Applied and Interdisciplinary Mathematics Seminar

University of Michigan

Fall 2004
Friday, 8 October, 3:10-4:00pm, 1084 East Hall

Modeling the flow of freeway traffic

L. C. Davis

University of Michigan


Abstract

Modern analyses of traffic flow frequently involve simulations of a large number of individual vehicles whose motion is dictated by simple rules. Large-scale cellular automata calculations can be done to study emergent phenomena such as propagating jams, which are analogous to kinematic waves in the hydrodynamic theory of Lighthill and Whitham. To describe small-scale behavior, car-following and optimal velocity models, where equations of motion are integrated, are appropriate. Using a modified optimal velocity model that explicitly includes driver reaction time, I will explore the dynamics of jamming and synchronous flow. The latter is a new phase of traffic flow identified by Kerner in instrumented-autobahn data. The effects of adaptive cruise control on traffic flow will be discussed.