Accident prediction models for urban roads

Abstract

Traffic accidents prediction has an important meaning to the improvement of traffic safety management, and urban traffic accidents prediction model. Different approaches for developing Accident Prediction Models (APMs) are used such as multiple linear regression, multiple logistic regression, Poisson models, negative binomial models, random effects models and various soft computing techniques such as fuzzy logic, artificial neural networks and more recently the neuro-fuzzy systems. This paper reviews application of these approaches for developing APMs and advantages of neuro-fuzzy system in modelling accidents in urban road links and intersections.