Safety Analysis & Information

How is safety defined?

The efficient and responsible investment of resources in addressing safety problems is a difficult task. Since crashes occur on all highways in use, it is inappropriate to say of any highway that is safety or unsafe as road safety is a matter of degree. We have refined the assessment of the magnitude of safety problems through the use of Safety Performance Functions (SPF). The SPF reflects the complex relationship between traffic exposure measured in Average Daily Traffic (ADT), and crash count measured in crashes per year. The SPF models provide an estimate of the expected crash frequency and severity for a range of ADT among similar facilities.

Colorado Safety Performance Functions (SPF)

All dataset preparation was performed using CDOT's crash database. Crash history for each facility was prepared using the most recent five to ten years of available crash data. The ADT for each roadway and/or intersection approach (major and minor) over the five to ten years was entered into the same dataset. Each dataset is corrected for the regression to the mean bias using the Empirical Bayes (EB) procedure. This resulted in the SPF models for highway segments, intersections, ramp intersections. CDOT has developed maps to display the safety performance of our roadways, which could be found here.

Normative Baselines

The roadways and intersections were tested for the presence of patterns related to crash type, severity, direction of travel, road conditions, spatial distribution of crashes, time of day, and behavioral attributes. Pattern-recognition analysis is performed using normative percentages for diagnostics of safety problems for comparable roadways and intersections. These diagnostic norms were developed using the same data points as the SPF analysis.