Gap Acceptance on Urban Expressway Merging Sections: An Application of Inverse Time to Collision

Among driver behaviors at merging sections, gap acceptance is regarded as the most important ones. Although, many studies have been carried out to investigate the gap acceptance of merging vehicles (MVs), various influencing factors such as geometries and traffic conditions have not been thoroughly considered. This paper tries to close this gap by proposing an approach for modeling gap acceptance of MVs based on valuable microscopic trajectory data extracted from video data. The video data were collected at two urban expressway merging sections in Nagoya City, Japan (left hand traffic) in three periods of extending the acceleration lanes including “before”, “during” and “after” with wide ranges in geometries and traffic conditions. In this study, the effects of geometries and traffic conditions (density) were thoroughly taken into account. They are found to be significant effects on the choices of MVs. Moreover, the study adopted the concept of inverse Time to Collision to represent the reactions of MVs to the mainline vehicles (MLVs). The inverse Time to Collision between the MV and the leading MLV (TTCL⁻¹) as well as between the MV and the following MLV (TTCF⁻¹) were applied in the models. The estimation and sensitivity analysis results indicate that the decision of MVs to reject or accept the gap is significantly dependent upon TTCL⁻¹ and TTCF⁻¹. Generally, the MVs accept the gap in the cases of low potential collision with both leading and following MLVs. By contrast, they reject the gaps to avoid collision with the leading and following MLVs.