of traffic accidents and there are many factors including qualitative and quantitative ones, accident rate was selected as microscopic predict objective after four traffic accident indexes and accident rate were analyzed. Seven factors, including driving years, number of lanes, radius of horizontal curve, longitudinal grade, road surface status, type of intersection and width of road surface, were selected as major influencing factors. The influencing factors were analyzed from a systematic perspective with the combination of human, vehicle and the road. A microscopic prediction model for traffic accidents based on quantitative theory has been established which developed from quantitative theory. To verify the model, the paper takes the data of road accident and actual structure from 222.888km to 377.387 km on national highway No.307. Furthermore, the road black-spot from 333.5km to 334.0km is selected to applied the model concretely, and some proposals are put forward to improve the present road conditions.