An agent-based model of bicyclists accessing light-rail stations in Salt Lake City

Many of the existing transportation analysis methods rely on aggregate data analysis. While these models have been well established in the transportation community, there are many components of transportation that these methods neglect. Among these marginalized components are multimodal trips. Using more than one mode of transport to get from origin to destination is not nearly as common as a single-mode trip, but multimodal trips can have a significant impact on the transportation system as a whole. Additionally, traditional analysis methods have difficulty capturing nuances in travel behavior that are best observed at an individual level such as trip-chaining. This thesis shows how an agent-based model (ABM) can be used to analyze a multimodal transportation problem. Specifically, the model presented here describes the behaviors of bicyclists in Salt Lake City, Utah who use the TRAX light-rail system in conjunction with cycling to complete their trip. The model in this thesis includes the effects of elevation on mode choice, which is often ignored in traditional analysis methods. The model serves as a proof-of-concept that an ABM is a worthy means of analyzing multimodal trip patterns. The model is constructed in the NetLogo environment, which shows the usefulness of freely available software. Lastly, the outputs of the model show that an ABM can successfully capture trends that cannot be observed using traditional methods.