A joint tour-based model of tour complexity, passenger accompaniment, vehicle type choice, and tour length
MetadataShow full item record
Tour-based model systems are increasingly being deployed to microsimulate daily activity-travel patterns of individuals. There are a host of tour attributes of interest that are modeled within these systems. However, a dimension that is often missed is that of vehicle type choice, a variable of considerable importance in the energy consumption and emissions estimation arena. Another issue that arises is that most tour attributes are modeled independently or sequentially with loose coupling across the models, thus ignoring important endogeneity effects that may exist across multiple tour dimensions. This paper considers four key dimensions of tours – tour complexity, passenger accompaniment, vehicle type choice, and tour length – with a view to developing a joint simultaneous equations model system of tour choices while accounting for the presence of correlated unobserved attributes affecting multiple dimensions through appropriate error covariance structures. The paper makes an important methodological contribution by describing and formulating a multi-dimensional joint choice model system capable of accommodating a variety of endogenous variable types (discrete and continuous). The paper makes an important empirical contribution by providing evidence on the nature of the relationships among these tour dimensions of interest within the context of a joint model. The model system is estimated on a sample of tours from the 2009 National Household Travel Survey of the United States. In general, it is found that there is significant evidence of correlated unobserved factors across these tour dimensions and that vehicle type choice affects tour length, a finding that could have important policy implications.
At the time of publication Rajesh Paleti and Chandra R. Bhat were at the University of Texas at Austin, and Ram M. Pendyala and Karthik C. Konduri were at Arizona State University.