The goal of this study is to analyze the factors affecting the accidents based on design vehicle. The traffic accident data (from 2013 to 2015) are collected from the “traffic accident analysis system (TAAS)” data set of the Road Traffic Authority...
The goal of this study is to analyze the factors affecting the accidents based on design vehicle. The traffic accident data (from 2013 to 2015) are collected from the “traffic accident analysis system (TAAS)” data set of the Road Traffic Authority. Count data model is utilized in this study to develop the traffic accident models. 17 explanatory variables such as geometry and traffic volume factors are used. The main results are as follows. First, the null hypothesis that the design vehicle does not affect the number of accidents of roundabout is rejected. Second, 4 count data accident models are developed, which are all statistically significant. Third, the inscribed circle diameter, entering volume, mean entry lane width, mean exit lane width, the number of circulatory roadway, the entry radius, the number of bus stop, the number of leg and crosswalk, and circulatory roadway width are analyzed to see how the variables affect the accident of roundabout. Through the development of models by design vehicle, the fact that the accident factors at roundabout is different from design vehicle is revealed. Also the test results of fitness using RMSE and others show that Poisson model in the case of small size vehicle, and negative binomial model in the case of large size vehicle, and semitrailer and total model are appropriate to the given data.