Fuzzy probabilities have been widely used in the areas of risk assessment and decision making. Here, we propose a system based on fuzzy probabilities for assessing the intensity of pedestrian collisions and levels of injury. The proposed approach uses possibility probability distributions for demonstrating fuzzy probabilities. A composition of possibility probability distributions (fuzzy probabilities) of collision intensities coming from different environmental variables yields the final fuzzy probability of intensity with higher reliability. The proposed approach is then applied to an actual database of pedestrian collisions from the City of Mashhad for over a period of five years. Keywords_ Collision Intensity, Collision Variables, Fuzzy Probability, Possibility Probability Distribution.
Farzaneh Tatari, Mohammad R. Akbarzadeh-Totonchi