Reducing urban traffic congestion improves overall transportation system efficiency and reliability in support of improving the nation's economy, the environment and the societal welfare. The objectives of this Faculty Early Career Development Program (CAREER) award are to: 1) investigate relationships that exist between congestion dynamics at the local (e.g., individual intersections or roadways) and regional (e.g., city-wide) spatial scales; and, 2) use this knowledge to identify strategies that can effectively combat urban traffic congestion at individual intersections, along corridors and across entire regions simultaneously. The relationships between local and regional congestion dynamics will be integrated into a novel multi-scale congestion modeling framework for optimizing and refining contemporary traffic control policies as well as supporting the development of advanced congestion management strategies that will become available as connected and automated vehicles are integrated into urban environments. The proposed research will be integrated with a set of educational activities to advance our knowledge in traffic management and promote national prosperity and welfare. Specific education activities include collaboration with local transportation agencies who will provide data to verify the identified relationships and support real-world project-based learning activities and demonstrate a holistic, multi-scale view of traffic congestion. These activities will be shared through the creation of an electronic repository of undergraduate course materials on traffic operations. The PI will also develop interactive, online simulations to demonstrate the relationships between local and regional congestion dynamics and integrate these into "Expanding Youth Horizons" STEM outreach workshops targeted to local female middle school students.
This research will integrate novel models of regional congestion dynamics that relate network productivity and accumulation--the network Macroscopic Fundamental Diagram (MFD)--with existing link-based modeling paradigms. These two disparate approaches will be combined by identifying and quantifying interdependencies between local and regional network properties and congestion dynamics along each of these spatial scales. Theories and methodologies will be developed to unveil network characteristics and local congestion patterns that are conducive to the existence of reproducible and predictable regional congestion models for realistic urban traffic networks. This includes the identification of the building blocks necessary to characterize regional traffic network performance for networks made up of hierarchical streets and that serve heterogeneous traffic streams. Insights obtained for these heterogeneous network structures will also serve as the basis to study and design more sophisticated systems, particularly multimodal traffic networks with multiple unique and interacting vehicle classes. Integration of this regional MFD approach with existing link-based modeling paradigms will provide a framework that facilitates automated traffic control policies to coordinate regional and local traffic performance.
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.
