Evaluating the Impacts of Mobility-as-a-Service in Prototypical North American Cities via Agent-based Simulation

Abstract

Mobility-as-a-Service (MaaS) is a natural evolution of trends in shared mobility. As new mobility options emerge users are faced with an increasingly splintered mobility landscape. By unifying emerging on-demand modes with traditional mobility options, MaaS can increase users’ transportation accessibility, empowering users to pick between and use different modes. Furthermore, by offering users incentives MaaS has the potential to shift users towards more sustainable mobility patterns, improving network congestion and emissions. The objective of this thesis is to analyze the potential demand for MaaS in cities in the United States using agent-based simulation. Three different MaaS menus are designed, comprising of MaaS plans varying in price and incentive structure. The adoption of, subscription to, and impacts of these MaaS menus is simulated in three major city-types present in the U.S. using the state-of-theart simulation laboratory developed by the MIT Intelligent Transportation Systems (ITS) Lab, SimMobility. SimMobility is a multi-scale, multi-modal activity- and agent-based simulation software suitable for high-fidelity simulation of hypothetical transportation scenarios. This thesis demonstrates that MaaS has the potential to penetrate the United States, capturing both users of both sustainable and unsustainable modes. MaaS has the potential to improve transportation sustainability by promoting car-agnostic transit such as mass transit and Mobility on Demand to car users, while also having the potential to trigger switches away from sustainable active mobility modes.