Abstract
This paper presents a general approach to simulate heterogeneity of car-following behaviors in freeway traffic. This is achieved by proposing a new car-following model called General Individual Based car-following Model (GIBM) which takes into account the characteristics of the individual driver. The model parameters are expressed as functions with regard to “aggressiveness level” of drivers. The numerical simulation of the perturbation propagation in dense traffic flow by using the GIBM shows that small perturbations could develop into stable moving jams, growing moving jam, and decaying moving jams randomly, due to the heterogeneity of individual car-following behaviors in traffic flow. The result of the simulation is consistent with the empirical observation. The GIBM appears promising for application to microscopic simulations of traffic dynamics including randomness. It presents a general approach to simulate the heterogeneity of car-following behaviors efficiently and simply.