Safety Evaluation for Conversion From Protected-Only Left-Turn Phasing to Time-of-Day Protected-Permissive Left-Turn Phasing Using Flashing Yellow Arrows

Abstract

This study investigated the safety effects of the conversion from a protected-only left turn to protected-permissive left turn with flashing yellow arrow (FYA−PPLT) with time-of-day operation. The observational before–after study with the comparison group method was used to develop crash modification factors (CMFs) for the total crashes and two types of target crashes that involve left-turning vehicles on treated approaches (left-turn−same roadway and rear-end crashes). For all potential crashes, crash reports, crash diagrams, and narratives were manually inspected to identify the actual target crashes. The CMFs were separately developed for a full 24-h day and specific time-of-day based on FYA-PPLT operation, and for two severity categories (all severities and fatal & injury). The results showed that the total crashes for 24 h had a slight increase, whereas the target crashes had substantially larger changes (increase in left-turn−same roadway crashes and decrease in rear-end crashes). When looking at the specific time-of-day use of FYA-PPLT, the increase in left-turn−same roadway crashes was significantly higher than the full 24-h period. This implies that engineers could extend protected-only use more on the fringes of the peak periods to mitigate the increase in left-turn-related crashes. The analysis results also showed a trade-off between left-turn–same roadway and rear-end crashes after the signal conversion. The findings reveal how important it is for engineers to properly understand the safety effects on the different categories of target crashes and time periods when considering the signal conversion from a protected-only left turn to FYA-PPLT.