Swapping Smarts for Asphalt

Smart traffic lights are no cure-all for Virginia’s congested road network but VDOT increasingly regards them as part of the solution.by James A. Bacon

Northern Virginia traffic was quieter than usual in the early morning of January 17. The weather forecast was calling for snow later in the day, and the Federal Office of Personnel Management had issued an alert recommending that government workers telecommute or take unscheduled leave. Other commuters were staying home, too, especially workers who worried about getting back to day care in time to pick up their kids.

Despite lighter-than-usual conditions, two traffic accidents tied up traffic along Interstate 66, and the Virginia Department of Transportation had to close off some eastbound lanes. To avoid the congestion, many motorists diverted their routes to Nutley Road and the Lee Highway. Back at the Northern Virginia operations center, VDOT employees used video cameras to keep a close eye on the changing traffic patterns.

Had the alternate routes gotten overloaded, VDOT had the option of altering the traffic-light timing: lengthening the green lights for eastbound traffic. But it wasn’t a decision to be made lightly, for it would have thrown off the careful sequencing designed to accommodate the every-day traffic. In the end, the traffic engineers decided not to to tinker with the lights. Emergency response teams cleared the accidents and driving conditions on I-66 eventually returned to normal.

Xiaoling “Ling” Li

Although it went unused that day, the ability to adjust traffic lights real time in response to traffic accidents or disruptive events like the presidential inauguration is a powerful tool in VDOT’s ongoing effort to keep the traffic moving in Northern Virginia. Of the 1,300 traffic lights under its jurisdiction in the region — everywhere but Arlington County, Alexandria and a handful of smaller cities — VDOT actively manages about 900, says Xiaoling Li, Northern Virginia operations engineering manager.

However, while better coordination of traffic signals can reduce congestion, it is no miracle cure for Northern Virginia gridlock, says Garrett W. Moore, former district administrator for the Northern Virginia district who was recently elevated to statewide chief engineer. In numerous Bacon’s Rebellion commentaries, I had urged Virginia transportation policy makers to consider investing more money in “smart road” technologies like traffic light synchronization as an alternative to expensive concrete, steel and asphalt. Moore invited me to observe Virginia’s state-of-the-art traffic management facility and see what traffic-light synchronization can and cannot do.

The VDOT operations center sits in a large complex of state and Fairfax County buildings just off I-66. From the outside, it’s just another boxy government building. But inside it’s a bee-hive of activity. Entombed within the $125 million facility, a vast, high-ceilinged room houses clusters of cubicles manned by dispatchers with VDOT, the state police, Fairfax fire and rescue, and the Fairfax police. These dispatchers field roughly one million calls yearly.

The response to traffic accidents calls for close collaboration in the field. VDOT dispatches a crew to lay out traffic cones and manage traffic. When there are injuries, Fairfax fire and rescue takes control over the rescue phase. Then state troopers move in to investigate who’s at fault. Finally, when everyone else is finished, VDOT cleans up the mess.

Grouping inter-governmental personnel in the same facility makes it easier to coordinate traffic-response teams and to brainstrom measures to boost productivity, says Hari K. Sripathi, VDOT regional operations director.

One recent initiative reduced the crash-investigation time so roads could be cleared more quickly, says Sripathi. The operations center procured a laser scanner that enabled police to measure precise distances of the vehicles, lanes, topography and other relevant details with a quick electronic sweep — much quicker than measuring everything by hand. As a bonus, the police get a 3-D model of the crash site that is far more detailed than anything they could produce before.

VDOT oversees the traffic lights from a side room. When I visited, two employees were keeping tabs on traffic via a large overhead video screen as well as their own PCs. They had a wealth of data at their fingertips. Besides getting a visual fix on traffic flow, they tracked numbers on major corridors and compared them to historical averages, which can vary by time of day, day of the week, and season of the year. Read more.


Share this article



ADVERTISEMENT

(comments below)



ADVERTISEMENT

(comments below)


Comments

6 responses to “Swapping Smarts for Asphalt”

  1. re: smart traffic signals

    there are two methods currently in use:

    1. – pre-programmed to use historical traffic data but can be changed from control operations

    2.- actual real-time signals that sense the changing traffic conditions and adapt automatically without external intervention.

    Most of what we have is 1. and 2. is more a mathematics problem properly designing software algorithms that actually work correctly.

    Number 1 is basically an extension of VDOTs conventional approach to traffic signals that basically use road or optical sensors to detect traffic at THAT signal but NOT in conjunction with other nearby signals.

    Number 1 uses a computer controller to network the adjacent signals but the program it uses is historical data. Historical data on a network is better than no network but it has some significant disadvantages.

    The data can and does get stale especially in areas where new development is occurring. But the biggest flaw is when traffic conditions change to not normal that do not match the historical data and that’s when the control centers have to think about doing something but keep in mind the control centers are looking at a networked array of signals that they would consider altering manually – with the attendant risk that their changes may well not be coordinated across all the signals – properly – in real time.

    It’s not a really good solution.

    The real solution involves expertise that VDOT has not had “in house” because it institutionally is a road-building organization rather than a traffic management organization and the skills required to build and maintain a true real-time adaptive network of traffic signals is a computer and math problem – not a physical signal /sensor problem.

    The expertise to do the true real-time adaptive signals does exist and is being implemented in some places and I hope that it’s on VDOT’s list of things to do.

    When you combine true real time adaptive signal technology with the dynamic tolls on the adjacent beltways – you may well be able to wring out additional de-facto network capacity across the region and even operate – routinely – in a casualty mode rather than gridlock chaos.

  2. Ain’t gonna happen. The Algo writers are tied up working for Wall Street. They won’t work for peanuts.

  3. “Algo”? ok.. Darrell, educate me… is that the software language that would be used in this application?

  4. In all the kerfuffle about transportation, roads and METRO, several news stories were circulating about the transportation mode requirements for a new location for the FBI headquarters and they start out with (but then “relaxed”) requirements that any new location had to be near METRO.

    that’s a pretty powerful argument for a subway system… right?

    the impression I get from NoVa folks like TMT and DJ is that METRO is indispensable but a sorry mess anyhow…

    ….however.. if push came to shove.. METRO stays no matter what.

    Of course they also believe that RoVa should help pay for it!

  5. Peter Galuszka Avatar
    Peter Galuszka

    Sounds pretty cool actually. Like Richmonders, DC folk freak out at the first sign of snow.
    They ought to try wintering in Moscow…Ha!

  6. Metrorail makes a difference. It takes cars off the road, but also enables additional development that makes traffic worse by putting more cars on the road. It would be interesting to see whether less development without Metrorail or more development with Metrorail provides a better return to the public.

Leave a Reply