Safety effects of protected left-turn phasing at signalized intersections: an empirical analysis

Abstract

Left-turn crashes occur frequently at signalized intersections and often lead to severe injuries. This problem can be addressed through the implementation of protected left-turn signals. This gives left-turning vehicles the right to enter the intersection free from conflict with opposing drivers and pedestrians. The present study analyses the effect of this measure on crash occurrence. The study included 103 signalized intersections with left-turn signals in highways in Flanders, Belgium, of which 33 received only changes in the signal control and 70 also received additional changes. The effect on traffic safety is analyzed through an Empirical Bayesian before-and-after study on crashes, in which general trend effects and regression-to-the-mean are controlled. On the 33 intersections that received only changes in the signal control, the number of injury crashes decreased significantly (−46%, 95% CI (−36%; −55%)) during the after-period. This was mainly attributable to a decrease in left-turn crashes: −60%, 95% CI (−39%; −74%). The number of rear-end injury crashes did not change significantly after the implementation of a protected left-turn signal. A larger effect was identified for more severe crashes (involving serious injuries and fatalities) as compared with crashes resulting in lighter injuries: −66%, 95% CI (43%; −80%). Furthermore, the effect of left-turn phasing on the number of injured car occupants, cyclists, moped riders and motor cyclists was examined, and favorable effects were found for each of these groups.