Sign In

Communications of the ACM

ACM TechNews

Waze Data Can Help Predict Car Crashes, Cut Response Time


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Traffic in Montreal.

City leaders, academic researchers, and the U.S. Department of Transportation (DOT) want to use data from crowdsourced traffic app Waze to predict car crashes and reduce emergency response times.

Credit: Patrick Lauzon/Getty Images

City leaders, academic researchers, and the U.S. Department of Transportation (DOT) want to use data from the crowdsourced traffic app Waze to predict car crashes and reduce emergency response times.

For example, a team of researchers at the University of California, Los Angeles and the University of California, Irvine compared data from Google-owned Waze with crash data from the California Highway Patrol, and found that Waze users notify the app of crashes an average of two minutes, 41 seconds before anyone alerts law enforcement.

Separately, researchers at DOT's Volpe Center analyzed six months of Waze accident report data and found they could build a computer model that closely follows the crashes reported to the police.

The researchers reported the model could "offer an early indicator of crash risk," identifying where crashes might happen before they do.

From Wired
View Full Article

 

Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account