Safety Effectiveness of Flashing Yellow Arrow

Abstract

The purpose of this project was to develop crash modification factors (CMFs) for the implementation of a flashing yellow arrow (FYA) on the basis of the specific before-and-after period conditions of a signalized intersection. Although this countermeasure has been used for years in North Carolina and other states, no published studies to date have provided CMFs for left-turn crashes specific to the treated approaches, and none has provided CMFs for the three-section FYA for permissive-only left turns. Crash data from 222 intersections in North Carolina with an FYA protected–permissive left turn (FYA-PPLT), three-section FYA permissive-only left-turn installations, or both were used to provide CMFs for five category types: Category 1 (permissive only to FYA-PPLT), Category 2 (protected only to FYA-PPLT), Category 2A (protected only to FYA-PPLT with time of day operation), Category 3 (five-section PPLT to FYA-PPLT), and Category 4 (permissive only to FYA permissive only). A before-and-after crash analysis with consideration given to increases in traffic was used to determine the safety estimates. Safety performance functions were used to account for the effect of traffic volume trends. In Categories 3 and 4, the change was exclusive to the left-turn display and not to a change in phasing. All CMF results were statistically significant for Category 3, and target and injury CMF results were statistically significant for Category 4. On the basis of the results from the study sites, a statistically significant decrease was found in target left-turn crashes and injury crashes after a site signal underwent a change from a solid green ball to an FYA for permissive left turns when phasing remained unchanged. This finding applied whether the left phasing was protected–permissive or fully permissive.