Construction and estimation of discrete choice models for assessing and forecasting the effects of sustainable mobility strategies

The current transport system, based on the use of vehicles fueled almost exclusively by petroleum products, generates non-negligible social, environmental and economic impacts. To overcome the problems associated with the current system, in recent years, particular attention has been paid to sustainable urban mobility at different levels. Governmental bodies, policy makers, transportation operators and academic researchers have focused on strategies and policies to reduce automobile travel and induce a modal shift in travel behavior. In particular, there has been increasing interest in soft measures which use information and communication tools to heighten people’s awareness and encourage them to change their travel behavior. The primary objective of this thesis is to explore the methodological processes able to guarantee that sustainable mobility policies seeking to reduce car use are accompanied by a quantitative assessment of the effects that this generates in the transport context. In particular, we focused on the role and weight of psycho-attitudinal variables (attitudes, perceptions, habits, etc.) in the process of choosing a sustainable mode of transport. The first part of the thesis presents the findings of a study focusing on unraveling the linkage among psycho-attitudinal factors related to bike use and the choice to cycle. In doing so, we constructed and estimated different econometric models (Integrated Choice Latent Variable model, GHDM model, Multivariate ordered probit) each aimed at shedding light on those aspects overlooked in the research. The context of the study were the urban areas of Cagliari and Sassari, main cities in Sardinia (Italy), where, despite the implementation of policies supporting bike use, cycling levels for commuting trips are still low. Our modeling estimation results reinforce the idea that promoting cycling through the implementation of awareness campaigns and educational programs, intended to improve peoples’ perceptions of the bike mode, can persuade them to consider the bike as an alternative means of transport to private motorized vehicles. Further, investments aimed at supporting use of the bike for leisure (e.g. cycle routes) may increase the number of people who choose to use the bike as an alternative means of transport for commuting or shopping. The second part of the work attempted to assess the short-term effect on travel mode choice of introducing a new sustainable form of transport into the choice set (hard measure) when implementing a VTBC program (soft measure). The transport context chosen for this experiment is a corridor linking the city center of Cagliari (Italy) to a university/hospital complex, where a new light rail route went into service in February 2015. For assessing changes in travel behavior in the short term, the modal shares observed in the first and second wave surveys were compared. Our results show that the combination of hard and soft measures achieved a change in travel behavior of 34%, when the measure is not personalized, and 46% with the VTBC program. Finally, we evaluated the long-term effects of these measures and we investigated if any changes in the psycho-attitudinal factors and/or in socio-economic characteristics exist after implementation of those measures. In particular, the objective of the study is to analyze whether these changes in individual characteristics are able to affect mode choice from a modeling perspective, through the specification and estimation of Integrated Choice Latent Variable models that use, for the same sample, the data collected for these two moments in time. Our results indicate that psycho-attitudinal variables were not significantly different. over waves, showing that the impact of the psychological construct remained stable over time, even after the introduction of the new light rail. Additionally, we found some evidence that the variables that explain the psycho-attitudinal variables could change over time.