Addressing Liability Issues of Proactive Safety Improvements (2026)
Chapter: I. INTRODUCTION
ADDRESSING LIABILITY ISSUES OF PROACTIVE SAFETY IMPROVEMENTS
Fine Points, Ltd., Jefferson City, MO, in conjunction with Pam Clay-Young and Rachel Catchings, University of Kentucky, Lexington, KY
I. INTRODUCTION
The foundation of every highway improvement project is engineering judgment. Engineering judgment, put simply, is the use of the education, specialized training, and experience of the engineer to inform the decision-making process of the agency. Engineering practices are based upon generally accepted industry guidance in roadway design, traffic, maintenance, and construction, as well as data that informs the decision-making process. Data can be used to analyze existing roads to identify collision clusters and features that could benefit from treatment. Data can also be used to plan road improvements, guide maintenance activities, route traffic, and improve the safety and reliability of the transportation system.
The use of data to guide the decision-making process is not a new concept—manually collected traffic counts, intersection turning movements, seatbelt use observations and information found in crash reports, and public outreach meetings have been used for decades to guide the planning and engineering decisions of agencies. Methods of data collection are presently much more robust than they have been in the past.
Rapidly evolving data collection practices provide transportation agencies with the ability to more quickly adapt to changing traffic conditions and trends. Instead of waiting for written reports, agencies can collect data from cameras, interactive smartphone applications, law enforcement agencies, and the solicitation of information from the public via websites and chatboxes. Data may also be gathered through smart work zones, sensors and video analytics, GPS systems, and artificial intelligence.
Federal regulations require the collection and reporting of safety data as a condition of the receipt of federal funds. The predictive methodology found in the Highway Safety Improvement Program1 is different from the more established and reactive methods used in the past to identify and prioritize locations for roadway improvements. Using traditional methodologies, public agencies targeted areas for improvement based primarily on past collision history. Locations with high collision rates would be identified, then the features at these locations that could be causing or contributing to cause collisions were analyzed. Once an engineering analysis occurred, changes to the road or road environment would be made based upon a “look back” analysis.
A. Data Analysis
The federally required collection method is cumulative in its approach. Information about road features, locations, and driver behavior patterns are captured and considered to assist the agencies in identifying types of locations and road features that correlate with collision frequency. Those methods are intended to assist agencies in prioritizing and planning improvements on existing roadways before a history of collisions and injuries accrues. The data can be used to inform the roadway design process and multiple other transportation processes.
Predictive analysis2 helps identify roadway sites with the greatest potential for improvement and quantifies the expected safety performance of different project alternatives. Predictive approaches combine crash, roadway inventory, and traffic volume data to provide estimates of an existing or proposed roadway’s expected safety performance. The results can inform roadway safety management and project development decision-making.
Systemic analysis uses crash and roadway data in combination to identify roadway features that correlate with crash types. The term “systemic safety improvement” means an improvement that is widely implemented based on high-risk roadway features that are correlated with crash types, rather than crash frequency.3 A systemic analysis identifies locations that may experience severe crashes, even if there is not a high crash frequency at the time of the review. If those locations are identified, practitioners may be able to use low-cost countermeasures at those locations.4
The abundance of data and the difficulties in accessing and analyzing it create challenges for the state transportation agency. The problem with the predictive data collection process, from the perspective of a risk management analysis, is that the data identifies locations that are “dangerous” in time for the agency to address that condition of their road. Often funds are not available to address those conditions, and the agency is in the position of having identified a condition or location that needs to be improved, without improving it.
A plaintiff’s attorney might argue that the presence of predictive data in the agency’s data collection systems provides constructive or actual notice of potentially “dangerous” conditions of roadways. Following that logic, a plaintiff might
___________________
1 23 U.S.C. § 148.
2 FHWA, Data-Driven Safety Analysis (DDSA), FHWA SAFETY PROGRAMS, https://highways.dot.gov/safety/data-analysis-tools/rsdp/data-driven-safety-analysis-ddsa (last visited Dec. 7, 2024).
3 U.S.C. § 148(14).
4 FHWA, supra note 2.
