IBM working on traffic prediction phone app

IBM is working on software for smartphones that will predict traffic jams and warn motorists to avoid congested routes.

The company revealed that its staff at Silicon Valley and San Francisco are testing the technology, which employs location-tracking systems in smartphones to pinpoint a vehicle and feed live updated information over the Internet from pattern-identifying computers.

The software will be called Traffic Prediction Tool and can give 'reasonably reliable' predictions of how traffic will develop over a journey even 40 minutes before the journey has begun. It can also be personalised to suit the travel habits of an individual.

The data is collected from roadside censors, normally used to report existing traffic, but it is then fed into analytics software to identify trends and come up with predictions on how traffic will change over the course of time.

This marks a distinct difference to current traffic reporting systems, which often report what has already happened. By that time it is too late for many motorists to change their routes, ensuring that they get stuck in a traffic jam and add to the congestion.

The program manager for the project, John Day, told AFP: “We wanted to take advantage of analytic tools to provide predictive capabilities; to get correlations with minor slowdowns and major ones that happen after that.”

The project has been ongoing for the last five months and combines IBM research teams with authorities from the California state highway and a Mobile Millennium Team from Berkeley University.