Analyzing Driver Behavior to Understand the Factors Contributing to Highway Crashes
Concept to Countermeasure – Research to Deployment Using the SHRP2 Safety Data
Despite the best efforts of highway engineers, law enforcement, and highway safety advocates, traffic crashes still result in tragic loss of life and serious injuries. In 2012, there were more than 33,000 fatalities and 2.2 million injuries in the United States. Driver behavior is a significant factor in more than 90 percent of these crashes. Research to date has only studied driver behavior indirectly by examining crashes and attempting to reconstruct the events that produced them. Detailed and direct observational data on driver behavior are needed so that the highway safety community can better understand how the driver interacts with other vehicles, the roadway, passengers, and distractions.
The SHRP2 Safety Data provides the largest and most comprehensive database ever assembled of what happens in the vehicle before and during crashes and near-crash events. The SHRP2 Safety Data consists of the Naturalistic Driving Study (NDS) database and the Roadway Information Database (RID).
The NDS compiled 2 petabytes of trip data from 3,150 drivers aged 16-80 while they traveled 49.5 million miles under actual driving conditions. The effort compiled an unprecedented amount of detailed data of actual driver behavior under normal conditions, near-crash events, and crash events. The data included detailed video of the driver and the roadway. Also included were data on the vehicles’ speed, acceleration, braking, and other maneuvers. Information such as seatbelt usage and the presence of alcohol is available.
Linked to the NDS trip data are roadway data from the RID, such as the roadway location, curvature, grade, lane widths, and intersection characteristics. The RID also provides environmental data such as time of day and weather. These two databases will support innovative research leading to new insights into crash causation and identification of countermeasures.
FHWA is partnering with the American Association of State Highway and Transportation Officials (AASHTO) and the Transportation Research Board to solicit promising research using the SHRP2 Safety Data. AASHTO’s objectives are to support research leading to promising new countermeasures, to demonstrate the use of the SHRP2 Safety Data, and to familiarize AASHTO members with its potential.
The FHWA/AASHTO Implementation Assistance Program will offer financial and technical support to State departments of transportation and their affiliated researchers in using the safety data to conduct research into new or improved countermeasures, driver education efforts, or enforcement strategies. Beginning with a Proof of Concept implementation assistance phase, the safety data can be used to gain new insights into crash causation and the countermeasures that could prevent crashes. This implementation assistance will provide financial and technical support to help applicants acquire a reduced data set, test a research hypothesis, and identify potential new safety countermeasures. If the Proof of Concept is successful, additional financial and technical support may be provided to conduct more in-depth research and to implement promising new countermeasures. FHWA’s Safety Training and Analysis Center will be available to provide advice, education, and technical assistance on how to use the SHRP2 Safety Data.
In the Field
|Florida||Implementation Assistance Program – The Florida Department of Transportation is studying interactions between drivers and pedestrian features (signs, signals, crosswalks) at signalized intersections to develop more effective engineering, education, and enforcement countermeasures to improve pedestrian safety.||Proof of Concept||
|Iowa||Implementation Assistance Program – The Iowa Department of Transportation’s research will focus on driver speed and distraction, roadway geometry, and how roadway design countermeasures may affect road departures. The results will allow agencies to better understand which roadway safety countermeasures and roadway characteristics are associated with fewer and less severe road departure crashes.||Proof of Concept||
|Michigan||Implementation Assistance Program – The Michigan Department of Transportation’s research will address how a driver’s behavior changes in response to posted speed limits, while controlling for the effects of other important factors, such as roadway geometry, particularly horizontal and vertical alignment. This research hopes to provide critical insights regarding the setting of maximum speed limits, the use of advisory sign location and designs, and important differences among drivers.||Proof of Concept||
|Minnesota||Implementation Assistance Program – The Minnesota Department of Transportation (MNDOT) is exploring the role speed and distraction in work-zone crashes and near crashes. The impacts of distraction and speeding will be quantified so that countermeasures can be developed to command a driver’s attention and provide cues. MNDOT expects to provide recommendations for these applications, which may lead to changes in work zone standards.||Proof of Concept||
|Nevada||Implementation Assistance Program – The Nevada Department of Transportation is studying how driver, vehicle, roadway, and environmental factors affect turning movements, as well as vehicle operators seeing and yielding to pedestrians at intersections. Research findings will be used to guide the development of effective countermeasures to improve pedestrian safety at intersections.||Proof of Concept||
|New York||Implementation Assistance Program – The New York State Department of Transportation is evaluating whether high-visibility marking of crosswalks (HVC) improves pedestrian safety at uncontrolled intersections. They will research the most effective HVC designs and recommend HVC-related improvements.||Proof of Concept||
|North Carolina||Implementation Assistance Program – The North Carolina Department of Transportation is investigating the effects of different road alignments on road departure crashes, and will research how speed, driver-controlled variables, time of day, curves, and grades may also affect near-crashes. This study is expected to lead to more effective countermeasures or a better understanding of where additional countermeasures may be necessary. Countermeasures include different types of advance warnings, in-lane rumble strips, and improved roadway delineation.||Proof of Concept||
|Utah||Implementation Assistance Program – The Utah Department of Transportation is exploring how driver behavior and performance are impacted while in the vicinity of closely spaced interchange ramps (less than 2000 ft). The DOT seeks to identify new roadway design criteria, signing and striping practices, and other safety countermeasures.||Proof of Concept||
|Washington||Implementation Assistance Program – The Washington State Department of Transportation (WSDOT) is analyzing episodic speeding (e.g., 10 mph or more over the posted speed) in and around Seattle. Examining speed behaviors will help WSDOT to determine whether specific aspects such as infrastructure, bicycle and pedestrian access, roadside parking, number of lanes, visual cues to motion, and shoulder width result in excessive speeding.||Proof of Concept||
|Washington||Implementation Assistance Program – The Washington State Department of Transportation is evaluating how roadway lighting characteristics affect safety performance and driver behavior for different types of roadways, and the type of lighting needed to improve safety performance and driver comfort in various roadway settings. The DOT will also investigate cost-sensitive lighting design standards and policies that can improve safety and lower energy consumption.||Proof of Concept||
|Wyoming||Implementation Assistance Program – The Wyoming Department of Transportation (WYDOT) is researching how drivers respond to adverse weather and road conditions (e.g., speed adaptation, lane maintenance, vehicle headways). WYDOT will gain insight into what cues are the most effective in providing drivers with a more realistic variable speed limit system and how drivers adjust their behaviors to compensate for increased risk due to reductions in visibility.||Proof of Concept||
Field activities also performed in Indiana and Pennsylvania during the research phase.