Most of the existing tunnel safety studies have focused on single tunnels, while only few have examined safety for freeway tunnel groups which consists of consecutive tunnels. Furthermore, no study has identified factors associated with the injury severity of crashes in freeway tunnel groups, which is crucial for employing effective countermeasures to improve safety. Therefore, this study aims to investigate factors influencing the injury severity of crashes in the freeway tunnel groups using five years of crash data (2012-2016) from Hunan Province, China. A mixed logit model was developed to identify potential factors for the injury severity. The developed model was compared with the conventional logit model, and it was found that the mixed model significantly outperforms its counterpart. The results indicate that: (1) the summer, weekdays, and daytime are positively associated with a higher crash severity, while the winter is negatively associated; (2) severe crashes are more likely to occur in the access, entrance, connecting zones, and downgrades while they are less likely to occur on curves; (3) drivers with less driving experience, involved in speeding, and fatigue driving bring about a higher likelihood of severe crashes; and (4) sideswipe collisions are associated with a lower tendency of severe crashes, while rollover with a higher tendency. Subsequently, multiple countermeasures are recommended to improve the safety in tunnel groups.

Traffic safety;tunnel group;crash severity;random parameter model