Smartphones to accelerate urban planning
In the Proceedings of the National Academy of Sciences — the world’s most cited general scientific journal — Ford and MIT demonstrated today [Aug. 29, 2016] how a mere six weeks of historical cell phone location data could be nearly instantly analyzed to provide optimal plans for infrastructure development and resource allocation that city planners might take years to sift out.
“The great advantage of our framework is that it learns mobility features from a large number of mobile phone users, without having to ask them directly about their mobility choices. Based on that we create individual models to estimate complete daily trajectories of the vast majority of mobile phone users,” professor Marta Gonzalez at MIT told EE Times in an exclusive interview in advance of the announcement today. “Likely, in time, we will see that this brings the comparative advantage of making urban transportation planning faster and smarter, and even allowing to communicate the recommendations directly to the devise users.”
By giving EE Times advance notice of its breakthrough analytics, Ford and MIT were probably expecting an article crammed with buzzwords like Big Data, Crowdsourcing and Disruptive Technologies. The significance of their feat, however, is more important than stringing together tech-talk to describe it. City planners today get paid six-figure salaries to provide this caliber of accurate commuter surveys (which they usually farm out to consultancies which charge them seven-figure prices). By feeding Ford and MIT’s algorithms the realtime anonymous data already available from cellphone carriers, the years- to even decade-long urban planning cycles are over.
Because most smartphones have not only location data but accelerometer, gyroscope and magnetometer readings, Ford and MIT’s algorithms can also determine whether their users are walking, biking, taking mass transit (and which kind — bus, trolley, train) or driving in a car. They can even determine whether those traveling by car are alone or car-pooling (by merely counting the number of smartphones in the vehicle). Likewise, the software can determine how many riders there are on each particular bus/trolley/train.
In the past, all these data types had to be gathered and correlated from massive surveys, which were notoriously inaccurate since people tend to lie (exaggerate) about their use of mass transit to impress the surveyor.
MIT’s Human Mobility and Networks Lab in its Department of Civil and Environmental Engineering worked with Shounak Athavale, an information technology manager at Ford Motor’s Palo Alto Research and Innovation Center. In cooperation with MIT professor Daniele Veneziano, doctoral candidates Yingxiang Yang and Siddharth Gupta, the group analyzed six weeks of data from the Boston area to prove the concept to their peers who admitted their paper to the Proceedings of the National Academy of Sciences. They also compared their data with that collected the old fashioned way with comparable results, albeit with faster and smarter conclusions from the algorithmic version.
Get all the details in the Aug. 30th issue of the Proceedings of the National Academy of Sciences.
Related articles:
German carmaker trio grabs Nokia’s map service
Aware of pedestrian flow: autonomous car services to predict demand
Toyota partners with taxis to streamline automated driving
Ford Motor and Chinese search engine Baidu invest into Velodyne LiDAR
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
