Planners and modelers at all levels of government have been grappling with the problem of how to better represent bicycling when modeling and forecasting future scenarios. Planners want to know how and to what extent infrastructure projects can increase bicycling activity and reduce the health, environmental, and financial burdens imposed by single-occupant vehicle travel. Fueled by rapidly advancing research on bicyclist preferences and behavior, efforts to develop modeling capacity have largely evolved along two parallel tracks: 1) strategic planning models at the state and regional level, and 2) travel demand models at the regional and local levels. Strategic planning models are typically easier to apply, more transferable, and produce a wide range of policy outputs but to date lack the ability to capture bicycle travel at a detailed level. Travel demand models, on the other hand, have in a few locales developed highly sophisticated bicycle demand models but have been challenging to apply, transfer, and respond to a full range of planning needs in practice. This research proposes to bridge the gap between these two application branches, transferring ideas between the models and from emerging research to advance both tracks and improve the consistency between them.
Project researchers and partners have identified a unique opportunity in Oregon for research to engage two parallel model development processes. The Oregon Department of Transportation (ODOT) has been developing a family of strategic planning models under the banner of VisionEval. The models accept demographic, land use, transportation system, and various policy inputs to produce detailed estimates of vehicle miles traveled (VMT), emissions, and travel cost outcomes. At present, bicycling is incorporated only via a fixed future mode share expectation and is insensitive to specific network infrastructure or connectivity changes. Meanwhile, Portland Metro MPO (Metro) has developed a sophisticated bicycle sub-model as part of their trip-based travel demand model, sensitive to highly detailed network variables including facility types, connectivity, traffic volume, slope, and intersection controls. ODOT would like to improve the sensitivity of their strategic model to cycling-related factors. Metro would like to increase the versatility of their bicycle model and better integrate it into the larger regional modeling process. Both see opportunities to learn from and contribute to the two model streams. This research would leverage the combined knowledge and experience of both agencies and the research team to advance the state of the practice for strategic and regional bicycle modeling and provide a template for future modeling efforts.