One Small Miracle of Traffic Light Synchronization

Here’s an illustration of what rejiggering traffic light timing can accomplish to ameliorate traffic congestion on a micro level: The Virginia Department of Transportation has cut in half the time it takes to drive a three-stoplight stretch of State Route 2 in Spotsylvania County, traversed by more than 22,000 cars a day.

Reports the Free Lance-Star:

VDOT has changed the signal timing to cycle more cars through the bottleneck. Now, once a driver reaches the Lansdowne signal, it should take about 2 1/2 minutes to pass Shannon Park Drive, which is adjacent to a Wawa gas station.

That’s down 46 percent from the previous wait of 4.7 minutes.

VDOT engineers added a new sequence of timing options. The three signals have weekday morning, midday and afternoon schedules, a weekend schedule, and an off-peak schedule. Under the old timing schedule, the signals had only morning and afternoon patterns.

So simple. So cost effective. Sadly, there’s no juice in it for the political class, so Virginia doesn’t give the attention to traffic light synchronization that it deserves.

Share this article


(comments below)


(comments below)


4 responses to “One Small Miracle of Traffic Light Synchronization”

  1. Larry Gross Avatar
    Larry Gross

    This is a good first step.

    But what is needed is signals that communicate with each other and adapt to changing traffic conditions.

    I sat in a left lane with a red light and traffic backing up behind me for several minutes when there was no oncoming all.

    and guess what.. when the next slug of cars just released from the upstream light arrived – the light turned red on them and THEN let the left-turning traffic go.

    and of course when I get to the next light.. it turns red – with no side traffic… at all.

    This is a really, really stupid way to deal with traffic and yet this is the way that it is delt with while the “guys” at VDOT say “this is nothing we can do because there is no money for new roads”.

    VDOT also refuses to put in roundabouts in many places that they would work just fine at letting traffic self-regulate rather than have it sit at red lights ..for no reason.

    So.. I give them KUDOs .. AND a thump to the noggin..

    and if there are two different VDOT guys.. Major Kudos to the guys working on traffic management technologies and double thumps on the head to those guys who pine away for new roads or nothing.

    This is really frustrating. I would expect the tax & spend folks to want more money for transportation but the fiscal conservatives that we have seem incapable of motivating VDOT to transform itself into a customer-oriented agency even with JLARC writing out the recipe in long hand.

  2. Anonymous Avatar

    Traffic light timing is like 5% of the cause of congestion. Still, every little bit helps.

    Now tell me why this is any different from your other argument wherein we fix one little bottleneck and that speeds upt the flow to the point that it creates a new bottleneck a little farther down the road?

    Particularly with lights, you fix one and the change propagates throught the system, so you fix a few more, and a few more until you start having negative feedback: The ones you adjusted first have to be readjusted,etc. etc.

    I think this is a much harder problem tha just say ing let’s fix all the lights.


  3. Larry Gross Avatar
    Larry Gross

    you’re confusing the difference between optimzing an existing network as opposed to adding capacity to one part of the network… which basically shunts traffic to the next place where capacity has not been added.

    You would not … purposely “tune” traffic signals to cause bottlenecks.

    That’s exactly what I was talking about.. which was using “dumb” timing of lights that do not talk to the next light.

    Timing lights won’t fix every bottleneck but that’s not the purpose in the first place.

    The purpose is to move traffic as efficiently as you can by utilizing intelligent strategies to optimise your existing infrastructure.

    Timing 3 consecutive lights by laboriously tweakint each light’s controller is, in my view, a DUMB way to trying to accomplish the intended goal because first of all it’s labor-intensive and requires continuous monitoring and tweaking.. and changing if traffic patterns are affect (like with a new light or a new road)..

    What you need is signals that communicate with other signals and that can and do adapt in real-time to changing conditions.

    It’s the same exact technology that is used by computer network routers.. or cell phone towers or any number of other technologies

    and should be.. in my mind.. one of the top priorities of VDOT in our congested ageas.

    For bottlenecks that cannot be easily delt with by technology, then consider on a per site basis if new or redesigned infrastructure can be a solution.

    Building roads as the first response to congestion is DUMB. It’s using a 50 year old approach to problem solving when we need 21st century solutions.

    The most maddening thing about all of this is that this is a natural mission for any agency whose mission is mobility but our agency VDOT chooses to make road-building the priority instead.

  4. Anonymous Avatar

    I was just going by the post. It is a three stoplight stretch. That doewn’t sound like optimizing the system to me.

    Then there is this

    “Now, once a driver reaches the Lansdowne signal, it should take about 2 1/2 minutes to pass Shannon Park Drive, which is adjacent to a Wawa gas station.

    That’s down 46 percent from the previous wait of 4.7 minutes.”

    It sounds like tht means in the same amount of time nearly twice as many drivers could arreve at the WaWa.

    What happens after that?

    I’m agreeing with you that this is dumb, and it is exaclty like fixing one physical bottleneck in the end result.

    “It’s the same exact technology that is used by computer network routers”

    Not quite: routers are designed to allow a certain number of packet “collisions” and if a packet does not arrive, it gets resent. A packet may get split and sent by different routes to be reassembled on arrival.

    It is, as I said, a very complicated problem, more so with independently piloted vehicles than with ephemeral electron sequences that we can make an instantaneous new copy of.

    And, there ar lot of really smart engineers working on the network problem because there is a lot of maoney available………

Leave a Reply