The assessment of driver compliance at highway-railroad grade crossings based on naturalistic driving study data

Document Type

Conference Proceeding

Publication Date



Department of Civil, Environmental, and Geospatial Engineering; Center for Human-Centered Computing


The last several decades have seen a significant effort toward reduction in the number of accidents at highway-rail grade crossings (HRGCs), with remarkable results. However, HRGCs remain one of the key locations for loss of life within the rail system, so safety improvements at HRGCs remain a high priority. Literature indicates driver behavior is a leading cause of accidents. This suggests that further reductions in accident rates might be achieved by better understanding the risks drivers take at HRGCs. Some of these key risk factors to accidents are highway traffic volume/speed limit, type of traffic control devices (TCDs), trains per day, etc. This study used over 5,000 individual crossing traversals from the SHRP2 Naturalistic Driving Study data to evaluate compliance of drivers at HRGCs and to perform comparative analysis on the risk factors mentioned above. To quantify the quality of drivers’ behavior during traversals, the study developed a “compliance score” that was based on the visual scanning and the speed adjustment. The early result indicate that there is a significant difference in behavior at HRGCs with different TCDs. Lower compliance scores are found at HRGCs with active warning devices, and higher scores at passive HRGCs. Preliminary results for numerical parameters suggest potential trending, but more data must be analyzed to obtain statistical significance. Overall, this study provides a starting point toward better understanding of driver behavior at HRGCs. In the future, the results could also be useful in the design and evaluation of new approaches to improve safety at HRGCs.

Publication Title

Transportation Research Board 97th Annual Meeting