Tradable mobility credit (TMC) schemes as a way to manage travel demand have received significant attention in the transportation domain in recent years. In a TMC scheme, a regulator initially provides an initial endowment of mobility credits or tokens to all potential travelers. In order to use the transportation system, users need to spend a certain amount of tokens that could vary with the choice of travel mode, route, departure time etc. The tokens can be bought and sold in a market at a price that is determined by token demand and supply. This project will design, model and evaluate a system of TMCs--named Trinity--that operate in real-time. Trinity is personalized for each user and is designed to improve system-level objectives such as social welfare using future predictions of the transportation system. This study has the potential to accelerate the understanding and real-world deployment of TMCs and aid transportation agencies and operators in achieving long-term societal goals of sustainability by mitigating congestion and reducing energy and emissions. Algorithms and software developed under the project will be open-source and publically shared. They will be integrated with a transportation system prediction and control platform, DynaMIT, which will provide a useful tool for researchers to further extend the scientific knowledge. Educational benefits from the project include the training of graduate students and the involvement of postdoctoral researchers in undergraduate and graduate teaching activities and advising.
Specifically, the project pursues the following goals: (1) Design and implement a novel bi-level optimization framework tailored for 'online' applications that includes two components: a system-level optimization that periodically (in real-time) determines 'optimal' token (mobility credit) rates for different mobility options by utilizing short term predictions of the transportation network, and a user-level optimization that will determine a personalized 'optimal' menu of mobility options to display to each individual user subject to the optimal system-level token charging policy; (2) Model the operation/dynamics of the token market by considering actions of the user (individual buying and selling decisions and plausible behavioral models incorporating heterogeneity) and regulator; and (3) Perform extensive simulation?based experiments on real-world networks to gain insights into how the design of the TMC schemes, user behavior, and network conditions can impact performance and market dynamics (for example: how should the allocation/acquisition/expiration of tokens be designed; how and when does the regulator intervene in the market, how should the token policies be designed to allow for scalability/effectiveness of optimization and network control?). The Trinity system, in employing complex simulation-based disaggregate transportation and market models, will allow a community of researchers to gain insights into system behavior when the token charges, traffic flow patterns and market price evolve together?a topic of immense interest which has received limited attention thus far.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
