A Simulation Study of Predicting Conflict-prone Traffic Conditions in Real-time

Current approaches to estimate the probability of a traffic collision occurring in real-time primarily depend on comparing the traffic conditions just prior to collisions with the traffic conditions during normal operations. Most studies acquire pre-collision traffic conditions by matching the collision time in the national crash database with the time in the aggregated traffic database. Since the reported collision time sometimes differs from the actual time, the matching method may result in traffic conditions not representative of pre-collision traffic dynamics. This may subsequently lead to an incorrect calibration of the model used to predict the probability of a collision. In this study, this is overcome through the use of highly disaggregated vehicle-based traffic data (i.e. vehicle trajectories) from a traffic micro-simulation (i.e. VISSIM) and the corresponding traffic conflicts (i.e. dangerous concurrences between vehicles) data generated by the Surrogate Safety Assessment Model (SSAM). In particular, the idea is to use traffic conflicts as surrogate measures of traffic safety, and data on traffic collisions are therefore not needed. Two classifiers are then employed to examine the proposed idea: (i) Support Vector Machines (SVMs) – a sophisticated classifier and (ii) k-Nearest Neighbors (kNN) – a relatively simple classifier. Substantial efforts are devoted to making the traffic simulation as representative to real-world as possible by employing data from a motorway section in England. Four temporally aggregated traffic datasets (i.e. 30-second, 1-minute, 3-minute and 5-minute) are examined. The main results demonstrate the viability of using traffic micro-simulation along with the SSAM for real-time conflicts prediction and the superiority of 3-minute temporal aggregation in the classification results. Attention should be however given to the calibration and validation of the simulation software so as to acquire more realistic traffic data resulting in more effective conflicts prediction.