Stochastic Modeling of Traffic Dynamics

Stochastic models of traffic flow are used in a variety of applications, e.g., traffic state estimation, travel time reliability, and traffic control. This talk will present techniques used to develop stochastic models. A main source of uncertainty in traffic dynamics is heterogeneity among drivers. This is captured using parametric uncertainty, resulting in stochastic microscopic models. These, in turn, are used to develop probabilistic traffic relations and stochastic Lagrangian models of traffic dynamics. Applications of the stochastic models, namely data assimilation, are presented.

Speaker: Saif Jabari, New York University, Abu Dhabi