Analysis of accident rate prediction in overtake maneuvers: an hazard analysis application

Accident data bases show that the overtake maneuver is one of the most frequent risky event. The aim of this paper is to define a methodology able to assess the risk connected with this maneuver along a road segment under different operating condition. Referring to the risk theory and using probabilities calculated through the "tree of events", this study develops an innovative method for road accidents predictive estimation. The hazard analysis has been applied since able to keep into account of the sequence of the physical situation and the decisions taken by the drivers. These evaluations have been performed in CRISS (Inter University Research Center for Road Safety - Roma TRE University) laboratory with a driving virtual simulator which allowed to evaluate a key variable for estimating accidents risk under different conditions: the driver's behaviour in terms of ordinary prudence. In case of overtake maneuvers it has been built the “tree of events” and it has been defined the algorithms to calculate the probability of each “final event”. The model has been calibrated analyzing accidents data base along homogeneous road sections. The analysis has been carried out for a total length of 800 km. All the sections have been organized into four groups characterized for functionality and geometrical properties. The results show good convergence between observed and calculated values. The results of this predictive model are more precise as the homogeneity of the road sections increase.