Abstract
This thesis introduces a new method to analyze network-wide bicycle infrastructure to quantify dangerous situation exposure for bicyclists. The method is intended for sketch-level scenario planning. Lack of bicycle volume data is a common impediment to calculating exposure, our method overcomes this by extrapolating short-duration citizen-volunteer count data to estimate community-wide bicycle volumes. First, the count data is extrapolated spatially using an origin-destination centrality technique. Second, the count data is extrapolated temporally using adjustment factors for hour, day, and month. This two-step extrapolation produces a rough estimate of Annual Average Daily Bicyclists (AADB) for streets, trails, and intersections across a community. Next, we propose using public participation to define community-specific "dangerous situation metrics" that can be used to compare AADB exposure for alternative improvement scenarios. We demonstrate the process with a case study by comparing exposure under current conditions and after implementing a proposed bicycle improvement master plan. For example, the case study showed a 5% decrease in AADB exposure to the dangerous right hook situation (bicyclists going straight through an intersection where there is a high volume of vehicles turning right). As another example, the improvement master plan was shown to potentially reduce the need for bicyclists to cross harsh intersections by 7%. The method introduced in this thesis can provide engineers, planners, and other decision-makers a means to compare improvement scenarios for investment decision-making. Furthermore, the literature review and discussion provides a starting point for communities to define their own dangerous situation metrics.