Stochastic modeling of traffic flow breakdown phenomenon: Application to predicting travel time reliability

This paper presents a modeling approach to generate random flow breakdowns on congested freeways and capture the subsequent wave propagation among heterogeneous drivers. The approach is intended for predicting travel time variability caused by such stochastic phenomena. It is assumed that breakdown may occur at different flow levels with some probability, and would sustain for a random duration. This is modeled at the microscopic level by considering speed changes that are initiated by a leading vehicle and propagated by the following vehicles with correlated-distributed behavioral parameters. Numerical results from a Monte Carlo simulation demonstrate that the proposed stochastic modeling approach produces realistic macroscopic traffic flow behavior and can be used to generate travel time distributions.