Bike-Sharing Is Transit: Building Tools to Plan and Optimize Bike-Sharing Networks

Abstract

Urban America is infamous for its broader lack of public transit connectivity. Millions of Americans are termed as “transit-dependent”, which means they do not have immediate access to a vehicle or otherwise cannot drive and must use an alternative means of transportation. These Americans are likely barred from vital services and getting to work. As urban populations grow, the importance of sustainable, accessible urban transportation options grows, and bike-sharing systems provide an effective solution to both, offering an environmentally friendly, healthy, and congestion-limiting option for commuters. This thesis explores how interest groups and political stakeholders affect transit planning, as well as how those policies diffuse amongst larger and smaller cities. Then, using Boston Public Schools teachers as a case study due to their geographic distribution, this thesis details a multi-agent simulation model as a novel planning mechanism for bike-sharing which accounts for different weather and ridership based scenarios. Finally, this thesis offers a cellphone GPS data-driven method of discovering and predicting bike-sharing demand at possible station locations.