Crash prone 'modern roundabouts'

[tradephoric]

There are a few roundabouts that have had multiple fatal crashes.  One of them is 96th & Westfield Blvd in Carmel Indiana:

Police say two young men died early Monday morning at a roundabout in Carmel.  It happened around 3:00 am at Westfield Blvd. and 96th Street. Police found a Honda Accord in the center portion of the traffic roundabout. Police say the car crashed into the concrete retaining wall.
http://www.wthr.com/article/two-die-at-carmel-roundabout

A Carmel resident died March 17 after crashing a vehicle into a concrete barrier in the middle of a roundabout.  Police said Lambert Doll was driving a black 2014 Toyota Scion southbound on Westfield Boulevard and failed to negotiate the roundabout at 96th Street. The driver was taken to St.Vincent Hospital in Indianapolis and died a short time later.
http://currentincarmel.com/carmel-driver-dies-after-crashing-into-roundabouts-concrete-barrier

Another roundabout that has experienced multiple fatal crashes is US 62 & Big Tree Road in Hamburg, New York:

Police have identified the man who crashed into the boulders at the new roundabout on South Park and Big Tree Friday morning.  The victim is 54-year-old Kenneth D. Braun of South Park Avenue in Blasdell. Braun was the owner of the vehicle in the accident. He was identified by the Medical Examiners Office through Dental Records.
http://hamburg.wgrz.com/news/news/60422-victim-identified-hamburg-fatal-crash

Hamburg man died early this morning after he was ejected from his motorcycle at the intersection of Big Tree Road and South Park Avenue, according to Hamburg police.  The man, 42, was found in the roundabout shortly after 2 a.m. and taken to Mercy Hospital, where he later died, authorities said.
http://www.buffalonews.com/city-region/hamburg/motorcyclist-killed-at-hamburg-roundabout-20130811

There are many scenarios where someone may blow through the center of a roundabout.  I just posted a video of a driver being chased by the cops and flipping his car in the middle of a roundabout.  You also have drunk drivers, drowsy drivers who fall asleep, inattentive drivers, drivers with medical conditions, drivers who misjudge the road conditions and simply enter the roundabout too fast, and even drivers who experience mechanical failure and are unable to slow down.  You might lose your brakes or be driving one of those pesky Toyota's that have a sticking accelerator pedal.

So a simple question.  Should roundabouts be built with fixed objects in the middle of them?  Had there not been a retaining wall in the middle of that Carmel roundabout we probably wouldn't be talking about it right now.  As it stands, 3 people are dead from striking that retaining wall.

[jakeroot]

So a simple question.  Should roundabouts be built with fixed objects in the middle of them?  Had there not been a retaining wall in the middle of that Carmel roundabout we probably wouldn't be talking about it right now.  As it stands, 3 people are dead from striking that retaining wall.

Fixed objects? Maybe a town-center roundabout, with a statue. But most roundabouts should have heavy shrubbery in the middle, no fixed objects (especially not concrete ones). Maybe raise the land in the middle a little too, so there's like a small hill, where you could elevate some of the bushes and trees (or whatever you want to place in the middle).

[tradephoric]

The FHWA published a report reviewing the fatal and severe injury crashes at roundabouts.  According to the report there have been 46 fatal roundabout crashes from 2005-2013.  Of the 46 fatal crashes, 39 involved vehicle striking a fixed object (85% of all fatal roundabout crashes).  This excerpt is found on page 22 of the report:

Many of the fixed object crashes involved vehicles striking the splitter and central islands of roundabouts, striking curbs and sometimes other fixed objects such as sign posts, light poles, landscaping walls, boulders, and trees. At least 35 of the 39 fixed object crashes involved vehicles striking a curb.
http://safety.fhwa.dot.gov/intersection/innovative/roundabouts/docs/fhwasa15072.pdf

It appears they consider it a fixed object crash when a vehicle strikes a curb but it also mentions drivers hitting other fixed objects such as sign posts, light poles, landscaping walls, boulders, and trees.

[tradephoric]

There have been 134 crashes, 9 injury crashes, and 2 semi's that have tipped over in the first 9 months of a roundabout opening in Jeffersonville, Indiana.  It opened in September 2015 as part of the Ohio River Bridges project and has been a point of contention ever since.  Here's a picture of one of the tankers that overturned inside the roundabout:

http://www.newsandtribune.com/news/saturday-spotlight-new-jeffersonville-roundabouts-see-accidents-in-nine-months/article_649f5dca-5b4a-11e6-a6f8-c318a61be753.html

When it first announced the building of the roundabouts, INDOT boasted national figures that show roundabouts are safer, more efficient and greener.  According to the Insurance Institute for Highway Safety, roundabouts compared to four-way intersections increase traffic capacity by 20 to 50 percent and decrease traffic delays by 20 percent, thus saving fuel use and reducing pollution. And roundabouts are proven to decrease fatalities by 90 percent and all crashes by 35 percent, according to the Federal Highway Administration.

Sixteen years ago the Insurance Institute for Highway Safety analyzed a few dozen single-lane and simplified 2x1 multi-lane roundabouts.  Of course roundabouts being built today are much different than roundabouts built back then.  There was no triple-lane roundabouts analyzed in the IIHS study.  That doesn't stop agencies from citing the out-of-date IIHS study to push any roundabout they are proposing.  Agencies are misleading the public (either knowingly or unknowingly) into thinking some complex 3-lane roundabout will be safer than they actually will be.  If a triple-lane roundabout is being proposed, maybe a study that looks at the safety of triple-lane roundabouts should be cited.  That would be transparent. 

Accident numbers in its first year of operation may not reflect the true safety of the traffic configuration.  In fact, INDOT excludes data from the first three to 12 months of operations in follow-up monitoring because "it should be expected that motorists are adjusting to the new intersection during this time frame."

Of course later on in the article it states this....

INDOT is working to improve the safety of the roundabouts. The agency plans to install larger yield signs and tweak other signs and pavement markings to address the most common accident causes.

So which is it?  They exclude data from the first 3 to 12 months of operations in follow-up monitoring yet there are already plans for INDOT to "tweak"  out the roundabout with larger yield signs and pavement marking changes.  How can INDOT determine tweaks are needed if they don't do any follow-up monitoring for the first three to 12 months of operations?

[english si]

So a simple question.  Should roundabouts be built with fixed objects in the middle of them?  Had there not been a retaining wall in the middle of that Carmel roundabout we probably wouldn't be talking about it right now.  As it stands, 3 people are dead from striking that retaining wall.

A retaining wall is surely structural?

Its also surely quite hard to drive straight across the middle of a roundabout without some other mitigating circumstances - if it wasn't the retaining wall in the roundabout, it gonna be something else.

But yes. Roundabouts, like all streetscapes, should minimise the amount of unyielding solid objects that are possible to crash into at speech, and protect (just as bridge parapets have crash barriers near, something like a wall would need to to) the necessary ones.

[tradephoric]

A retaining wall is surely structural?

Its also surely quite hard to drive straight across the middle of a roundabout without some other mitigating circumstances - if it wasn't the retaining wall in the roundabout, it gonna be something else.

It looks like they added the retaining wall so that decorative plantings could be raised a few feet off the ground.  That appears to be the only purpose of the retaining wall.  There is also shrubbery that somewhat hides the view of the retaining wall.  Here's the view of the retaining wall as drivers approach it traveling SB Westfield Blvd:


I can foresee a scenario where teenagers are joyriding around at 1 in the morning and decide to "turf"  the middle of the roundabout (not realizing there is actually a retaining wall hidden by shrubbery).  The teenagers are just looking for some cheap thrills but instead hit a concrete barrier doing 30 mph.  I don't know if this scenario has happened but I could see it.  Do you want your 18 year old son or daughter to be involved in a fatal crash at this roundabout just because they are acting like stupid teenagers?

[jeffandnicole]

Do you want your 18 year old son or daughter to be involved in a fatal crash at this roundabout just because they are acting like stupid teenagers?

It's about time someone thought about the children.

[cjw2001]

What happened to the concept of personal responsibility?   If you drive like an idiot you live (or die) with the consequences.  These are the same people that would have blown the stop sign and taken out another car when it was a stop controlled intersection -- I'm much happier with them being stopped by an inanimate object rather than another vehicle.

[english si]

It looks like they added the retaining wall so that decorative plantings could be raised a few feet off the ground.  That appears to be the only purpose of the retaining wall.  There is also shrubbery that somewhat hides the view of the retaining wall.  Here's the view of the retaining wall as drivers approach it traveling SB Westfield Blvd:

Well that's someone not thinking things through and designing for stupidness.

You can't blame roundabouts for this though - the problem isn't the roundabout as a road feature, but the really poor thinking about where a garden is by its designer. I would say it's little different to a tree blocking a stop sign, but that's far worse as there's no legitimate reason people should be driving across the roundabout's island.

Do you want your 18 year old son or daughter to be involved in a fatal crash at this roundabout just because they are acting like stupid teenagers?

No, but
1) They aren't acting stupid, they are acting reckless. It's not that they didn't see the roundabout, or didn't know what to do when driving through one, but that they chose to drive over it at 30mph.
2) This recklessness doesn't absolve them from blame, but you seem to put the entire blame on the traffic engineer, who is entirely blameless as they didn't design the garden nor choose to drive across the roundabout, and his choice of junction for the intersection.

[tradephoric]

There have been 3 deaths already at this Carmel roundabout, all caused when a vehicle struck the retaining wall.  How decorative is that retaining wall to begin with?  Would the roundabout really lose that much aesthetics if the retaining wall was torn down? 

If you drive like an idiot you live (or die) with the consequences.  These are the same people that would have blown the stop sign and taken out another car when it was a stop controlled intersection -- I'm much happier with them being stopped by an inanimate object rather than another vehicle.

The teenager who is willing to turf a roundabout for fun isn't going to be blowing every red light that they come across.  They are reckless not suicidal.  Don't get me wrong, someone turfing a roundabout is an idiot, and they may very well pay with their life if they don't recognize the retaining wall in the middle of it. But your assumption is anyone who dies at a roundabout would have somehow killed or seriously injured someone else if they had not died in the roundabout.  You really believe that?

[kalvado]

What happened to the concept of personal responsibility?   If you drive like an idiot you live (or die) with the consequences.  These are the same people that would have blown the stop sign and taken out another car when it was a stop controlled intersection -- I'm much happier with them being stopped by an inanimate object rather than another vehicle.

If we rely on personal responsibility only, there would be no airbags, no seatbelts, no handrails on stairs, no many other things. You're not supposed to crash, not supposed to slip on stairs etc. However those things happen, and someone made an effort to minimize the consequences.
Job of engineer is to accommodate common human mistakes in the design. Or choose another job. Did I mention we have some open position in janitorial service?

[tradephoric]

The double-lane roundabouts analyzed in the Michigan study actually saw a 39% increase in fatal & incapacitating injury crashes (1.50 / 1.08).  Based on some of the recent comments in this thread, we should be embracing the increase in fatal crashes at Michigan's double-lane roundabouts because they are killing off bad drivers.


https://www.michigan.gov/documents/mdot/MDOT_Research_Report_RC1566_379286_7.pdf

[lordsutch]

The double-lane roundabouts analyzed in the Michigan study actually saw a 39% increase in fatal & incapacitating injury crashes (1.50 / 1.08).  Based on some of the recent comments in this thread, we should be embracing the increase in fatal crashes at Michigan's double-lane roundabouts because they are killing off bad drivers.


https://www.michigan.gov/documents/mdot/MDOT_Research_Report_RC1566_379286_7.pdf

Today in cherry-picking: the triple-lane roundabouts in the study actually saw over a 50% decrease in "A-level" crashes (from 1.98 to 0.83, page 56, redacted from your table). So clearly drivers are too stupid to understand two-lane roundabouts but do just fine with three-lane jobs...?

[jakeroot]

Job of engineer is to accommodate common human mistakes in the design. Or choose another job. Did I mention we have some open position in janitorial service?

According to tradephoric's study from several pages ago, 52% of all fatal collisions at roundabouts involved impairment. How exactly do you propose that we curb impaired driving collisions at roundabouts? Roundabouts require a quick reflex, which is something that usually drops when you're impaired (though I know this isn't the case for all drugs). Drunk-driving is less of an issue at signals because driving straight doesn't necessarily require a lot of thought -- you just keep your hands on the wheel, don't twist them, and keep going.

Now, I'm not saying that we have to ignore drunk-driving. But by suggesting that roundabouts are poorly designed because they don't accommodate drunk drivers is ludicrous.

[english si]

There have been 3 deaths already at this Carmel roundabout, all caused when a vehicle struck the retaining wall.

And therefore recklessness and landscapers not accounting for recklessness are to blame, rather than the highway engineer and their junction choice.

How decorative is that retaining wall to begin with?

As you point out, entirely. And therefore out of the remit of highway engineers.

Would the roundabout really lose that much aesthetics if the retaining wall was torn down?

No it wouldn't. Hence why your use of that one fatal collision as part of your argument that multilane roundabouts are death traps is totally and utterly disingenuous.

[kalvado]

Job of engineer is to accommodate common human mistakes in the design. Or choose another job. Did I mention we have some open position in janitorial service?

According to tradephoric's study from several pages ago, 52% of all fatal collisions at roundabouts involved impairment. How exactly do you propose that we curb impaired driving collisions at roundabouts? Roundabouts require a quick reflex, which is something that usually drops when you're impaired (though I know this isn't the case for all drugs). Drunk-driving is less of an issue at signals because driving straight doesn't necessarily require a lot of thought -- you just keep your hands on the wheel, don't twist them, and keep going.

Now, I'm not saying that we have to ignore drunk-driving. But by suggesting that roundabouts are poorly designed because they don't accommodate drunk drivers is ludicrous.

I am not aware of any roundabout which cannot be fixed to accommodate diversity of human skills. Couple of high power bulldozers would do most of the job within couple of hours.
And reflexes are not only impaired by alcohol and related stuff. Besides, not everyone is a NASCAR driver. Something like a flu (did you ever got sick while at work, for example?) or lack of sleep (ever had to work too much overtime?) may do the trick. Of course,  there is an option of letting Darwin do the job...

Oh, and you know - 1/3 of all fatal traffic accidents in US involve alcohol. 50% of drinking age pedestrians hit with cars were actually drunk.
Please present your case for elimination of airbags and crosswalks based on that data. Would save tons of money...

[tradephoric]

Today in cherry-picking: the triple-lane roundabouts in the study actually saw over a 50% decrease in "A-level" crashes (from 1.98 to 0.83, page 56, redacted from your table). So clearly drivers are too stupid to understand two-lane roundabouts but do just fine with three-lane jobs...?

I provided a link to the full study.  Anybody that took the time to read the report could easily point out that the triple-lane roundabouts saw fatal crashes drop by over 50% (which is still well short of the 90% reduction in fatal crashes advertised in the national studies).  The study highlights that the reduction in fatal crashes at the double and triple lane roundabouts in Michigan are lackluster at best.  When you combine the double and triple lane roundabouts analyzed in the study, fatal and incapacitating injuries only dropped by 24%.  That's nowhere near a 90% reduction.

The only roundabout type that had a stellar safety record in the Michigan study were the single-lane roundabouts.  They saw a 50% reduction in total crashes, a 56% reduction in injury crashes, and a 88% reduction in fatal and incapacitating injury crashes. That's the kind of safety result you want to see. 

[lordsutch]

Yes, ideally we'd get the improvements of single-lane roundabouts. However, single-lane roundabouts won't accommodate higher traffic volumes without either signalization of entrances (which reduces the traffic flow and efficiency benefits, although at least signalization retains the angle-of-incidence benefits and probably reduces crashes overall over free-flow) or more lanes (which reduce, but certainly doesn't eliminate, the injury/fatal crash benefits on average).

Personally, I'm ok with 25-50% fewer people burying their loved ones due to crashes at high-volume intersections if that means a few more people end up in fender-benders, if that's what the ultimate improvement of multi-lane roundabouts turns out to be. I suspect it's closer to the single-lane case over the long term with more examples and years of data, but even 25% would be a substantial improvement especially in light of the other quality of life benefits (time, stopped vehicle emissions, etc.).

[tradephoric]

Would the roundabout really lose that much aesthetics if the retaining wall was torn down?

No it wouldn't. Hence why your use of that one fatal collision as part of your argument that multilane roundabouts are death traps is totally and utterly disingenuous.

If I'm trying to prove that multilane roundabouts are "death traps"  wouldn't it be more effective for me to focus on fatal crashes that have occurred at multilane roundabouts?  The roundabout with the retaining wall we have been discussing is the single-lane roundabout at 96th & Westfield Blvd in Carmel.

The theory is that roundabouts with fixed objects in the central island will have more injuries than roundabouts with grassy central islands where there is nothing to hit.  Regardless if the roundabout is single-lane or multi-lane, there will be times when drivers blow through the roundabout's central island.  The question is should these drivers be severely punished for their mistake? 

[kalvado]

Today in cherry-picking: the triple-lane roundabouts in the study actually saw over a 50% decrease in "A-level" crashes (from 1.98 to 0.83, page 56, redacted from your table). So clearly drivers are too stupid to understand two-lane roundabouts but do just fine with three-lane jobs...?

I provided a link to the full study.  Anybody that took the time to read the report could easily point out that the triple-lane roundabouts saw fatal crashes drop by over 50% (which is still well short of the 90% reduction in fatal crashes advertised in the national studies).  The study highlights that the reduction in fatal crashes at the double and triple lane roundabouts in Michigan are lackluster at best.  When you combine the double and triple lane roundabouts analyzed in the study, fatal and incapacitating injuries only dropped by 24%.  That's nowhere near a 90% reduction.

The only roundabout type that had a stellar safety record in the Michigan study were the single-lane roundabouts.  They saw a 50% reduction in total crashes, a 56% reduction in injury crashes, and a 88% reduction in fatal and incapacitating injury crashes. That's the kind of safety result you want to see.

It is impossible to tell for sure without full datasets - but looks like there are some roundabouts that work, and there are some that don't.  Probably depends on traffic patterns.. Bust since roundabouts are the most bestest thing since sliced bread, there is no time to think about it, we need to get more contracts out. 

[tradephoric]

"I'd say in the past ten years, you've seen an increase in modern roundabouts just because research has proven that they are safer, so we have started implementing them in New York State," said Surdej.  "And we have started implementing them successfully, we are seeing a reduction in accident rates in those intersections where there are modern roundabouts."


http://hamburg.wgrz.com/news/news/60422-victim-identified-hamburg-fatal-crash

Put another way... the IIHS assured us that roundabouts are safer so we can build any roundabout we want... hell we're going to put some boulders in the middle of this one.  No harm will be done when drivers strike them... ooh wait, maybe a driver will hit those rocks, causing their vehicle to flip over and catch on fire, burning the driver alive where they can only be identified through dental records.  Yeah, that could happen too i guess.  Maybe not all roundabouts are created equal and maybe it was a really bad idea to put those boulders in the middle of the central island.

[7/8]

Does anyone know what these markings are called? I've circled them in the satellite image below of the Fairway/Kossuth/Fountain roundabout in Cambridge, ON



I feel like the fact that they point in the opposite direction of traffic makes them confusing. They're also inconsistently applied. This roundabout in the same city, at Franklin Blvd and Sheldon Dr has them at both entrances and exits (notice the first image has no markings on exits).



I drove the above roundabout a few weeks ago and for a split second, I saw the markings and wondered if I somehow went the wrong way! Of course I didn't, but the fact that these could make someone think that makes me wonder if they should even be there at all?

[kphoger]

Those are Yield markings. They're at roundabout entrances to indicate yielding to traffic within the roundabout. The fact that your second example has them at exits is due to there being crosswalks near the exits, and the markings are actually to indicate yielding to pedestrians.

[jeffandnicole]

Yep...they're the same shape as a yield sign. Its basically a stop bar, but with diamonds instead. There should be a yield sign posted in line with the markings.

They may be confusing if you haven't seen them before, which happens with many traffic control devices that one doesn't come across very often.

[7/8]

^ Oh okay, that explains my confusion.

But I noticed this nearby roundabout at Fairway/Zellar/Woolner in Kitchener has crosswalks but no yield markings on exits.



They have these signs, which could be argued have the same function. But the Franklin/Sheldon roundabout with the exit markings also has these signs. I guess the Region decided these markings were no longer needed on exits?