Crash prone 'modern roundabouts'

[tradephoric]

Pontiac Trail roundabout most accident-prone in Michigan

An Oakland County intersection is named the most-accident prone in Michigan.  The roundabout at Pontiac Trail and M-5 in Commerce Township had 186 crashes last year.  That's 54 more than any other intersection in the state. Coming in at number two was Telegraph and 12 Mile in Southfield while Eleven Mile and Van Dyke in Warren -- placing third.

http://www.fox2detroit.com/news/local-news/154684840-story

[7/8]

Waterloo Region (in Ontario) has dozens of roundabouts, but there is one that has an especially high collision rate; the roundabout at Homer Watson and Block Line in Kitchener.
http://www.therecord.com/opinion-story/6127192-the-record-s-view-this-roundabout-must-be-fixed-at-homer-watson-boulevard-and-block-line-road-in-ki/

There were 51 collisions at the roundabout in 2011, the year it opened. That number crept upward the next year, when 53 collisions were recorded, then soared in 2013 when there were 72 collisions.

To its credit, the region responded by changing signs, altering lanes and lowering speeds in the vicinity. The result? There were 107 collisions at the roundabout in 2014 – one every three or four days and more than double the number in 2012.

It originally had 3 lanes on one side, but after a bus hit one of my classmates at my high-school, they decided to close the third lane off. They also reduced the speed limit by 20 km/h, but most people ignore that and it hasn't improved the collision rate.

I personally like using roundabouts, but I think it would help a lot if more people would use their signals when exiting.

[tradephoric]

Manufacturers are designing bulky a-pillar in a quest to receive five star safety ratings.  Do bulky A-pillar designs reduce driver visibility and lead to drivers pulling out into the roundabout thinking it's clear when it's not?  The curves of a roundabout can make it easy for a vehicle to get lost in the a-pillar.


http://www.safespeed.org.uk/bike005.pdf

[jeffandnicole]

Manufacturers are designing bulky a-pillar in a quest to receive five star safety ratings.  Do bulky A-pillar designs reduce driver visibility and lead to drivers pulling out into the roundabout thinking it’s clear when it’s not?  The curves of a roundabout can make it easy for a vehicle to get lost in the a-pillar.


http://www.safespeed.org.uk/bike005.pdf

Trebek: Over-exaggerations for $3 please?

And just wondering...why is the picture taken as if the driver is sitting in the middle of the front, rather than in front of the steering wheel?

[tradephoric]

^I agree that "Killer Pillars" is over the top. 

That said, the point is still valid.  Just today i was approaching a roundabout and my a-pillar was blocking the view of a white pickup circulating through the roundabout.  I started to pull out infront of them just as they appeared from my a-pillar's "blind spot".  Luckily i was able to stop in time and no harm was done.  It was the first time i realized that my a-pillar was blocking the view of a circulating vehicle and i instantly took note of the scenario.  I was diligently scanning for circulating vehicles, but that white pickup still came out of nowhere to me.  It's not that i wasn't looking.     

Now the Monday morning quarterbacks will suggest a million different things i should have done to keep myself out of that scenario.  Ultimately, i feel like I almost got in an accident because i can't see through steel.  If I come to a complete stop before entering and turn my head from left-to-right to ensure a vehicle isn't in my a-pillar blind spot, then you get impatient drivers on the verge of rear-ending you because you aren't entering the roundabout fast enough for them.  If you just go assuming nobody is in your a-pillar blind spot, you risk pulling out infront of someone.  Honeslty, i'll piss off people behind me and let them rear end me before I pull out infront of a white-pickup again.

[jakeroot]

I've had large A-pillars block me from seeing pedestrians about to cross, other vehicles when pulling out at a t-junction, literally anything when pulling straight out of a parking spot (after having backed in, which I do quite often). At roundabouts? Sure, but not any more often than at any other type of junction.

[jeffandnicole]

I've had large A-pillars block me from seeing pedestrians about to cross, other vehicles when pulling out at a t-junction, literally anything when pulling straight out of a parking spot (after having backed in, which I do quite often). At roundabouts? Sure, but not any more often than at any other type of junction.

Exactly.  Blind spots are hardly unique to roundabouts. 

[kalvado]

I've had large A-pillars block me from seeing pedestrians about to cross, other vehicles when pulling out at a t-junction, literally anything when pulling straight out of a parking spot (after having backed in, which I do quite often). At roundabouts? Sure, but not any more often than at any other type of junction.

Exactly.  Blind spots are hardly unique to roundabouts.

They are somewhat unique with respect to angles  being unusual. On "normal" intersection, 99% of cases traffic comes from exact left and exact right. There is a certain habit of looking for those directions, known head positions  - I don't know those positions but my body does.
A somewhat similar situation is (also roundabout related) a newly constructed merge to arterial - road was rerouted due to roundabout. I drive that spot every day, and it took me quite a while to get used to new alignment. I am not alone, I just witnessed a close call at the location with bus having to move half a lane to avoid collision.
Problem is that ramp merges to road in a curve, and main road traffic is in unusual spot somewhere between 7 and 8 o'clock - neither over the shoulder, nor in the mirror. Lots of drivers cut off main road traffic without seeing it...

I would say this is same effect.

[tradephoric]

The blue line represents a vehicle path that may be in the blind spot of someone waiting to enter the roundabout.  The angles present at a roundabout can make it pretty easy for a vehicle to get lost in the a-pillar.



The most logical alternative to the roundabout pictured above would be a signaled intersection.  It's true that blind spots are hardly unique to roundabouts.  Here's another revelation ... traffic is controlled by blinky lights at a signalized intersection.  The potential dangers of blind spots at signalized intersections can be alleviated by the use of traffic control devices (ie. no turn on red signs, protected only left turns, conflicting vehicle movements never come on together in the phasing, etc.).  You don't have the same control at a free-flowing roundabout (you just got to live with the crashes). 

[jeffandnicole]

The blue line represents a vehicle path that may be in the blind spot of someone waiting to enter the roundabout.  The angles present at a roundabout can make it pretty easy for a vehicle to get lost in the a-pillar.



The most logical alternative to the roundabout pictured above would be a signaled intersection.  It's true that blind spots are hardly unique to roundabouts.  Here's another revelation ... traffic is controlled by blinky lights at a signalized intersection.  The potential dangers of blind spots at signalized intersections can be alleviated by the use of traffic control devices (ie. no turn on red signs, protected only left turns, conflicting vehicle movements never come on together in the phasing, etc.).  You don't have the same control at a free-flowing roundabout (you just got to live with the crashes). 

You're fucking kidding me, right? You're looking out the drivers door window, inches from ones face. I've been driving in circles and roundabouts all my life, and not once have I ever had such a gastrically huge blind spot.

[Brian556]

I don't think this should be a problem. It is very easy to see, and you and always lean forward if nessessary.

[tradephoric]

Experimental fencing has been installed at the crash prone roundabout of M-5 & Pontiac Trail in an effort to slow drivers down.  According to the article, the average speed of traffic coming off is 30-33 mph.  They want it to be 18-23 mph.  What does everyone think of this idea? 

http://www.wxyz.com/news/region/oakland-county/drivers-have-mixed-feelings-about-new-fences-at-m-5-roundabout 

[kphoger]

Experimental fencing has been installed at the crash prone roundabout of M-5 & Pontiac Trail in an effort to slow drivers down.  According to the article, the average speed of traffic coming off is 30-33 mph.  They want it to be 18-23 mph.  What does everyone think of this idea? 

http://www.wxyz.com/news/region/oakland-county/drivers-have-mixed-feelings-about-new-fences-at-m-5-roundabout 

Not in favor.  Reduce visibility to improve safety?  Not cool.  Driver speed through a roundabout should be controlled by geometry; in this case, I'd say they need to both flare the approaches and reduce the ICD.

[kalvado]

Experimental fencing has been installed at the crash prone roundabout of M-5 & Pontiac Trail in an effort to slow drivers down.  According to the article, the average speed of traffic coming off is 30-33 mph.  They want it to be 18-23 mph.  What does everyone think of this idea? 

http://www.wxyz.com/news/region/oakland-county/drivers-have-mixed-feelings-about-new-fences-at-m-5-roundabout 

Not in favor.  Reduce visibility to improve safety?  Not cool.  Driver speed through a roundabout should be controlled by geometry; in this case, I'd say they need to both flare the approaches and reduce the ICD.

That is a common problem of roundabouts as far as I can tell: when it is being built, it is cheap, it is efficient, it is maintenance-free
And then reality kicks in. Need to add crosswalk signals, rebuild approaches, add lighting, improve pedestrian and bike safety, provide additional training for snow plow drivers - and additional hours for them...   
How often they would need to replace those mesh fences - every two weeks?

[kphoger]

Experimental fencing has been installed at the crash prone roundabout of M-5 & Pontiac Trail in an effort to slow drivers down.  According to the article, the average speed of traffic coming off is 30-33 mph.  They want it to be 18-23 mph.  What does everyone think of this idea? 

http://www.wxyz.com/news/region/oakland-county/drivers-have-mixed-feelings-about-new-fences-at-m-5-roundabout 

Not in favor.  Reduce visibility to improve safety?  Not cool.  Driver speed through a roundabout should be controlled by geometry; in this case, I'd say they need to both flare the approaches and reduce the ICD.

That is a common problem of roundabouts as far as I can tell: when it is being built, it is cheap, it is efficient, it is maintenance-free
And then reality kicks in. Need to add crosswalk signals, rebuild approaches, add lighting, improve pedestrian and bike safety, provide additional training for snow plow drivers - and additional hours for them...   
How often they would need to replace those mesh fences - every two weeks?

I'm curious to know when the roundabout in question was constructed.  The engineers should have known better than to build that large of an ICD and that straight of approaches if it was constructed in the last decade.

[jeffandnicole]

Experimental fencing has been installed at the crash prone roundabout of M-5 & Pontiac Trail in an effort to slow drivers down.  According to the article, the average speed of traffic coming off is 30-33 mph.  They want it to be 18-23 mph.  What does everyone think of this idea? 

http://www.wxyz.com/news/region/oakland-county/drivers-have-mixed-feelings-about-new-fences-at-m-5-roundabout 

Not in favor.  Reduce visibility to improve safety?  Not cool.  Driver speed through a roundabout should be controlled by geometry; in this case, I'd say they need to both flare the approaches and reduce the ICD.

That is a common problem of roundabouts as far as I can tell: when it is being built, it is cheap, it is efficient, it is maintenance-free
And then reality kicks in. Need to add crosswalk signals, rebuild approaches, add lighting, improve pedestrian and bike safety, provide additional training for snow plow drivers - and additional hours for them...   
How often they would need to replace those mesh fences - every two weeks?

That stuff is generally taken care of in the planning process.

I drive a snowplow for a state.  We've never been given any additional training for roundabouts or circles. They are just another piece of asphalt that we have to clear snow from.

Again, this tells me that we're just making up excuses why roundabouts shouldn't be built, rather than using any actual concrete evidence why roundabouts shouldn't be built.  Again, every once in a while traditional intersections undergo modifications as well after they are constructed due to unforeseen events.  And snowplowing...training for a roundabout?  Ha!

[kalvado]

I drive a snowplow for a state.  We've never been given any additional training for roundabouts or circles. They are just another piece of asphalt that we have to clear snow from.

And that is exactly why extra training is needed. Because that one incoming lane flares into two lanes for different roundabout lanes. So how many plow passes are required to clean that approach? Exactly, three passes because approach is 2 lanes + gore wide.
Some heroic plow drivers do two passes, leaving approximately 1.5 driving lane and then designed flow pattern is not completely screwed up.  Not really a problem since that snow usually melts in a month or so, but still.

And yes, this tells me that even DOTs have no understanding of how roundabouts work.. But hey, this is just a second decade of learning curve!

[jakeroot]

I'm curious to know when the roundabout in question was constructed.  The engineers should have known better than to build that large of an ICD and that straight of approaches if it was constructed in the last decade.

The approaches seem to have plenty of flair. You could add chicanes, like they do here in Washington, and all over Australia, but many drivers simply ignore the lane lines, drifting all over the place, in an attempt to keep up a high speed. The best way to slow down drivers, is to have them slow down naturally....

The wide visibility really is the biggest issue here. Roundabouts can have high-speeds, but only once you're in the circle...entering speeds need to be much lower. Not so low that a stop sign is the preferred method, but low enough that T-bone collisions don't suddenly become features of roundabouts.

High entry speeds are generally the result of good visibility. While that sounds like a good thing, driver's need to be giving way at or near the yield line, not 400 feet before the roundabout (in other words, you need to decide whether or not you can enter based on conditions present within 20 to 30 feet of the yield line). Because driver's had such excellent visibility, they were deciding whether or not they could enter well before reaching the circle ... some drivers, apparently, chose poorly.

If you decrease the size of the decision-making zone, entry speeds should decrease, because drivers will need to slow down to confirm that they can enter (the smaller size of this zone will naturally result in lower speeds).

The good news is that, if this is successful (which I'm sure it will be), the county will be installing more permanent decoration, such as bushes).

[tradephoric]

The M-5/Pontiac Trail roundabout saw a big spike in crashes in 2015.  It just so happens the roundabout was part of a major detour route in 2015 during the Haggerty Road closure.  The traffic patterns the roundabout was exposed to during the detour didn't match the traffic patterns the roundabout was designed for.  Sure, roundabouts can work well under specific scenarios, but when the moon and stars don't align you can get into trouble.   

My concern is any report dealing with the effectiveness of the experimental fencing will be biased towards comparing 2015 to 2016 crash data.  That to me won't be a fair comparison.   People will think the fencing is effective simply because there was a year over year drop in crashes.  Again, the traffic patterns at the roundabout during the detour last year doesn't match the traffic patterns occurring during this experimental fencing test.  That's a huge point that shouldn't be ignored.

[jeffandnicole]

I drive a snowplow for a state.  We've never been given any additional training for roundabouts or circles. They are just another piece of asphalt that we have to clear snow from.

And that is exactly why extra training is needed. Because that one incoming lane flares into two lanes for different roundabout lanes. So how many plow passes are required to clean that approach? Exactly, three passes because approach is 2 lanes + gore wide.
Some heroic plow drivers do two passes, leaving approximately 1.5 driving lane and then designed flow pattern is not completely screwed up.  Not really a problem since that snow usually melts in a month or so, but still.

And yes, this tells me that even DOTs have no understanding of how roundabouts work.. But hey, this is just a second decade of learning curve!

What?

You do understand I plow in a state with traffic circles wider than nearly any roundabout, and dealing not only with normal roads entering and exiting, but driveways and parking lots accessing the circle as well.

Circles are just like any other road feature. Put the plow down and push the snow. Spread salt to help melt the snow. Repeat as necessary.

There's an art to it. It takes time to clear a road, and keep it cleared. It's not like there's a wall of 20 inches of snow that traffic can't travel thru.  But with snow falli!g off vehicles, even a plowed road will become slushy again.

A 90 degree right turn at an intersection is worse. Those trucks don't turn on a dime, and snow sticks out far from the curb. Plow operators just keep going at it until it's acceptable.

Please...offer your services to plow a road. No doubt you'll do better than everyone else.

[kalvado]

What?

You do understand I plow in a state with traffic circles wider than nearly any roundabout, and dealing not only with normal roads entering and exiting, but driveways and parking lots accessing the circle as well.

Circles are just like any other road feature. Put the plow down and push the snow. Spread salt to help melt the snow. Repeat as necessary.

There's an art to it. It takes time to clear a road, and keep it cleared. It's not like there's a wall of 20 inches of snow that traffic can't travel thru.  But with snow falli!g off vehicles, even a plowed road will become slushy again.

A 90 degree right turn at an intersection is worse. Those trucks don't turn on a dime, and snow sticks out far from the curb. Plow operators just keep going at it until it's acceptable.

Please...offer your services to plow a road. No doubt you'll do better than everyone else.

Possibly things are much better in NJ; but roundabouts I deal with in NY often had 2/3 lane under  snow a week after snow ended.  At least that was the case during last winter, 2014/2015 (we didn't have any winter in 2015/2016).  I don't remember any problems with right turns except for barricades of snow pushed from  the main road blocking side roads/driveways. Once again, could be different vehicles used in different places.
And well, thing is turning around and doing another pass at roundabout approach seem less than straightforward. You either need to look for a u-turn spot, or back along curved approach. I am not sure if it is done really carefully

[tradephoric]

I'm curious to know when the roundabout in question was constructed.  The engineers should have known better than to build that large of an ICD and that straight of approaches if it was constructed in the last decade.

You seem to be implying that a smaller ICD would have been more effective at reducing the number of crashes.  To your original question, the M-5/Pontiac Trail roundabout was constructed in 2011.  While there was an increase in crashes after the roundabout was constructed, the crash rate wasn't as high as other multi-lane roundabouts cited in this thread.  It wasn't until 2015 when the roundabout had a big spike in crashes (likely due to it being part of a major detour route).  Now you are throwing the roundabout design under the bus because of an abnormally high number of crashes in 2015.  Yes, a roundabout that experiences 100 crashes in a year is a problem, but you are trying to convince me a smaller ICD would have done better.  I'm not buying it. 

[tradephoric]

A 3x2 roundabout at Superior & 14th in Lincoln, NE was constructed in 2012.  Due to the large increase in accidents, one of the circulating lanes was removed in 2013.  It is now a 2x1 roundabout.  In addition, they added fencing at all 4-approaches in 2014.  Now they are talking about spending $888,000 to add more permanent features to replace the "ugly" fencing.  I haven't read any reports indicating how effective the fencing was, but i'm assuming the roundabout saw a big drop in crashes simply because it went from a 3x2 to a 2x1 roundabout.

http://www.1011now.com/content/news/More-updates-likely-coming-to-14th--Superior-roundabout-373306771.html

[tradephoric]

The problem is we don't know if the fencing was successful at the Lincoln, NE roundabout.  They compared the crash rate of the 3x2 roundabout with no fencing to the 2x1 roundabout with fencing.  I could have told you the crash rate would drop if you convert the roundabout to a simplified 2x1 roundabout.  But what effect did the fencing have?  We have no idea because they lumped in the lane configuration changes with the fence change.

When the 14th and Superior roundabout was initially built and opened to traffic the crash rate was higher than anticipated. We had the configuration of the initial roundabout studied and the results showed that we needed to reduce the number of lanes and also reduce the sight distance at the roundabout. One lane in each direction was removed with traffic control devices. The site distance was limited by constructing chain link fence with slats in the medians.

The devices have been in place for approximately two years. The results show that the crash rate has been reduced significantly by the lane reduction and the fencing. At this point we are moving ahead with making the temporary adjustments to the roundabout more permanent. Concept plans have been developed to show putting tack on median concrete in place of the lane reductions. This will allow us to remove the traffic control devices along the outer lanes. A landscaping plan was developed for the center island of the roundabout and the medians. The landscaping and a short decorative fence have been shown to take the place of the chain link fence.

Along with this work a piece of artwork will be loaned to the city from the Duncans to be placed in the center of the roundabout. Construction on the more permanent fixes for the roundabout could begin as early as Fall of 2016.
https://www.lincoln.ne.gov/city/pworks/projects/14th/superior-safety/

[tradephoric]

Venice is proposing changes to the crash prone roundabout at Jacaranda Blvd and Venice Avenue.  Since it's completion in 2008, the roundabout has held the dubious distinction as being the top intersection for wrecks in Sarasota and Manatee Counties.  To combat this, FDOT plans on eliminating a circulating lane inside the roundabout.

https://www.youtube.com/watch?v=M5g7zzBC6Bw

Edit:  The proposed roundabout changes slated to cost $585,000 has been approved by the transportation agency.

http://www.wftv.com/news/florida/roundabout-changes-okdjacaranda-and-venice-avenue-intersection-deemed-high-crash-zonekd/332095543