Journal of Korean Society of TransportationISSN:1229-1366(Print) 2234-4217(Online)대한교통학회

This study was conducted in an attempt to identify the factors influencing traffic accident severity according to gender (All, Male, Female) of bus drivers, and to propose a method for future introduction of advanced driver assistance system. Traffic accident data was gathered from the ｢Traffic Accident Analysis System (TAAS)｣ of the Korea Road Traffic Authority, of accidents caused by local/inter city buses in 31cities of Gyeonggi-do Province within 3years (2014－2016) of time. Analysis was carried out by first reviewing the gender variable with the decision tree model, then analyzing the accident severity according to gender using the ordered logit model. The results of the decision tree model show that light violation and vehicle to vehicle variables have the greatest influence in both All and Male drivers, and that the chances of severe injuries and casualties increase the more the driver violates the traffic lights, and the less the accident is a vehicle to vehicle type. On the other hand, there were no selected variable for female drivers. Results of the severity model analysis show that light violation variable holds significance in all three categories. In both All and Male category, Vehicle to Pedestrian, Speed violation, and Median Encroachment were selected as the explanatory variable. In the Male category, Intersection variable was additionally selected. Also, Intercity bus and Safety distance violation were selected as unusual variable for Female drivers. Thus, with the analyzed results, this study offers a policy implication for the introduction of the advanced driver assistance system according to the accident severity of each gender.