In the transportation business, the compensation for the traffic accidents has an effect on the gain and loss of a transportation company. Moreover, the traffic accident influences the life and properties of both drivers and passers-by. This study uses statistical methods and data mining techniques to discover the implicit information from the data of P Company’s transportation management system. This research also devises a driving behavior scale to analyze drivers’ characteristics, including four driving indicators (hard acceleration, rapid deceleration, speeding, idling) and the occurrences of traffic accident. Then, the decision tree model is used to classify the drivers based on their characteristics and driving behavior in order to find out the related factors that influence the occurrences of the traffic accident. Based on the analyzed result, the transportation company can evaluate the personal characteristics of newcomers before making the hiring decisions. Additionally, the result also can be referenced by the company for managing its employees and preventing the happening of traffic accident.