Headway distribution modeling with regard to traffic status

Knowledge of vehicle headway distribution is very important for intelligent transportation and intelligent vehicle simulations. Various headway distribution models have been proposed, but most of them only fit for a certain traffic situation. To solve this problem, we study the dependence of headway distributions on traffic status in this paper. Results show that the log-normal distribution model is adequate in fitting headway data when the traffic is in free flow status; while the log-logistic distribution model is more suitable in fitting headway data when the traffic is in congestion status. This conclusion is useful in the traffic signal optimization algorithm, since it indicates that we should apply different delay estimation models during different traffic status so as to design optimal timing plan.