This paper presents a before-after crash and injury study of intersectionsignalization in an urban area. Signalization of 54 intersections in Copenhagen, Denmark, is analyzed. Police recorded crashes, which occurred at these intersections as well as crashes up to 500 meters away on roadways leading up to the intersections being signalized, are included in the study.The method used accounts for general crash and injury trends, changes in traffic volumes, and also regression-to-the-mean effects in the before period. The best estimates for safety effects of intersection signalization are decreases in crashes and injuries of respectively 21 and 17 percent at 3-armed intersections, and 39 and 33 percent at 4-armed intersections. These findings are in line with previous studies. These safety benefits at intersections are primarily due to very significant reductions in right-angle crashes. Safety is also improved on roadways up to about 100 meters awayfrom the 3-armed intersections, which were being signalized, and up to approximately 200 meters away from the 4-armed intersections. Around 60 percent of the total safety improvement that took place due to intersection signalization occurred on the roadways and only about 40 percent at the intersections. In total, for both intersections and roadways up to 200 meters away, the best estimates for safety effects of intersection signalization are decreases in crashes and injuries of respectively 19 and 21 percent near and at the 3-armed intersections, and 26 and 33 percent near and at the4-armed intersections.
Abstract