This study considers the scenario of a multimodal transit system that integrates autonomous mobility on demand (AMoD) service with mass transit services as a first- and last-mile solution. A hybrid system such as this may potentially reduce the total miles traveled by AMoD services, which in turn may reduce congestion and emissions.
This concept is explored through the development of a model for assigning the user demand between an AMoD system and mass transit, then a framework for planning the fleet management tasks of the operator is developed, and finally the performance tradeoffs between user, operator, and city/governmental objectives is examined. Multiple objectives and rebalancing strategies are considered, with a focus on practical considerations for the fleet operator. Performance is analyzed with simulation of a hybrid AMoD and mass transit system for the Washington, DC area using data from car2go (carsharing operator) and Metro (transit provider).
Under baseline, free-flow conditions, 81.09 percent of trips are completed fastest by the AMoD service, 1.25 percent of trips are completed fastest by mass transit, and 17.66 percent of trips are completed the fastest by taking a multimodal approach.
During rush hour the distribution changes somewhat with 59.33 percent of trips being completed fastest by AMoD, 1.97 percent of trips completed fastest by mass transit, and 38.70 percent of trips completed fastest by taking a multimodal approach. During rush hour, more than twice as many trips are fastest by taking a multimodal approach and about 37 percent fewer trips are fasted with AMoD only.
An integrated AMoD and mass-transit system can provide up to a 50 percent reduction in total AMoD vehicle miles traveled, thus reducing the contributions to congestion and emissions from these vehicles.