Analysis and Modeling of Driving Behavior using In-Vehicle Data Recorder

This paper develops a general theoretical framing of driver behavior expressing the relations among the different factors of driving behavior. Several sources of data should be used in order to estimate models based on this framework. This paper uses newly available In Vehicle Data Recorders (IVDR) data, as well as data about driving behavior, characteristics, attitudes and perceptions from traditional selfreported questionnaires. The modeling framework integrates these various data sources in addition to costs and environmental data to identify relevant driving behaviors that may be related with the risk of accident involvement.