Based on the traffic conflict technique, this paper analyzes the aerial photography video of the highway interchanges, obtains the survey data, and calibrates the parameters of VISSIM and SSAM. Traffic conflict prediction models for single lane ramp and two-lane ramp are established. The negative binomial distribution model is selected as the single-lane ramp collision prediction model, and the Poisson distribution model is selected for the two-lane ramp collision prediction model. The test results show that the model has high accuracy, and MAPE is 8.41 % and 12.0% respectively. The results show the models can provide a basis for the safety assessment of highway merging area and the decrease of traffic accidents.
Collision Prediction Model for Interchange Mergring Area Based on Traffic Conflict Technique
2018-09-01
6673804 byte
Conference paper
Electronic Resource
English
Traffic conflict severity analysis method used for interchange
European Patent Office | 2016
|