Incorporating the standstill distance and time headway distributions into freeway car-following models and an application to estimating freeway travel time reliability

摘要

Standstill distances and following time headways are two important microsimulation model parameters associated with driver aggression. This paper investigates the distributions of standstill distances and time headways and incorporates these distributions into car-following models to estimate travel time reliability. By incorporating standstill distance and following headway into car-following models as stochastic parameters, a speed-density region can be generated, based on which various travel-time-reliability measures can be calculated. Key findings of this study are as follows: (1) Both standstill distances and time headways follow fairly dispersed distributions. Therefore, it is suggested that microsimulation models should include the option of allowing standstill distances and time headways to follow distributions as well as to be specified separately for different vehicle classes. (2) By incorporating stochastic standstill distance and time headway parameters in car-following models, travel-time-reliability measures can be estimated more precisely and faster compared with using VISSIM.