Fostering sustainable mobility for secure and livable communities is key to address the current environmental and energy crises. There are successful examples of cities for which cycling is playing a major role in their paths toward sustainability. For example, 5.8% of commuters in Portland cycle to work. The percentage in New York City is only 0.6%, despite 345 miles of bicycle routes being added in the last decade. To encourage the use of non-motorized alternatives we need to better understand the motives underlying demand. Econometric travel demand models are highly valuable for assessing the effect of policies and incentives seeking to reduce the indiscriminate use of car. In fact, forecasting demand using discrete choice models has proved to be successful in the case of modal split among motorized alternatives. However, there are several challenges in applying choice modeling to non-motorized options. Users of the transportation system may be motivated to cycle or walk not because of the tradeoff between cost and time, but because of health and environmental benefits of these alternatives. At the same time, there are several factors that may discourage the use of non-motorized transportation, such as poor accessibility, safety concerns, and unfavorable route and weather conditions. For instance, it is often argued that the North East has poor climate to encourage the use of biking. Accounting for factors beyond traditional compensatory attributes is not straightforward and requires a deep understanding of user behavior.

This research project will focus on two related problems that are relevant for better informing policies targeting sustainable transportation as well as safer and more livable cities. The first research project is to derive a new latent segmentation approach to discrete demand to model non-motorized transportation choices and characterize both utilitarian and recreational cycling users. For this project, a survey instrument will be designed to collect stated-preference and attitudinal data to test new estimators. The second research project is to improve the analysis of cycling demand subject to weather conditions by building a new methodology to study and update time series of automatic cycling counts, which will be tested using case studies of 2 major cities in North America. Results of both projects are expected to contribute in the methodological and practical characterization of current and potential bicycle users, as well as to provide essential information capturing the motives and barriers associated with cycling decisions.