Understanding Commuters' Daily Travel Time: Application of a Hazard-Based Duration Model

Travel behavior researchers have long been intrigued by the amount of time people allocate to travel in a day. This paper revisits this issue by introducing the hazard-based duration model to travel time analysis. In the hazard-based duration model daily travel time is analyzed through the likelihood of ending the duration depends on the length of elapsed time since start of the duration, which incorporates the impacts of covariates and enables the test of travel time dependence on the elapsed time. Using travel data from the Beijing Third Comprehensive Transport Survey, this paper sheds new light on daily travel time by analyzing the relationships between daily travel time and socio-demographic attributes and activity and travel characteristics. Furthermore the estimated results illustrate that the behavior of most commuters can be represented by travel time minimization mechanism. It is demonstrated that hazard-based duration model, integrating the notion of temporal dynamics, is a suitable tool for duration data analysis in travel behavior research.