Computational and Data-Enabled Analysis for Sustainable Transportation Systems

16. Abstract Transportation planners and traffic engineers are faced nowadays with immense modeling challenges arising from several emerging policy, planning, and engineering developments. In fact, the recent emergence of mobile sensing and traffic monitoring technology has provided an unprecedented amount of information and data for traffic analysis, demanding the adaptation of mathematical and physical models to a new generation of cyberinfrastructure. Some of the major challenges that traffic models meet include: computational tractability, very large-scale deployment, real-time application decision support, multiple time scales, and fusion of dissimilar data. This research takes a major step in developing analytical, holistic mathematical models and traffic analysis tools capable of addressing data-enabled traffic modeling, estimation, control and optimization problems, leading to a more efficient, reliable, and sustainable transportation system. In the project, we constructed: (1) a class of Mathematical Programming with Equilibrium Constraints (MPEC) problems to understand and mitigate congestion externalities and mobile source emissions, and (2) a class of Mixed Binary Integer Programs (MBIPs) for data fusing, real-time traffic estimation and prediction, as well as data-enabled traffic control. 16. Abstract Transportation planners and traffic engineers are faced nowadays with immense modeling challenges arising from several emerging policy, planning, and engineering developments. In fact, the recent emergence of mobile sensing and traffic monitoring technology has provided an unprecedented amount of information and data for traffic analysis, demanding the adaptation of mathematical and physical models to a new generation of cyberinfrastructure. Some of the major challenges that traffic models meet include: computational tractability, very large-scale deployment, real-time application decision support, multiple time scales, and fusion of dissimilar data. This research takes a major step in developing analytical, holistic mathematical models and traffic analysis tools capable of addressing data-enabled traffic modeling, estimation, control and optimization problems, leading to a more efficient, reliable, and sustainable transportation system. In the project, we constructed: (1) a class of Mathematical Programming with Equilibrium Constraints (MPEC) problems to understand and mitigate congestion externalities and mobile source emissions, and (2) a class of Mixed Binary Integer Programs (MBIPs) for data fusing, real-time traffic estimation and prediction, as well as data-enabled traffic control.