کاربرد مدل های رگرسیون پرصفر در مدل سازی تصادفات بزرگراه های درون شهری دوره 11، شماره 4، پاییز 1390،

we handle in this paper, the research that have been performed to recognize the factors that affect crash frequency and severity in urban highways and use crash data of mashhad urban highways as a case study. Statistical models that have been used in this research include Poisson, Negative binomial, Zero-inflated poisson and Zero-inflated negative binomial regression models. Traffic flow related and road geometric related variables have been used as independent variables of models. Special task that have been done in this research ,is separation of total traffic volume into passenger cars, heavy vehicles and light non-passenger car vehicles volume to have an especial and thorough look at the role of traffic volume in crash occurrence , whether crashes with property loss only or more severe crashes, and see ,precisely which part of traffic have an effective role or more effective role in crash occurrence. In this research, the efficiency of Zero-inflated models is investigated against Poisson and Negative binomial regression models in modeling accidents on urban highways. Results of the research show that the number of access roads and horizontal curves on highway segments and the volume of light non-passenger car vehicles have increasing role in likelihood of accidents. Furthermore, Zero-inflated negative binomial regression model is the best and fittest model for modeling accidents, whether with property loss only or more severe and consequently the efficiency of Zero-inflated models in modeling accidents on urban highways of Mashhad is approved.