Statistical Models for Preferred Time Headway and Time Headway of Drivers in Steady State Car-Following

This paper proposes two shifted continuous distribution models for preferred time headway (Tp) and time headway of drivers in steady state car-following (Tws), i.e. the lognormal and gamma models. Lognormal distribution with a shift of 0.21 seconds provides a good fit to Tp, whereas a shift of 0.26 is proposed for Tws. Similarly, shifted Gamma distribution is proposed for Tp and Tws with shifts of 0.21 seconds and 0.26 seconds, respectively. Detailed data sets (FHWA Data) observed at one-second intervals collected from aerial photography for four sites have been used in this paper. The data sites are a basic freeway section, a ramp merge, a lane drop and a ramp weave section. Tp and Tws were found to be the shortest (mean, median and st. dev.) on the ramp weave section compared to all other sites. It was also found that the shifted Lognormal distribution provided a better fit to all the observed sites compared to the shifted Gamma distribution.