IBM is testing smartphone technology that can predict traffic jams and warn commuters before they ever take to the road, something it promises “will ultimately help drivers around the world.”
IBM is one of several companies exploring how to use the computing power of smartphones to make our commutes less hellish. There’s tremendous incentive to do so, because the average American spent 34 hours sitting in traffic in 2009 — and paid $808 for the privilege.
Smartphones are a logical way to help us manage traffic because so many people have them. Mobile Millennium already combines data from a smartphone app and traditional traffic sensors to provide real-time monitoring of traffic conditions. Waze offers a smartphone app that does much the same thing.
But IBM is taking the tech one step further by creating personalized forecasts that predict what users’ commutes might look like even before they leave home.
The heart of the system is a learning and predictive-analytics tool developed by IBM Research. The Traffic Prediction Tool, now being tested in the San Francisco Bay Area, analyzes real-time traffic data and commuter habits to identify any problems that might tie them up.
GPS data from participants’ smartphones identifies their typical commute time and route. That info is combined with data from existing traffic sensors in roads, toll booths, bridges and intersections. The Traffic Prediction Tool creates personalized predictions of when and where someone might run into gridlock, then alerts them by e-mail or text while there’s still time to chose an alternate route.
“The idea is to learn a traveler’s habits, then run it on the predictive model to see what traffic they can expect,” John Day, head of IBM’s Smarter Traveler program, told Agence France-Presse. “The objective was to make it much more personal and provide it to them just before they were about to leave.”
IBM says the predictive nature of its system makes it better than anything else on the road because, despite advances in GPS navigation, real-time traffic alerts and the like, updates on traffic conditions and alternative routes usually reach commuters after they’re stuck in traffic.
“Unlike existing traffic-alert solutions, we’re helping take the guesswork out of commuting,” said Stefan Nusser of IBM Almaden Services Research. “By actively capturing and analyzing the massive amount of data already being collected, we’re blending the automated learning of travel routes with state-of-the-art traffic prediction of those routes, to give travelers timely information that can help them make decisions about the best way to get to their destination.”
Looking further down the road, IBM hopes to incorporate transit information, so commuters could decide whether it might be smarter to hop on the bus or the train.
IBM is developing the technology with the California Department of Transportation and the California Center for Innovative Transportation.
