IBM Systems Magazine, Mainframe - March/April 2017 - 22
PHOTO: S-F / SHUTTERSTOCK.COM
IBM Research and partners help traffic run smoothly in Madrid
early every city plan takes traffic into account to help commuters. But
people still face congestion on the road, whether they drive or take public
transportation. IBM researchers, such as Paula Ta-Shma, research staff
member, IBM Research-Haifa, are working to rectify that, using machine learning
and a host of open-source tools.
Jim Utsler is a
senior writer for
and has been
Ta-Shma is collaborating with
COSMOS to use both historical
and real-time data to improve
traffic monitoring to optimize the
use of roadways and vehicles in
the city of Madrid.
COSMOS is a European Unionfunded research project that
comprises use-case partners
(including the Madrid Council
and the EMT Madrid bus company) and technology partners
(including IBM, Atos and the
University of Surrey), and aims
to have sensors in the Internet
of Things (IoT) interact with one
another socially-the way people
do on social networks.
According to Ta-Shma, that
research could have implications
well beyond transportation.
IBM Systems Magazine (ISM):
How are you using machine
learning in the Madrid traffic
Paula Ta-Shma (PTS): What
we're doing is collecting traffic data
[regarding speed and intensity].
In the case of Madrid traffic, it's
open data published by the Madrid
Council. Around 3,000 sensors
record traffic in various fixed locations throughout Madrid, and we
continuously collect this data and
store it long term. At the same time,
we have continuous access to the
real-time feed of data. What's really
important is making use of both
22 // MARCH/APRIL 2017 ibmsystemsmag.com
pg 22-23.indd 2
2/7/17 1:27 PM
If you would like to try to load the digital publication without using Flash Player detection, please click here.