I had the distinct pleasure in working with two people this year on papers for the upcoming 2014 TRB Annual Meeting where it was wonderful to read their work and support as I could with key thoughts and edits.
Effect of Reducing Maximum Cycle Length on Roadside Air Quality and Travel Times on a Corridor in Portland, OR: Christine M. Kendrick, David Urowsky, Willie Rotich, Peter Koonce, & Linda George
Mitigating the RightTurn Conflict Using Protected Yet Concurrent Phasing for Cycletrack and Pedestrian Crossings
Peter G. Furth, Northeastern University, Peter J. Koonce, City of Portland, Miao Yu, Northeastern University Fei Peng, Northeastern University
Wednesday, July 31, 2013
First TRB Paper - 15 years ago
I am celebrating an anniversary of sorts that happened 15 years ago. In 1998, on this day, I was busy working on the Transportation Research Board Annual Meeting paper deadline (August 1st every year as every graduate student in the U.S. can probably tell you) and produced the following paper:
Koonce, P. J. V, T. Urbanik, and D. Bullock. Evaluation of Diamond Interchange Signal Controller Settings by Using Hardware-in-the-Loop Simulation. In Transportation Research Record 1683. Transportation Research Board. Washington D.C., 1999, pp. 59-66.
It was a very auto oriented research document on traffic signal settings and I really had my heart into the research at the time. I was fascinated with the concept of exploring how traffic signals worked and how I was going to make lives better by reducing traffic congestion at diamond interchanges. There were no pedestrians let alone bicycles in College Station at the SMART Diamond interchange at Highway 6 and Harvey Road.
View Larger Map
Tonight, I am collaborating with researchers related to topics as diverse as air quality assessments of adaptive traffic signal control, thermal imaging detection cameras for bicycle traffic, and effects of traffic signal phasing for bicycle travel. A pretty diverse set of topics still with the common theme of traffic signals.
Koonce, P. J. V, T. Urbanik, and D. Bullock. Evaluation of Diamond Interchange Signal Controller Settings by Using Hardware-in-the-Loop Simulation. In Transportation Research Record 1683. Transportation Research Board. Washington D.C., 1999, pp. 59-66.
It was a very auto oriented research document on traffic signal settings and I really had my heart into the research at the time. I was fascinated with the concept of exploring how traffic signals worked and how I was going to make lives better by reducing traffic congestion at diamond interchanges. There were no pedestrians let alone bicycles in College Station at the SMART Diamond interchange at Highway 6 and Harvey Road.
View Larger Map
Tonight, I am collaborating with researchers related to topics as diverse as air quality assessments of adaptive traffic signal control, thermal imaging detection cameras for bicycle traffic, and effects of traffic signal phasing for bicycle travel. A pretty diverse set of topics still with the common theme of traffic signals.
Friday, July 26, 2013
Vehicular Cycling Debate
I have been pulled into the Vehicular Cycling debate because I dared question the National Committee members on the topic through the NCUTCD listserv. Wading into that conversation is disconcerting.
The first round of dialogue centered on the "fact" that education was the answer. All we need to do is get more people educated about positioning their bicycle and we should have savvy cyclists. As an accomplished cyclist myself, I enjoy having a bike lane when there's an ability to provide one as opposed to sharing a lane with high speed traffic.
The second foray was worse:
I was basically told that my 7 year old should keep her bicycle on the playground and that Portland's safety hasn't gotten better, that we're just masking the data with fancy statistics.
The City Club's recent report "No Turning Back: A City Club Report on Bicycle Transportation in Portland " highlights the steps a group of prominent citizens believe we should take with cycling in the Best Bike City in America. It's interesting that the Vehicular cycling folks are in largely suburban areas where cycling is largely representative of less than a half percent of the population.
Pedaling Revolution has a nice Chapter on Overcoming the Safety Barriers to bicycle transportation that's worth sharing, so I have taken photos of the book from Jeff Mapes' book, which I highly recommend, Buy it here at Powells!
The first round of dialogue centered on the "fact" that education was the answer. All we need to do is get more people educated about positioning their bicycle and we should have savvy cyclists. As an accomplished cyclist myself, I enjoy having a bike lane when there's an ability to provide one as opposed to sharing a lane with high speed traffic.
The second foray was worse:
I was basically told that my 7 year old should keep her bicycle on the playground and that Portland's safety hasn't gotten better, that we're just masking the data with fancy statistics.
The City Club's recent report "No Turning Back: A City Club Report on Bicycle Transportation in Portland " highlights the steps a group of prominent citizens believe we should take with cycling in the Best Bike City in America. It's interesting that the Vehicular cycling folks are in largely suburban areas where cycling is largely representative of less than a half percent of the population.
Pedaling Revolution has a nice Chapter on Overcoming the Safety Barriers to bicycle transportation that's worth sharing, so I have taken photos of the book from Jeff Mapes' book, which I highly recommend, Buy it here at Powells!
 |
| A summary of some of the Vehicular cycling concept arguments.. |
 |
| Part of the argument in Pedaling Revolution for protected lanes... |
Wednesday, July 24, 2013
Are Traffic Signals Bad? The Bikeportland "debate"
A friend wrote in on Bikeportland.org about how traffic signals are bad.
Here's my response:
I respectfully disagree with you on the topic of traffic signals and support the dialogue on speed limits. There are 1,070 traffic signals in Portland and not all are designed for cars to "move faster". A good number of them (250) are in downtown and are used to manage traffic carefully to insure speeds are safe for all users (especially pedestrians crossing right turning traffic and people on bikes sharing the lane or in a bike lane). Our policies in downtown prioritize multimodal movement and safety and have for thirty years. We have been extending that concept to recent projects like the Burnside-Couch couplet which are set up to progress all traffic at 20 miles per hour. Granted, cars can move faster then that, but our intent is to manage the traffic so the traffic signals are a positive influence on the safety of the street. This is not a one sized fits all proposition. The adjacent buildings and activity of pedestrian and bicycle travel play a significant role. What works in downtown doesn't on 82nd Avenue.
Traffic signals aren't a panacea for safety either. In situations with two way traffic, we can't manage the speeds like we do in a grid of streets with nearby signals. There are a number of corridors where the signals are so distant that they do move cars faster than if they weren't there, but any intersection treatment would have the propensity to do that.
I can't stress enough that context is important, but there is also difficulty in changing the status quo with only a change to the traffic signals. There has to be the right context and other supportive efforts to make change happen. Just as the concepts of Missouri aren't likely to show up in Portland any time soon, the ideas of one intersection in England would have to be used with the right context and community support.
Your later comment on lower speed limits and the resulting dialogue should be considered further in a Vision Zero type effort. Reducing speed limits, combined with effective enforcement, including working with the judicial system, would only be effective with a comprehensive focus on the problem. That's what leadership looks like.
If there are traffic signals that you think encourage speeding, ask City staff to take a look at them. If there are streets that seem to have speed limits that are set unreasonably high, call the 503-823-SAFE hotline. The City uses good feedback to make change happen, sometimes incrementally but it all adds up.
UPDATE: Jonathan was kind enough to suggest that this was how to use comments effectively by citing my response. It must have been a slow Bike News day!
Here's my response:
I respectfully disagree with you on the topic of traffic signals and support the dialogue on speed limits. There are 1,070 traffic signals in Portland and not all are designed for cars to "move faster". A good number of them (250) are in downtown and are used to manage traffic carefully to insure speeds are safe for all users (especially pedestrians crossing right turning traffic and people on bikes sharing the lane or in a bike lane). Our policies in downtown prioritize multimodal movement and safety and have for thirty years. We have been extending that concept to recent projects like the Burnside-Couch couplet which are set up to progress all traffic at 20 miles per hour. Granted, cars can move faster then that, but our intent is to manage the traffic so the traffic signals are a positive influence on the safety of the street. This is not a one sized fits all proposition. The adjacent buildings and activity of pedestrian and bicycle travel play a significant role. What works in downtown doesn't on 82nd Avenue.
Traffic signals aren't a panacea for safety either. In situations with two way traffic, we can't manage the speeds like we do in a grid of streets with nearby signals. There are a number of corridors where the signals are so distant that they do move cars faster than if they weren't there, but any intersection treatment would have the propensity to do that.
I can't stress enough that context is important, but there is also difficulty in changing the status quo with only a change to the traffic signals. There has to be the right context and other supportive efforts to make change happen. Just as the concepts of Missouri aren't likely to show up in Portland any time soon, the ideas of one intersection in England would have to be used with the right context and community support.
Your later comment on lower speed limits and the resulting dialogue should be considered further in a Vision Zero type effort. Reducing speed limits, combined with effective enforcement, including working with the judicial system, would only be effective with a comprehensive focus on the problem. That's what leadership looks like.
If there are traffic signals that you think encourage speeding, ask City staff to take a look at them. If there are streets that seem to have speed limits that are set unreasonably high, call the 503-823-SAFE hotline. The City uses good feedback to make change happen, sometimes incrementally but it all adds up.
UPDATE: Jonathan was kind enough to suggest that this was how to use comments effectively by citing my response. It must have been a slow Bike News day!
Sunday, July 21, 2013
Multiple Lane Transition with a Green Bike Box
Boston has done a bit with green paint that I thought were notable. Here's a series of photos of an application that you don't often see in the American context. I didn't have a lot of time to study the intersections.
 |
| Here's a photo of a bike box across multiple lanes |
 |
| the bike lane lead is provided to go from the right hand side to... |
 |
| the left hand side, you can see there's also a transition that can occur from the lane |
 |
| the transition to the left hand side occurs for under the bridge so that you can continue on through the system avoiding a conflict. |
Traffic Signal Systems Committee Meeting in Portland
At the Summer meeting of the Transportation Research Board Committee on Traffic Signal Systems, we started off the three day meeting on a bike ride of inner Portland, giving the transportation engineers the opportunity to experience the streets on a bicycle. The 20 or so people that joined the trip got to see some of the good, bad, and brand new of the City. The visit included a trip on our City's downtown grid (slow speed progression - good), connections to the Waterfront across SW Naito Parkway (signals not really built for bicycle traffic - bad), the Eastbank Esplanade (relatively new), the Hawthorne Bridge counter (new), and the South Waterfront signals on SW Moody (new and improved).
The sit down meetings on Sunday occurred after lunch with multiple tracks of topic including (Simulation & Asset Management), (Multimodal & Technology - think driverless cars), and Signal Timing. The final session is a Research summary that comes up with Problem Statements that will be forwarded to the AASHTO Standing Committee on Research (SCOR) which results in some ideas that end up getting funded that can improve our practice in traffic signal systems.
These sorts of visits to bicycle infrastructure exposes engineers to a new mode that aren't always considered in traditional traffic engineering. There have been some cases where engineers have gotten an opportunity to travel to Europe to see the design used for multimodal solutions that have lead to 30% bicycle mode split.
That's a huge difference than Portland and it is largely because of the gaps that exist based on the various designs that have been used over the years and the lack of emphasis on this mode in the U.S.
More thoughts as the meeting continues throughout the next two days.
The sit down meetings on Sunday occurred after lunch with multiple tracks of topic including (Simulation & Asset Management), (Multimodal & Technology - think driverless cars), and Signal Timing. The final session is a Research summary that comes up with Problem Statements that will be forwarded to the AASHTO Standing Committee on Research (SCOR) which results in some ideas that end up getting funded that can improve our practice in traffic signal systems.
These sorts of visits to bicycle infrastructure exposes engineers to a new mode that aren't always considered in traditional traffic engineering. There have been some cases where engineers have gotten an opportunity to travel to Europe to see the design used for multimodal solutions that have lead to 30% bicycle mode split.
That's a huge difference than Portland and it is largely because of the gaps that exist based on the various designs that have been used over the years and the lack of emphasis on this mode in the U.S.
More thoughts as the meeting continues throughout the next two days.
Saturday, July 20, 2013
Detecting Bicycles at Traffic Signals Sign
The City of Portland has been working with Portland State University students to determine whether the detection of bicycles at traffic signals can be best done with inductive loop detection or some other type. We have studied a variety of different detection types including video and microwave and thermal imaging and there are good applications for these different types depending on the infrastructure at your particular location.The standard detection remains inductive loops and with that sort of equipment, it is important to get people on bikes in the right spot so the traffic signal infrastructure can work as designed. My experience in Boston, MA showed a new way of marking and signing where a person should locate themselves and their bike to get a green indication. I couldn't really determine what the sign meant. It's not intuitive to me and I wasn't sure where I was supposed to place my bicycle.
I am hopeful someone will be able to tell me via the blog comments, without seeing the loop cuts in the pavement (see the last picture), I didn't know what the wording of the sign meant for me to do. 
I would compare this sign to the following video on being detected using the standard sign and detector marking.
Detection of Bicycles at Traffic Signals in Boston
A few more varieties of markings and signage for detecting bicycles at traffic signals
 |
| Left turn lane detection marking (adjacent to the through lane marking) |
 |
| Through movement detection marking |
 |
| To request Green, wait on bicycle stencil |
Thursday, July 11, 2013
Bike Rental "Share" in Portland Maine
I was at the Portland, ME Amtrak railroad station and had a chance to try out the bike rental/bikeshare system called Zagster. Zagster had ten bikes lined up at a rack adjacent to the train station, less than 50 feet from the door. The system required a login and password (on the web) and I had set that up a couple of weeks before. I found the interface The bikes were Breezer bikes with full fenders a rear rack and a front basket which made carrying my luggage from the station easier than I anticipated. The one downside of reserving the bike online through the Zagster website is that they didn't indicate what size the bike was, so I reserved #209 and did not get one that was my size. It worked out, but it's something I shared with customer service as an idea to improve the web interface.
The docking station is not as elaborate as other bike share stations I have seen in Montreal, Washington D.C., etc. but was simple and did the trick. Rather than get a code from a kiosk adjacent to the bikes like these other units, it was delivered via text message on my phone. The text message allowed me to open the lock box that was simply bolted to the rear rack. It reminded me of the lock boxes that realtors use to control access to a spare key for a house they are showing, a simple solution to a problem with controlling access to the bikes. The key to the lock was inside the box and attached to a cable that could be pulled out to operate the lock. I did have trouble with the lock box in that I failed to close the box and that seemed to cause problems when I tried to reclose the bike. It also could have been user error.
The biked worked well for my trips and included a chain guard, so that is great for those that were just jumping on from the train station. One complaint was that the lights on the bike were really limited to the 1-LED Frog bike lights (originally by Knog). At least there were lights on the bikes, unfortunately they weren't worth much for actually seeing the potholes of Portland, ME.
The pricing was more traditional than the newest bike shares that are by the hour with a single day membership. At $20 a day, it was about the same as the local bike shop ($25) with the upside of the Zagster bike being that it was right at the train station and the bike was set up with all the necessary equipment (fenders, chain guard), that I could just get on and ride without any additional gear (I did wear a helmet with another tail light).
I would definitely use it again for day trips, of course the limitation of the system is that is just one dock in the City, but I could see a second dock in downtown Portland being useful for some of the bus transit that was part of the community.
Wednesday, July 3, 2013
Signal Timing Progression Speeds in Downtown
A question on the Institute of Transportation Engineers listserv came from Boise, Idaho related to downtown signal timing progression speeds. Here's my response.
In Portland, OR our downtown progression speeds are 13 to 16 mph depending on time of day. We use a quarter cycle offset system. This works well because the block spacing is limited. I wrote a short blog post on this a few years ago that might be worth a read if you have the time. http://koonceportland.blogspot.com/2010/10/portland-tribune-article-on.html
In Baltimore, we did some work north of the downtown streets in 2007 where we chose a 25 mph speed in the off peak in an attempt to calm the traffic in the neighborhood where there were lower off peak volumes.
Chapter 2 of the Signal Timing Manual describes this rather succinctly. I have excerpted it for your benefit, but we tried to write a summary of how policies like this should be considered carefully with the context of the urban form. Here's some of the text edited for brevity:
Broader land use transportation policies should define the objectives for signal timing plans. User expectations for a street network are often the guiding principles for an agency’s policies. In a downtown setting, the City should want to make sure pedestrian safety needs are prioritized, which is best done by keeping travel speeds consistent. This is best done by having adequate pedestrian crossing times and the lowest cycle lengths possible given the width of streets.
I would not use a 85th percentile speed. I would strongly urge you to use a speed that is reasonable to deliver the safest system possible for pedestrians. Using 20 to 25 mph would result in speeds that result in a sustainable system. A slower speed such that we have used in Portland is also good for people on bicycles. Transit efficiencies are also gained if there are numerous blocks where buses are stopping every other intersection.
There was another answer from a retired engineer from Houston. This was very interesting so I am copying it here.
In Houston in the 1950's a fixed time signal system with variable frequency operation was installed. The CBD is a grid of streets (12 X 15 blocks) with 330 feet c.-c. of the streets. The off-peak signal operation was based upon 40, 60, and 80 second timing which resulted in 18, 20, and 22 mph speeds based on 1/4 cycle offsets. The traffic on streets both east/west and north/south traveled at these speeds. The signal cycles were determined by traffic counts entering and leaving the CBD on a major parkway from the west. In turn, a signal system frequency generator determined the cycle length from the traffic counts. This signal system operated this way up until the early 80's when it was basically utilized as fixed time system operated by time clocks for peak and off-peak operation with 80 second cycles.
One problem that existed in the afternoon peak was the problem of parking garages and lots emptying into the streets and causing a queue of stopped vehicles in all lanes before the next signal turned green, thereby causing the progressive speeds to breakdown. As a result, the signal operation was scheduled for simultaneous operation during the peak hour
We also had a freeway feeder street system of 4 streets by18 blocks (330' c.-c.) long that operated on fixed time of 40, 60 and 80 second cycles with the same 1/4 cycle offset and same speeds as in the CBD. The only problem that existed was the 2-way Main St. that caused all traffic on the 4 streets to stop. The traffic on streets both east/west and north/south traveled at these speeds. This system continues to operate today. Peak hour operation was simultaneous.
In Portland, OR our downtown progression speeds are 13 to 16 mph depending on time of day. We use a quarter cycle offset system. This works well because the block spacing is limited. I wrote a short blog post on this a few years ago that might be worth a read if you have the time. http://koonceportland.blogspot.com/2010/10/portland-tribune-article-on.html
In Baltimore, we did some work north of the downtown streets in 2007 where we chose a 25 mph speed in the off peak in an attempt to calm the traffic in the neighborhood where there were lower off peak volumes.
Chapter 2 of the Signal Timing Manual describes this rather succinctly. I have excerpted it for your benefit, but we tried to write a summary of how policies like this should be considered carefully with the context of the urban form. Here's some of the text edited for brevity:
Broader land use transportation policies should define the objectives for signal timing plans. User expectations for a street network are often the guiding principles for an agency’s policies. In a downtown setting, the City should want to make sure pedestrian safety needs are prioritized, which is best done by keeping travel speeds consistent. This is best done by having adequate pedestrian crossing times and the lowest cycle lengths possible given the width of streets.
I would not use a 85th percentile speed. I would strongly urge you to use a speed that is reasonable to deliver the safest system possible for pedestrians. Using 20 to 25 mph would result in speeds that result in a sustainable system. A slower speed such that we have used in Portland is also good for people on bicycles. Transit efficiencies are also gained if there are numerous blocks where buses are stopping every other intersection.
There was another answer from a retired engineer from Houston. This was very interesting so I am copying it here.
In Houston in the 1950's a fixed time signal system with variable frequency operation was installed. The CBD is a grid of streets (12 X 15 blocks) with 330 feet c.-c. of the streets. The off-peak signal operation was based upon 40, 60, and 80 second timing which resulted in 18, 20, and 22 mph speeds based on 1/4 cycle offsets. The traffic on streets both east/west and north/south traveled at these speeds. The signal cycles were determined by traffic counts entering and leaving the CBD on a major parkway from the west. In turn, a signal system frequency generator determined the cycle length from the traffic counts. This signal system operated this way up until the early 80's when it was basically utilized as fixed time system operated by time clocks for peak and off-peak operation with 80 second cycles.
One problem that existed in the afternoon peak was the problem of parking garages and lots emptying into the streets and causing a queue of stopped vehicles in all lanes before the next signal turned green, thereby causing the progressive speeds to breakdown. As a result, the signal operation was scheduled for simultaneous operation during the peak hour
We also had a freeway feeder street system of 4 streets by18 blocks (330' c.-c.) long that operated on fixed time of 40, 60 and 80 second cycles with the same 1/4 cycle offset and same speeds as in the CBD. The only problem that existed was the 2-way Main St. that caused all traffic on the 4 streets to stop. The traffic on streets both east/west and north/south traveled at these speeds. This system continues to operate today. Peak hour operation was simultaneous.
Monday, July 1, 2013
Bicycle Turn Lane at Brookline & Beacon in Boston
I biked through several circles that are historic in Boston. There were challenges at many of these intersections because of the width of the intersections and the angles that folks were turning. This particular circle at Brookline and Beacon is a great example of a circle that has been recently reconfigured. Not designed for the 8 to 80 set , it does feature a bicycle turn lane that is badly faded from vehicles tracking along the dashed bike lane on either side. It was a busy intersection so there was no way to measure the width, but in riding this sort of configuration, it did provide an opportunity to take some space in order to make the crossing movement. It wasn't comfortable and I wouldn't take my 7-year old, let alone my 10-year on this sort of configuration (okay, maybe the 10-year old since she has done 2 years of Community Cycling Center Bike Camps)The other option would have been to continue through the intersection and then cross at the signal similar to a Copenhagen left. The last one of these installed in Portland at SE Water & SE Clay associated with the Portland to Milwaukie Light Rail project has not been well used, so in this instance, it is unclear whether that's a great concept for this particular situation. It's also clear that it is difficult to communicate to cyclists what we'd like for them to do at an intersection, obviously, signage is something that can be helpful when you're in a busy intersection that you are unfamiliar with.
