ABSTRACT. The starting point for this presentation is that radicle changes in technology, services, information and global policy analysis needs are presenting unprecedented disruptive challenges to the current travel demand modelling state of the art and practice. Improvements in travel demand models sufficient to meet these challenges first and foremost need to begin with improvements in the fundamental theories underpinning the models. They also need to exploit new, massive and passive, dynamic sources of data concerning travel that provide promising new avenues for research and for model and theory building. The next generation of travel demand theories and models may well (need to) be more dynamic, statistical, holistic and computationally powerful than the current generation of models. Ideally, they should also find their place within and contribute to the emerging “science of cities”.

ABSTRACT. It could be legitimately claimed that we are at a very fertile period in consumer choice modeling, with several important and exciting developments within the past decade in the data sources, theory, specification, estimation, and application of consumer choice models. These developments have taken place in many disciplines, including marketing, transportation, regional science, psychology, economics, statistics, political science, and sociology. In this presentation, the author will discuss the multi-disciplinary evolution of consumer choice models from being based on single source data to being based on multiple source data, from single discrete choice to multiple discrete-continuous choice, from single endogenous variable type to multiple endogenous variable types, from individual agent-based to social interaction-based, from traditional likelihood estimation to the use of other estimation methods, and from simulation methods for estimation to also considering analytic approximation methods for estimation, to name just a few. At the same time, it is critical that we do not lose sight of the need for consumer choice modeling to continue to be grounded on fundamental behavior theories, rather than being viewed as a "blind" data-driven exercise, especially as we move into a new landscape of data ubiquity. The issue of "causality" versus "association" becomes central in this discussion.

ABSTRACT. Understanding the determinants of activities and travel is critical for transportation policymakers, planners, and engineers to design and manage transportation systems. These sociotechnical systems, and their externalities, are interwoven with social systems in communities, cities, regions, and societies. But discrete choice models – the predominant modeling tool for researching travel behavior and planning transportation systems – are grounded in theories of individual decision-making. This research expands knowledge about the incorporation of social interactions into activity-travel choice models in the areas of informational conformity as well as social capital and social network indicators. This talk provides a review and discussion of how social networks are incorporated into discrete choice models of social interactions with specific application in travel behavior. The talk emphasizes the link between the context of decisions and social interactions to the social networks chosen for the analysis. Next, the state-of-the-art is described through two avenues of research: (1) methodological advancements in modeling social influence as taste variation and (2) efforts to develop a social capital theory of leisure activity behavior. The talk concludes with a look at the future of modeling social interactions in transportation systems and using social influence models for forecasting adoption of new technologies.

ABSTRACT. The field of choice modelling is facing the challenge of constantly improving its complex mathematical structures while correctly accommodating real-life behavioural aspects in these models. In this thesis, different methodological contributions aimed at improving the understanding and forecasting of discrete-continuous choices and representing heterogeneity are made, while substantial attention is devoted to a correct representation of social network processes and choices and to the investigation of the factors affecting time use. The thesis proposes both methodological and applied contributions making use of diverse revealed preference datasets from different countries, and puts forward ideas for improved data collection approaches. The findings from this thesis represent an advancement in the field of choice modelling and travel behaviour and a potential resource for policy making.