Evaluation of Blue Confirmation Lights on Red Light Running at Signalized Intersections in Lawrence, Kansas: A Case Study

Abstract

Red light running (RLR) continues to be a safety concern for many communities in the United States. The Insurance Institute for Highway Safety reported in 2012 that RLR resulted in 683 fatalities and an estimated 133,000 injuries nationwide. Currently, a wide range of countermeasures from low-cost strategies (e.g. signal timing adjustments, signal backplates, signage improvements, and targeted enforcement) to high-cost strategies (e.g. automated enforcement, lighted stop bar systems and intersection geometric improvements) exists to mitigate RLR violations and their related crashes. The severity of RLR crashes and high societal cost (about 14 billion dollars annually) have provided the impetus for increased red light enforcement programs throughout the country. In the State of Kansas, where automated red light cameras are not authorized, communities rely on targeted traffic law enforcement. Traditionally, red light enforcement has required a police officer to be located upstream of an intersection to observe the violation and another officer located downstream to pullover the offender and issue a ticket. This enforcement approach is labor intensive. In locations where enforcement resources are limited, having a single police officer monitor RLR raises intersection safety concerns. Confirmation light systems can aid a single police officer located downstream of an intersection to monitor RLR violations without having to travel through an intersection to pullover the offender. The city of Lawrence installed confirmation lights at six left-turn approaches of two signalized intersections (treatment sites) where RLR was prevalent. This study was conducted to evaluate the effectiveness of the confirmation lights. RLR violation data were collected before, one and three months after installation of the lights at the two treatment sites and 11 non-treated intersections which included six spillover sites (intersections nearby the treatment sites) and five control sites (intersections located far from the treatment sites or corridor under investigation). A Z test of proportion was used to determine if the changes in RLR violation rates from the before period to the after periods were statistically significant at the 95 percent level of confidence. The violation rates at the two treated sites were then compared to non-treated sites. Violation time into red (how long it took a driver to run a red light after red signal indication) was used as a secondary performance matrix to evaluate the confirmation lights. A Chi Square Test of Independence was used to analyze the violation times into red at the 95 percent level of confidence. Results of the analysis showed a 57.4 percent reduction in left-turn RLR violation rates at the treatment sites and a 55.7 percent reduction at the spillover sites one month after installation of the confirmation lights. The three months after study indicated a 42.7 percent decrease in violation rates at the treatment sites and a 31.7 percent decrease at spillover sites. Considering RLR violations in the City of Lawrence as a whole, the control sites showed no significant change in left-turn RLR violations during the study periods. Although the treatment intersections did not have confirmation lights installed for the through movement, the lights were effective in the short term for reducing RLR violations (84.1 percent reduction). The Chi Square Test showed that that the confirmation lights had no significant effect on the RLR violation times into red. In conclusion, the findings of this research study indicated that confirmation lights (both self-enforcing and aiding police officers) have a positive effect in reducing RLR violations at targeted signalized intersections.