Sunday, February 23, 2014

Speed Limits in Cities - ITE Listerv Discussion with Sam Schwartz

Occassionally there's time during the day to review the Institute of Transportation Engineers' Listserv. This is one of the greatest information exchanges that happens in the industry, which is a bit concerning because anyone can participate. That being said, I took an excerpt of a question by Sam Schwartz who is former Commissioner of NYCDOT.

Speed limits are one of the most challenging topics in the transportation industry and is closely linked to the enforcement by police. It's really an important part of any strategy to improve safety in a community. 

Speed limits city-wide
From:Mr. Samuel I. Schwartz, P.E.
To:All Member Forum
Posted:February 12, 2014 5:23 PM
Subject:Speed limits city-wide
NYC politicians and bike/ped advocates have been calling for a law lowering the city-wide speed limit to 20 mph from the current 30 mph.  I have been warning that it may have little effect if all that is done is pass a law and post signs.  I also have warned that it could create other issues including new crashes and more people breaking the law. I have been vilified by many for taking this position. I'd like opinion of peers.

I cite a quote from one of the studies "Expert System for Recommending Speed Limits in Speed Zones" published in November 2006 by researchers at the University of North Carolina Highway Safety Research Center for the NCHRP, TRB, and the NRC.
"Artificially low speed limits can lead to poor compliance as well as large variations in speed within the traffic stream. Increased speed variance creates more conflicts and passing maneuvers, which can lead to more crashes."
Also from "Effects of Raising and Lowering Speed Limits on Selected Roadway Sections" published in January 1997 by the Federal Highway Administration:
"Lowering or raising the posted speed limits at the experimental sites had little effect on driver behavior as reflected by the 85th percentile speeds. Lowering the speed limit by 5, 10, 15, or 20 mi/h (8, 16, 24, or 32 km/h) at the study sites did not result in major reductions such as 5 mi/h (8 km/h) or more in the 85th percentile speeds. "

Samuel Schwartz P.E.
President & CEO
Sam Schwartz Engineering,PLLC
New York NY

RE:Speed limits city-wide
From:Mr. Peter Koonce
To:All Member Forum
Posted:February 13, 2014 5:08 PM
Subject:RE:Speed limits city-wide
This is similar to the change & clearance interval debate. A good number of research efforts have been completed (mostly by graduate students in 12 month programs with good intentions) that show if you increase the yellow time, there is a resulting improvement in reduced crashes. The problem with these studies in my opinion  is that it doesn't necessarily track the human behavior and people learning over time. The Minnesota DOT has the largest study of the effect of clearance intervals I could find when my team was writing the FHWA Signal Timing Manual. My point in this is that there are situations where research is flawed because of the relatively short duration of the study. That wasn't what you asked about, but it is a frame that I use to consider these big picture questions. This is also the sort of thinking that I believe makes policymakers think that we are irrelevant at times (kudos for your article on this, excellent by the way).

In Portland, we have some data (that we're going to write up) that shows that the change of speed limits does have an effect over time. I am not suggesting that we're going to dispute the old adage that just changing the speed limit results in a change of speeds, but our experience  on some corridors is that by changing the speed limit, it begins a process of changing striping requirements and the multitude of other factors, which may change the nature of the street and the physical elements along the corridor over time. These changes don't occur over a one year period, but over a longer horizon (depending on the pace of development, level of activity, enforcement, etc).

So while I agree that a unilateral change in the speed limit may have little effect over the scope of a 18 month research project, our data suggests that there are cases where setting appropriate speed limits may  be cause / reason to support a policy decision to reduce the speed limit that would affect enforcement, engineering, etc in an attempt to meet the goals of the community.

Now, as for 30 mph to 20 mph, I think when you're reducing down to 20 MPH that's quite a low speed that is appropriate on very low volume streets, but not sure how well it would work in an urban context on busy streets. Playing devils advocate now with that, if you have a dense traffic signal spacing pattern then setting progression speeds for 20 MPH is very possible and is something that Portland has had for the past 20 years. Here's an example of that in the form of a video.


Peter Koonce
Principal Engineer
City of Portland
Portland OR

Troubleshooting Traffic Signals with Arrows

A colleague sent me this Boston Globe article on a traffic signal display.

Portland has 15 or so traffic signal locations with Flashing Yellow Arrows in order to display protected permissive left turns. The definitive research on the topic was completed by my colleagues at Kittelson & Associates and is available online here. I had a small part in the project and it was a great learning experience. The experience with implementation was very good initially. Portland's first location was NE 82nd & Tillamook. This location isn't a very busy pedestrian location. 

As their implementation widened, (our neighbors in Washington County - suburbs to the west of Portland) had trouble with left turning movements yielding to the pedestrian. Research that was done on NCHRP  did not consider pedestrians explicitly. The most basic crash that was assumed was the pedestrian approaching from "behind" the left turning motorist in the crosswalk. Our local transportation reporter touched on this issue of problems with the Flashing Yellow Arrow

There was some follow up research work, so the response based on science was underway.

This work by Oregon State University in collaboration with Portland State is a nice study on the topic: and there is also a video interview of the topic

There has been some follow up work that our regional traffic engineering group has done to protect pedestrians at these locations. Depending on the equipment at the intersection, their may be the ability to delay the flashing yellow until after the WALK is complete (like a leading pedestrian interval), or more conservatively, start the Flashing Yellow Arrow only after the Flashing Don't Walk has completed to make sure you can get a pedestrian all the way across the intersection before asking someone in their car to choose a gap. 

The one remaining problem with the specific example at this intersection is the two arrows in the same signal head or face. I would recommend that the left turn is separated from the right turn arrows so the display may be a little more clear to the user. 

Saturday, February 15, 2014

Peter Koonce Bio

Peter Koonce, P.E., manages the City of Portland Bureau of Transportation's Signals, Street Lighting, & ITS Division. He has served as an adjunct professor at Portland State University teaching graduate level courses in transportation engineering. He is a member of the Bicycle Technical Committee of the National Committee on Uniform Traffic Control Devices and was appointed Chair of the Transportation Research Board’s Committee on Traffic Signal Systems. He has served on numerous University Advisory Boards related to transportation engineering and is active as a volunteer for the Community Cycling Center.

Monday, February 10, 2014

Detecting bicycles at traffic signals study

The city is constantly trying to innovate to improve its traffic signals particularly for multi-modal transportation that has been neglected by the engineering profession. I have written about this before, but this time the City has spent a little effort with Portland State to learn more about the subject. 
One idea that we discussed for improving  the
understanding of the marking.

I saw this article in a another newspaper that describes the problem: 
Real answer: “Most traffic signal detectors utilize electro-magnetic induction technology located in the pavement to detect approaching vehicles,” said Anna G. Henderson, division traffic engineer with the Asheville office of the N.C. DOT. “Cyclists whose bicycles are constructed primarily of carbon fiber or other nonferrous materials may experience difficulty being detected by this technology. It is possible that the reader’s bicycle has less than the minimal amount of steel necessary for the signal to detect the bicycle’s presence. To improve the likelihood of receiving a green light, cyclists should stop their bicycle near the center of the travel lane at the white stop line.”
Green tactile marking in front of
bike detector stencil (another option)
This answer wouldn't meet expectations in Portland. How about instead of say near the center of the lane, why not make it more conspicuous by adding green color to the existing bike signal detection stencil applied. The hope is that this touch of green would improve the understanding of cyclists at the intersection and results in improved compliance with the traffic signal and better positioning in the intersection whether it be in a bike lane or in a shared Travel Lane . Feedback on Twitter was mostly positive. I had one traffic engineer that suggested the contrast might reduce the visibility. That's the great thing about Twitter is it gives us some quick feedback on a concept and then we can build some of those thoughts into the study design and test the subject to determine whether they agree with the expert who made that particular comment.
The beauty of this particular study is that it was completed by a student who needed a project to complete his degree. So, the cost to the city in this case is the materials and labor associated with installing the experimental markings and the staff time to investigate and review the work of the research team.  

So, the results? Published on the Portland State website here.

The initial feedback is good, it is the most downloaded undergraduate honors paper in the history of PSU posting them online. Here's the presentation from the PSU Friday Seminar