Reston Spring

Reston Spring
Reston Spring

Thursday, August 8, 2013

Traffic Jam Ahead: Level of Service Explained, Colin Mills, Reston Patch, August 7, 2013

The term "Level of Service" gets thrown around during planning discussions. What is it, and how does it affect Reston's future?


Last week’s post about office space per worker and its implications for Reston’s planning received a very favorable response.  I was happy to take a complex subject and make it understandable to the average person.  One of RCA’s mission is to keep citizens informed on issues that matter to our future.

One response I got contained a request for a layman’s explanation of traffic Level of Service.  This bit of jargon gets thrown around a lot in planning discussions, and it seems like an important measure of traffic.  But what exactly is it, and what does it mean in the context of Reston’s planning? 

The concept of Level of Service is simple enough to explain.  The different Levels of Service (LOS) are each assigned to a letter, from A to F.  These operate like old-fashioned school grades: A is the best and F is the worst.  (Unlike with school grades, there is a Level of Service E.) 

These “grades” are applied to intersections.  They measure the average vehicle delay, or the amount of time a car will have to wait before making it through an intersection.  So for intersections controlled by traffic lights, Level of Service A means that the average vehicle waits less than 10 seconds to get through the intersection.  By comparison, LOS C means that the average wait is between 20 and 35 seconds, and LOS F means the average delay is more than 80 seconds.

How do we know what the “average” delay is?  For existing conditions, this is based on traffic counts.  By figuring out how many cars that go through a given intersection over a period of time, it’s possible to calculate the typical delay.  For future conditions, transportation engineers use sophisticated modeling software that starts with current conditions and then projects future traffic counts based on the projected future growth and development. 

Easy enough so far, right?  But there are some complicating factors.  First, remember that Levels of Service are applied to intersections, not roads.  So there is no Level of Service that applies to, say, Reston Parkway as a whole: rather, each of its intersections has its own LOS. 

Also, we’re interested in service during the peak periods, when the most vehicles are on the road.  It’s nice that the intersections are free and clear at 3 AM, but for planning purposes, we care most about the times when people are getting to or from work.  We often call this “rush hour,” but as most commuters know, nowadays it lasts longer than an hour.  The peak period lasts about 3 to 4 hours in the morning and again in the evening; roughly 7 to 10 AM and 4 to 7 PM.

So if Levels of Services are like grades, we should be trying for an A, right?  Not necessarily.  As populations grow and more cars join the roads, LOS A becomes unrealistic.  You simply can’t build in enough road capacity to keep traffic flowing that quickly in urbanized areas.  Most urban areas set the LOS standard at D or E.  The current LOS standard in Fairfax County is a D (average delay of 35 to 55 seconds).  The County is considering moving to LOS E (55 to 80 seconds) for areas around Metro stations.  (Why the drop?  Because they’re trying to encourage walking and biking to the stations.  Pedestrians and bicyclists move more slowly than cars, so giving them time to cross the street means additional intersection delays for vehicles.)

So how will Reston do on the Level of Service scale?  The County’s Department of Transportation did modeling to gauge the impact of the Task Force’s current development plan (Scenario G) on Reston’s traffic.  The results were not good.  The worst intersections, as you would expect, are on either side of the Toll Road.  These are considered Reston’s “gateway” intersections: where Sunset Hills and Sunrise Valley cross Wiehle, Reston Parkway, and the Fairfax County Parkway.

Currently, those intersections are at LOS D or E during the peak periods.  If Scenario G is built out as planned, every one of those intersections except Wiehle/Sunrise Valley would earn an F.  The average delay at those 6 “gateway” intersections would be over 2 minutes each.  The worst of the worst, Wiehle/Sunset Hills, is forecast to have an average delay of over 4 minutes during the evening peak.  An F doesn’t adequately describe that kind of backup; that’s an F-, or even F--.

It gets worse: Those projected delays assume that the County invests heavily in road construction and improvement, bus transit, and pedestrian and bike access.  The forecast roadway improvements alone would cost more than $500 million.  You may have noticed that governments aren’t exactly flush with cash; where’s the money going to come from?
Also, those calculations don’t take into account “spill back.”  That’s what happens when an intersection is so congested that the traffic backs up through previous intersections.  If you’ve tried to go south on Reston Parkway during the evening rush, you’ve experienced this first-hand.  The intersection at Sunrise Valley backs up, which creates a jam for cars trying to get over the Toll Road, which worsens the backup at Sunset Hills, which creates gridlock going up past the Town Center.  And that’s today; imagine how much worse it could get!

This kind of traffic nightmare is what RCA is worried about, which is why we keep banging the drum about the traffic impacts of the Task Force scenarios.  Should traffic be the only factor driving Reston’s future development?  Of course not.  But given the amount of time most of us spend in the car, it’s an important quality of life factor.

We accept a Level of Service goal of E around the station areas; encouraging walking and biking to the stations is important, and aiming higher is probably not realistic.  But E is not F, and it’s certainly not F--.  If you’re going to aim for a standard of E, make it count.  That’s why we argue that the revised Comprehensive Plan should specify that development is only allowed if it allows the 6 gateway intersections I mentioned earlier to maintain LOS E during peak periods.

I hope this post has shed some light on Level of Service for you non-traffic experts out there.  If you have any questions, feel free to leave a comment.  And rest assured that RCA will keep fighting to ensure that Reston’s future growth doesn’t bring our roads to a halt.


Comment:  . . . And it gets worse:  The draft Comprehensive Plan language states, "An overall Level of Service (LOS) ‘E’ goal is expected for the street network in the Reston Transit Station Areas."  "Expected"?  That's not even a goal; somehow it will just happen. What does that mean?

More important, the concept of a network level of service is NEVER defined.  I can not find it defined in official Institute of Transportation Engineers documents either.  And its definition has an incredible impact on traffic congestion in Reston:
  • Does this literally mean every street intersection in the station area (soon to comprise a "grid of streets") or just the important ("gateway" is the term FC DOT uses) ones?  Does include only signalized (busy) intersections? 
  • Does it mean the average of all traffic all-day long seven days per week, including holidays?  workday peak periods? peak hour, or "peak of the peak" (highest 15 minutes)?
  • Does it weight traffic by the intersection measured?  The corner of Reston Parkway & Sunset Hills has a lot more traffic than the intersection of Isaac Newton Square North and West.
And this is just one of innumerable instances of vagueness and opacity throughout the draft Plan we continue to review.  Unless this kind of meaningless language is satisfactorily explained or changed in the draft Plan, it will be nearly impossible for RCA Reston 2020 to recommend to the RCA Board that it endorse the draft Plan. 

1 comment:


    Traffic engineers are really bad at telling what will happen in multimodal scenarios. All they know is road queue theory which is just that, just a theory.

    If you attack the demand side, it has a far better effect than attacking the supply side when it comes to transportation systems. In other words, making it safer for bikes and pedestrians will likely mean more people using bikes and walking, therefore less cars. Sometimes traffic is a healthy thing in that way, I sure as heck don't like crossing a road with 55mph LOS A.


Your comments are welcome and encouraged as long as they are relevant, constructive, and decent.