Stochastic Fundamental Diagram for Probabilistic Traffic
Dr. Daiheng Ni, Principal Investigator
Assistant Professor, Department of Civil and Environmental Engineering
University of Massachusetts/Amherst
Sponsor: New England University Transportation Center, Region 1
Advanced traffic signal control systems, innovative traveler information systems, and other intelligent system applications are designed in part to address disruptive changes in our transportation system. Central to the successful deployment of such intelligent systems is the need to employ traffic flow principles and models that capture the random nature of traffic flow. Existing traffic flow models based on conventional fundamental diagrams are deterministic in nature, and thus are unable to account for the inherent random nature in traffic flow. To address this limitation, this research proposes to develop a new fundamental diagram which represents the probabilistic relationship between traffic speed and density as a stochastic process. Such a stochastic fundamental diagram not only explains the variation of speed choices among driver populations but it also has the potential to improve the traffic flow model's predictive power. This proposed research is designed to respond to the U.S. Department of Transportation's objectives concerning congestion mitigation and the New England University Transportation Center's theme to improve strategic management in response to disruptive changes in the transportation system. The products of this research will contribute a new way to think about and to understand the complex interrelationships among traffic flow characteristics and will provide a basis to advance traffic flow modeling to a new level.