Discuss as:

Smartphones to ease traffic snarls

IBM

IBM's new Smarter Traveler initiative enlists your smartphone to collect data that could ease your commute by predicting traffic snarls and suggesting alternative routes.

All commuters have a personal bag of tricks to skirt traffic. Now, a new smartphone app under development promises to learn your tricks and let you know when to use them.

The opt-in system combines information on your typical driving patterns collected by your smartphone with mountains of historical traffic data collected by sensors at toll booths, in roads, bridges, and intersections to predict traffic snarls and ways to avoid them before you leave home.


This is a step up from the traffic report on local news radio or a real-time traffic map on the Internet. It's a prediction of what a driver's personal commute is likely to look like in 30 to 45 minutes, John Day, program manager for the IBM Smarter Traveler initiative, explained to me today.

"The idea is to get that information delivered to you before you leave," he said. "And you don't have to take an overt action to get that info, especially given how busy we all are."

California collaboration
IBM is developing the smartphone application in collaboration with the California Department of Transportation (CalTrans) and the California Center for Innovative Transportation at the University of California at Berkeley for deployment in the San Francisco Bay Area. A global rollout is on the horizon.

"The traffic problems can be very different in other parts of the world and we built it with that in mind," Day said. "It is a matter of partnering with the various agencies who are responsible" for managing transportation.

Key to the collaboration is IBM's Traffic Prediction Tool, which continuously analyzes congestion data, commuter locations and expected travel start times throughout a metropolitan region.

IBM researchers are teaming with California Department of Transportation and UC Berkeley to look at the problem of traffic as a data problem.

"It has a very high rate of accuracy in terms of predicting these issues once you have a good historical database," Day said. "As it happens, UC and CalTrans have done a real nice job of not only building this sensor network, but keeping the data for a long period of time."

The prediction tool is automatically updated every five minutes and produces a new model of what traffic will look like in 30 to 45 minutes. The new initiative blends in your own travel habits for a level of personalization.

People who opt-in to the program will be able to logon to a website where they can review traffic data, as well as specify things such as favorite alternatives and when they want alerts sent.

Future versions
Plans for the future include incorporation of real-time data on public transportation networks such as whether buses and trains are running on time and availability of parking at stations.

"When there is a major traffic issue, we can offer an alternative, maybe drive halfway, jump off (the freeway) and get BART (Bay Area Rapid Transit)," Day said.

This could help persuade people to get out of their cars and onto public transportation, a step that may be necessary as cities continue to grow and roadways become more clogged.

Currently, commuters across the U.S. spend an average of nearly a week's worth of time, 28 gallons of gas, and $808 a year stuck in traffic congestion, according to a 2009 study from Texas A&M University

More about traffic tech:


John Roach is a contributing writer for msnbc.com. Connect with the Cosmic Log community by hitting the "like" button on the Cosmic Log Facebook page or following msnbc.com's science editor, Alan Boyle, on Twitter (@b0yle).