Speeding across Texas: identifying high-risk locations using probe data

Speeding contributes to one-third of all motor vehicle fatalities in the U.S., making it crucial to understand speeding behaviors for traffic safety. This work compares a random sample of 39.1 million vehicle speeds in November 2024 to posted speed limits (PSLs) at 1.04 million roadway points across the Texas network, and reviews non-infrastructure strategies to ensure greater PSL compliance. Weighted least squares (WLS) and spatially weighted binomial logistic regression models control for design variables, congestion, and land use to examine how average speeds and shares of vehicles exceeding the PSL vary by time of day, day of week, and a host of other variables. Results indicate that speeding is most common during late-night hours and on weekends, especially on roads with lower PSLs (30 and 40 mph), where 43% of drivers exceeded the limit. Almost half the probe vehicles exceeded the PSL and 20% exceeded it by more than 15% (e.g., by more than 9 mph in a 60 mph zone) between 3 a.m. and 5 a.m. each day. PSL and rural settings were the top predictors of increased average traffic speed, with a 1 mph increase in PSL associated with a 0.72 mph rise in average speed - everything else constant. Access control and weekend variables were the factors most strongly associated with increased speeding behaviors, with weekends showing 7.7% higher speeding and access-controlled roads exhibiting 5.7% to 9.8% higher speeding, depending on the level of access control. This paper also reviews the effectiveness of non-infrastructure speed management strategies (like automated speed enforcement), and connects speeding behaviors to appropriate non-infrastructure strategies, highlighting enforcement for high-speed hotspots and community or operational measures for lower-PSL local roads. This work provides insights to help transportation agencies identify targeted enforcement locations and implement more effective speed management policies.