Our work on incorporating AI into the process of designing transit routes is now published. This helps transit agencies that periodically update their routes to both learn from and serve their users. It also makes a case for automated transit fleets that can grow organically as they learn and adapt to their markets. The work is funded by NSF CMMI-1652735 and C2SMART.
The Federal Transit Administration commissioned the BUILT lab to work on a white paper reviewing and exploring the range of public transit operations from fixed route to on-demand microtransit. The completed 156-pg report is now available to access. It includes almost 300 references and links to data (B63 bus route in Brooklyn) and code for a simulation tool to evaluate different transit operations (fixed route, flex-route, door-to-door microtransit). This should be useful to transit agencies (e.g. New York City Transit) looking for an overview of state-of-the-art methods in flexible solutions.
A link to the project and the compendium can be found here.