Quantifying the explanatory power of mobility-related attributes in explaining vehicle ownership decisions

Abstract

This paper shows the explanatory power of mobility-related attributes in explaining vehicle ownership decisions. In contrast with the state of the art, in which socio-demographic information is usually employed to explain vehicle ownership decisions, we highlight how the quality of travel offered by a given mode of transportation in serving one's activity pattern influences the ownership of a mode of transport. To implement this approach, we employ a two-level model: first, we estimate the mode choice for performing a set of activities, and second, we use the results of the mode choice calibration to estimate an ownership model. We also show the need to estimate this two-step model, originally modelled as a two-step multinomial logit, with more complex logit structures. With the aim to consider all possible correlations included in each step, a Mixed Logit model and a Cross-nested logit model are employed. Finally, the influence of the socio-demographic information on the quality of travel offered by a given transport mode in performing a particular activity is discussed.