TRAFFIC MANAGEMENT 65 Real time information P robably anyone reading this has been in this situation more times than they can count. e traffic on the motorway has already slowed to a crawl and the VMS sign ahead flashes: CONGESTION NEXT 20km. Brilliant! At this point you are committed to stay in the jam with the most viable alternate routes well behind you and the ones ahead now filling up as drivers bail out at the first opportunity, quickly tying up the secondary roads. You are on a modern, well-managed highway in a major city with a good current generation ITS infrastructure. e traffic management centre is fully aware of the situation. Roadside cameras and loop detectors in the system are all working perfectly generating a stream of images and probe data in real-time as they were designed and (procured at great expense) to do. So why didn’t all this technology prevent the traffic jam that you and about 30,000 others are sitting in right now or that the one that millions around the world experience on a daily basis? e simple answer is that real-time transport management, no matter how much data is available or how quickly and effectively it can be processed, simply isn’t good enough. For decision-makers from city managers to operations centre supervisors to individual drivers to have the means to avoid delays caused by any number of events, they must be able to anticipate, forecast and take action well ahead of what’s happening at that moment. MORE THAN AN EDUCATED GUESS is is exactly the situation that predictive analytics are designed to address – for virtually all transport modes whether highway, bus, ferry, rail, or any combination or sub-category thereof. As it has in other areas of both artificial intelligence and ‘big data’ management technology, IBM and other members of the global ICT David E Pickeral on when real time isn’t good enough Predictive text >>>
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
TRAFFIC MANAGEMENT
65
Real time information
Probably anyone reading this
has been in this situation more
times than they can count.
"e tra#c on the motorway
has already slowed to a
crawl and the VMS sign ahead $ashes:
CONGESTION NEXT 20km.
Brilliant! At this point you are
committed to stay in the jam with the
most viable alternate routes well behind
you and the ones ahead now &lling up as
drivers bail out at the &rst opportunity,
quickly tying up the secondary roads.
You are on a modern, well-managed
highway in a major city with a good
current generation ITS infrastructure. "e
tra#c management centre is fully aware of
the situation. Roadside cameras and loop
detectors in the system are all working
perfectly generating a stream of images
and probe data in real-time as they were
designed and (procured at great expense)
to do. So why didn’t all this technology
prevent the tra#c jam that you and about
30,000 others are sitting in right now
or that the one that millions around the
world experience on a daily basis?
"e simple answer is that real-time
transport management, no matter how
much data is available or how quickly and
e*ectively it can be processed, simply isn’t
good enough.
For decision-makers from city managers
to operations centre supervisors to
individual drivers to have the means to
avoid delays caused by any number of
events, they must be able to anticipate,
forecast and take action well ahead of
what’s happening at that moment.
MORE THAN AN EDUCATED GUESS"is is exactly the situation that
predictive analytics are designed to
address – for virtually all transport
modes whether highway, bus, ferry, rail,
or any combination or sub-category
thereof. As it has in other areas of both
arti&cial intelligence and ‘big data’
management technology, IBM and
other members of the global ICT
David E Pickeral on when real time isn’t good enough
Predictive text
>>>
66
TRAFFIC MANAGEMENT
66
community are developing predictive analytics as the essential
next step in ITS evolution.
Recognising the substantial investment in transport
monitoring technology to date, predictive analytics is about
enhancement not replacement. Enhanced assets such as
analytical tools, trackers, dashboards, ATMS and TIS become
an order of magnitude more e*ective by integrating multiple
sources of data – in far larger volumes than ever before possible
– to look ahead of real-time with a high degree of accuracy.
A SENSE OF HISTORYIBM’s Tra#c Prediction Tool (TPT), which has been piloted
in Singapore, is in the process of being selectively deployed
elsewhere around the world. "rough its synthesis of both
real time and historical data, TPT is able to provide extremely
accurate predictions about how and to what extent speci&c
incidents or events will a*ect all modes within the transport
ecosystem collectively and individually. As a result, all
stakeholders are empowered with accurate information
accompanied by actionable recommendations far earlier than
ever before possible. "is gives transport operators the ability
to mitigate or even eliminate problems in the transport system,
not merely react to them.
"e result is that predictive analytics, when used in concert
with advanced tra#c management, ensures the reliable
and unhindered $ow of goods and people through existing
Tra!c in Singapore which piloted IBM’s Tra!c Prediction Tool
www.barco.com/cool
Liquid-cooled video walls by Barco
Barco’s new generation of LED-lit rear-projection
video wall systems has been optimized with a
unique liquid cooling technology. Combined with
Barco’s Sense6 auto-calibration system, the new
LED video walls offer:
30% less power consumption
80,000 hours lifetime
usability in 24/7 applications
20% more brightness
4:3 and 16:9 compatibility
display sizes from 50” to 80”
Find out more on www.barco.com/cool
LED has neverbeen this cool
67
Real time information
David Pickeral is Global Development Executive for Smarter Transportation at IBM, based in Washington, DC