Can predictive analytics reduce roadway collisions in Indiana?

USDOT building out data tools for traffic safety

The U.S. Transportation Department (DOT) is looking to partner with state, local and tribal governments on developing data analysis tools to improve roadway safety.

In a Notice of Funding Opportunity published Nov. 14, the agency asked government agencies and related organizations such as metropolitan planning organizations to apply for grants to “develop, refine and implement safety tools that address a specific roadway safety problem.” Up to $3 million in funds are available through the program, and DOT expects each grantee to receive $250,000 to $300,000.

“This is part of the department’s ongoing effort to utilize predictive data analytics to identify and address systemic factors contributing to crashes and improve roadway safety in communities across America,” DOT Secretary Elaine Chao said in a statement.

Specifically, grantees will further refine and develop tools that resulted from DOT Safety Data Initiative (SDI) pilot projects aimed at saving lives by advancing roadway safety efforts related to data, analysis and policy-making. Grantees will be able to use other tools, too, that they specify in their application.

The tools, known as SDI Beta Safety Tools, include finalists of the Solving for Safety Visualization Challenge. A finalist in the second phase of the three-phase challenge is Ford’s RoadCode, which would combine crash data with connected vehicles, social media and local population data. Users would be able to use the web-based platform to view traffic flow and patterns, including near-crashes, to identify potential high-risk areas.

The other finalist is the University of Central Florida’s Department of Civil, Environmental and Construction Engineering, which built Real-Time Crash Risk Visualization Tools for Traffic Safety Management. This platform will integrate real-time and static data to pinpoint real-time safety conditions and allow for predictive analytics. It uses artificial intelligence to suggest immediate interventions and long-term solutions to improve safety.

Another tool up for grabs is the Pedestrian Fatality Risk Map, an interactive map that aims to better inform the risk to pedestrians at the neighborhood level across the nation. The model uses data from DOT’s Federal Highway Administration, the National Highway Traffic Safety Administration (NHTSA), the Environmental Protection Agency and the Census Bureau.

Developers created rural and urban models and broke down Census tracts into quintiles based on the ordinal ranking of a tract’s estimated pedestrian fatalities. The top 20% highest-risk neighborhoods are red or pink, while the rest are gray for at-a-glance insights.

NHTSA also is testing interactive ways of presenting data from its Fatality Analysis Reporting System (FARS) – a national census of fatal injuries resulting from vehicle crashes. The agency is using Tableau visualization software to recreate its Traffic Safety Fact Sheets on speeding and pedestrian fatalities.

For the former, NHTSA has a color-coded map showing the percentage of traffic fatalities related to speeding. Users can filter the data by occupant or driver, month and day of the week and time of day. They can also see at a glance where states rank in terms of speed-related deaths. (Washington, D.C., is in the lead with 77%, while Florida is safest at 10%.)

The pedestrian fact sheet is also color-coded to show which states have the most pedestrian fatalities. (California holds the top spot with 5,399.) Users can sort data by gender and age, year and vehicle type.

Another tool under development at DOT’s Volpe Center uses crowdsourcing from private-sector mobile applications such as Waze to provide real-time and historical data on roadway conditions, including accidents. SDI put the tool to work in the Tennessee Integrated Traffic Analysis Network. In collaboration with the Tennessee Highway Patrol, SDI found that integrating Waze and existing crash data could help THP identify high-risk times and areas.

In a similar test in Bellevue, Wash., SDI created dashboards with integrated views of three city crash datasets and locations where police-reported accidents coincide with 911 or Waze crash alerts. “SDI tools will allow the city of Bellevue to identify high priority locations to deploy systems such as video analytics for safety monitoring (in collaboration with Brisk Synergies) to identify specific causal information,” according to a DOT case study.

DOT will hold a webinar about the funding opportunity on Dec. 3 at 2:30 p.m. ET. Applications are due Jan. 17, 2020.

About the Author

Stephanie Kanowitz is a freelance writer based in northern Virginia.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.