Software Pinpoints Traffic Accident ‘Hotspots’
Scientists at Ohio State University have developed software that can identify areas on state roadways where accidents are likely to occur. The system uses statistics concerning injuries, fatalities, and the causes of accidents to generate predictions for all roads, conditions, and times.
“Everyone would love to be able to predict exactly where and when the next crash would be, but there are just too many factors involved, and too much randomness to do that,” says OSU Department of Statistics’ Christopher Holloman. “We can confidently make broad statements, like whether a particular piece of roadway is riskier at a particular time.”
The software serves as a supplement to the highway patrol officer’s expertise, and has mostly confirmed what they already knew. Although it cannot tell why a certain area might be prone to a certain type of accident, the software could help the highway patrol gain insight into such questions.
Holloman and his team have recently incorporated Google Earth technology into the software, which allows the system to color-code roadways so users can zoom in to see the probability of an accident in any area of the state.
The software uses a 900 Mb. database that contains details on every accident on Ohio highways between 2001 and 2005. In turn, the software produces 50 Gb. of output data.
To customize the software for another state would cost about half of what Ohio has spent, and the effectiveness of the system would depend on the quality of the state’s accident data.
Abstracted by the National Law Enforcement and Corrections Technology Center (NLECTC) from Ohio State Research News (03/19/07); Gorder, Pam Frost.