Departure headway distribution modeling for ordinary large cities based on video information

Departure headway distribution is significant for intelligent traffic control and traffic simulations. Jilin, an ordinary large city, is selected in this paper to study the departure headway distributions. Experimental results show that the log-laplace distribution model fits headway data better for free flow traffic condition, while the log-logistic distribution model is more suitable on the peak hours. The existed studies of departure headway distributions in extremely large cities, e.g. Beijing are different from those obtained in this paper. The results are useful for traffic signal optimization, since signal timing algorithms should be modified for different city scales as well as different traffic conditions.