Crash prone 'modern roundabouts'

[kalvado]

I'm confused.  So you're agreeing with us that they never concluded that any sort of collision would be eliminated entirely.

not entirely, but virtually.
I believe smallpox is the only thing people could entirely eliminate - and even then never say never.

[kphoger]

At this point, I'd say we've well established that roundabouts don't automatically reduce crashes to quite the degree they keep being said to–even injury crashes.  Let's step away from the semantics and agree on that.

Well, most accidents we have are preventable ones. People do make mistakes, and that is the major source of traffic problems.
To put things in perspective - 1/3 of all trafic deaths is caused by alcohol. That is 1 person per hour in US.
You don't care about that problem?

Wow, talk about putting words in someone's mouth.  Did I say I don't care about drunk driving at some point in this discussion?  That was out of the blue.

I'm not entirely sure what direction that argument is going, though.  My recent crash was caused by people (me) making a mistake.  Not drunkenness but inattentiveness.  And my point in even bringing it up is that roundabouts can help mitigate the risk of serious injury and death due to people making mistakes.

Well, let's mirror that: Too bad you survived your accident, we don't need idiots on the road.

Harsh? Sure. So try to be nice guy for once.

Point taken.  And I mean that.

But again, you're not actually mirroring what I said.  What I said was 'Drivers running into things while traveling at stupidly excessive speeds should not be our primary concern here.'  That is not the same thing as saying drivers who have accidents ought not to survive.  My simple point is that, every so often, someone doing something utterly reckless and foolish will have a terrible wreck, and it is not the job of engineers to eliminate that possibility.  That's all.

[kalvado]

At this point, I'd say we've well established that roundabouts don't automatically reduce crashes to quite the degree they keep being said to–even injury crashes.  Let's step away from the semantics and agree on that.

I don't mean "automatic" - but see my logic below.

But again, you're not actually mirroring what I said.  What I said was 'Drivers running into things while traveling at stupidly excessive speeds should not be our primary concern here.'  That is not the same thing as saying drivers who have accidents ought not to survive.  My simple point is that, every so often, someone doing something utterly reckless and foolish will have a terrible wreck, and it is not the job of engineers to eliminate that possibility.  That's all.

Well, I think you're wrong. Job of engineers is to create environment where mistakes are not getting aggregated. If you will, a job of engineer is to put a smoother curve and railings along dangerous spot. This is done to reduce both number of crashes (turn radius, warning signs)  and consequences of those (railings, divided highway). This is done exactly for people who fail to maintain control.. How is that different from treatment of  city intersection?

Or, looking at extreme... Of course, someone going 100 MPH over 35 MPH speed limit is asking for trouble - but putting automatic machine guns to ensure they get what they deserve is a bit excessive!

[Tarkus]

Being as advertised as super-duper safe, I expect roundabouts to be immune to driver mistakes.

That's where most people get it wrong. Roundabouts don't make people immune to mistakes...but they normally reduce the severity of people's mistakes.

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit.  Roundabouts, however, have a ton of stuff to hit--curbing, signs, lighting, landscaping, art--and due to the geometry of the intersection, it's easier to hit it if you're not paying attention/inebriated/trying to be a snazzbucket, because it obstructs the straight line path.  In these particular scenarios, they greatly compound the severity of people's mistakes.  There's trade-offs.

[kphoger]

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit. 

Except other cars, of course. At a conventional signalized intersection, there is substantial probability that the severe/fatal single-car wrecks that have been brought up here would have actually been severe/fatal multi-car wrecks.

[kalvado]

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit. 

Except other cars, of course. At a conventional signalized intersection, there is substantial probability that the severe/fatal single-car wrecks that have been brought up here would have actually been severe/fatal multi-car wrecks.

Interesting question.. Most of single vehicle crashes we discuss occur in low traffic situations.
And in those situations roundabout wouldn't prevent multivehicle event anyway.

[kphoger]

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit. 

Except other cars, of course. At a conventional signalized intersection, there is substantial probability that the severe/fatal single-car wrecks that have been brought up here would have actually been severe/fatal multi-car wrecks.

Interesting question.. Most of single vehicle crashes we discuss occur in low traffic situations.
And in those situations roundabout wouldn't prevent multivehicle event anyway.

A lone car on the crossroad at a stoplight creates that potential. The odds are less than at a busier time of day, but a T-bone is a T-bone, no matter the traffic volume.

At a stoplight, the cars are likely to be traveling at full speed and partially blind to each other (commonly 90°), whereas at a roundabout at least one of them is bound to be traveling slowly (correctly) and there's a 50/50 chance that the cautious vehicle will be pointed more towards the offender (with the offender approaching from the right rather than the left).

[kalvado]

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit. 

Except other cars, of course. At a conventional signalized intersection, there is substantial probability that the severe/fatal single-car wrecks that have been brought up here would have actually been severe/fatal multi-car wrecks.

Interesting question.. Most of single vehicle crashes we discuss occur in low traffic situations.
And in those situations roundabout wouldn't prevent multivehicle event anyway.

A lone car on the crossroad at a stoplight creates that potential. The odds are less than at a busier time of day, but a T-bone is a T-bone, no matter the traffic volume.

At a stoplight, the cars are likely to be traveling at full speed and partially blind to each other (commonly 90°), whereas at a roundabout at least one of them is bound to be traveling slowly (correctly) and there's a 50/50 chance that the cautious vehicle will be pointed more towards the offender (with the offender approaching from the right rather than the left).

there is a difference between potential and certain severe accident...

[Tarkus]

The types of severe/fatal single-car wrecks that have been brought up here are very unlikely to occur in conventional signalized intersections.  There's not much to hit. 

Except other cars, of course. At a conventional signalized intersection, there is substantial probability that the severe/fatal single-car wrecks that have been brought up here would have actually been severe/fatal multi-car wrecks.

In the scenario I was outlining, there are no other cars in the intersection.  If someone goes blazing through a red light at 100mph and there's no one else in the intersection, there's no wreck there. If someone goes blazing into a roundabout at 100mph, and there's no one else in the roundabout, they'll hit curbing, and most likely flip.

All that said, if you take the second scenario, and a car happens to be inside the roundabout on the same side, the benefits of the roundabout go out the window.

[tradephoric]

The Lee Road roundabouts at Whitmore Lake Road and U.S. 23 in Green Oak Township saw 40 accidents that resulted in injuries (between 2011-2016). Last year, 16 injury accidents were called in to 911, including 15 serious enough to require EMS assistance and one instance in which the crash victim refused EMS.

Livingston County's worst intersections for injury crashes
http://www.livingstondaily.com/story/news/local/community/livingston-county/2017/02/10/livingston-countys-worst-intersections-injury-crashes/97607006/

[jakeroot]

The Lee Road roundabouts at Whitmore Lake Road and U.S. 23 in Green Oak Township

I don't think I've seen a double roundabout diagram in the US before. The only one I'm familiar with is in Kelowna, BC. Here's the two side by side...

Michigan on left, British Columbia on right:

[tradephoric]

Traffic crashes continue to rise at Bluffton roundabout
http://www.wjcl.com/article/traffic-crashes-continue-to-rise-at-bluffton-roundabout/8800659
https://www.youtube.com/watch?v=wmVHFILIFfA

[tradephoric]

Less than a year after opening, the city of Dublin is dumping $29,046 at MTJ Engineering to study the 161-Riverside Drive roundabout design.  According to the article, there have been 77 crashes at the roundabout since opening on Aug. 13, 2016.  After six months, 77 crashes.  That's really bad considering the intersection only averaged 12 crashes per year before the roundabout was constructed.

77 crashes in 6 months or 12 crashes in 1 year?  Those safe roundabouts are really doing the trick in Dublin.

http://www.thisweeknews.com/news/20170214/riverside-161-roundabout-design-being-reviewed

[tradephoric]

It looks like they are using HAWK signals to meter traffic at the 136th and Keystone Parkway roundabout in Carmel.  Maybe i'm confused, but they aren't going to use traffic signals?  Here some news articles regarding the changes.

http://wishtv.com/2017/02/17/carmel-installs-new-metering-system-at-keystone-parkway-and-136th-street/
https://archive.org/details/Traffic_metering_of_136th_Street_and_Keystone_Parkway_Roundabout_Interchange 

[kphoger]

It looks like they are using HAWK signals to meter traffic at the 136th and Keystone Parkway roundabout in Carmel.  Maybe i'm confused, but they aren't going to use traffic signals?  Here some news articles regarding the changes.

http://wishtv.com/2017/02/17/carmel-installs-new-metering-system-at-keystone-parkway-and-136th-street/
https://archive.org/details/Traffic_metering_of_136th_Street_and_Keystone_Parkway_Roundabout_Interchange 

I wonder why they didn't choose "normal" two-lamp ramp meters like this one:

[photo courtesy of froggie]

[jeffandnicole]

When NJ metered traffic into a traffic circle, they simply used a traditional traffic light. 

[jakeroot]

I wonder why they didn't choose "normal" two-lamp ramp meters

When NJ metered traffic into a traffic circle, they simply used a traditional traffic light. 

A green light literally feet before a yield isn't the best idea. HAWKs can go dark when not in use, so there's less confusion.

[jeffandnicole]

I wonder why they didn't choose "normal" two-lamp ramp meters

When NJ metered traffic into a traffic circle, they simply used a traditional traffic light. 

A green light literally feet before a yield isn't the best idea. HAWKs can go dark when not in use, so there's less confusion.

It was about a 1/4 mile prior to the circle, with signage stating Traffic Metering Light.  They were designed to reduce traffic going into the circle when the circle started getting clogged up, but wouldn't stop it completely, as businesses and other driveways could be located between the light and the circle.  I glanced quickly at Google images seeing if I could find something, but came up empty.  The last circle with metering lights I'm aware of was the Somers Point Circle near Ocean City, and that circle was removed before 2010.

This was also at a time when roundabouts didn't exist, so traffic circles (and the lights) were pretty unique.  And Jersey traffic circles didn't have yield signs!

[lordsutch]

Signalized roundabouts in the UK are quite common, but they tend to control both the circulating roadway and the entrance, and are used more often in big multilane roundabouts that more resemble the old-style US rotaries than the designs FHWA favors.

As for why they're not using traditional ramp meters, I think it's because they want to cycle the lights less frequently and allow a variable number of cars through per "non-red" cycle. Green would also imply they don't need to yield to circulating traffic when entering, so unless it was significantly upstream of the roundabout it could be more confusing than the HAWK-style approach.

[cjw2001]

The meters will only come on during peak periods when traffic backs up (the rush hours when the nearby high school is discharging or pulling in mass traffic or when the evening rush hour is producing high volumes of traffic on the keystone ramp.   The vast majority of the time the signals will simply be dark. 

[tradephoric]

http://www.gooddaycarmel-bepartofthepositive.com/single-post/2016/11/18/Carmel-celebrates-historic-opening-of-100th-roundabout

As mentioned, Carmel opened its 100th roundabout at Rangeline Road and Carmel Drive on November 17th, 2016.  But to be honest, I have my concerns about this one.  This 2x2 geometry reminds me of Rockford's "wreck-it"  roundabout or Ann Arbor's Ellsworth roundabout.  This quite possibly could be the first roundabout in Carmel to average 100 crashes a year (would be fitting being the 100th roundabout and all!).  According to the Carmel Police Department Annual Reports, this busy intersection averaged 27 crashes a year from 2012-2014.  We will have to wait and see how many crashes occur moving forward.

Here's one crash to tally up at the Rangeline Road and Carmel Drive roundabout.  A Carmel Police officer was injured inside the roundabout a few days ago. 

http://fox59.com/2017/02/21/carmel-police-officer-injured-in-motorcycle-crash/

[cjw2001]

Likely this has more to do with the quality of the at fault driver than the properties of the roundabout:

"Guzman was arrested for operator never licensed, a class C misdemeanor. He was also cited for expired license plates and performing an unsafe lane change."

[tradephoric]

Likely this has more to do with the quality of the at fault driver than the properties of the roundabout:

"Guzman was arrested for operator never licensed, a class C misdemeanor. He was also cited for expired license plates and performing an unsafe lane change."

It has a lot to do with the properties of the roundabout.  The Rangeline Road and Carmel Drive roundabout has a 2x2 geometry.  These geometries are notorious for having high crash rates.  You ignore the fact that the average crash rate of 2x2 roundabouts are roughly 4x higher than signalized intersections and instead focus on trivialities. 

[kphoger]

Likely this has more to do with the quality of the at fault driver than the properties of the roundabout:

"Guzman was arrested for operator never licensed, a class C misdemeanor. He was also cited for expired license plates and performing an unsafe lane change."

It has a lot to do with the properties of the roundabout.  The Rangeline Road and Carmel Drive roundabout has a 2x2 geometry.  These geometries are notorious for having high crash rates.  You ignore the fact that the average crash rate of 2x2 roundabouts are roughly 4x higher than signalized intersections and instead focus on trivialities. 

I saw it that cjw2001 was simply poking a hole in your using the example of an unlicensed driver performing an illegal maneuver in an unregistered vehicle as evidence in support of a claim about safety.

While I agree that the design of multi-lane roundabouts is likely a contributing factor to many of the crashes at them (and this very thread has convinced me that it's likely), the specific example referred to was not particularly valuable to the conversation.  I might call it "clutter."

[kalvado]

Likely this has more to do with the quality of the at fault driver than the properties of the roundabout:

"Guzman was arrested for operator never licensed, a class C misdemeanor. He was also cited for expired license plates and performing an unsafe lane change."

It has a lot to do with the properties of the roundabout.  The Rangeline Road and Carmel Drive roundabout has a 2x2 geometry.  These geometries are notorious for having high crash rates.  You ignore the fact that the average crash rate of 2x2 roundabouts are roughly 4x higher than signalized intersections and instead focus on trivialities. 

I saw it that cjw2001 was simply poking a hole in your using the example of an unlicensed driver performing an illegal maneuver in an unregistered vehicle as evidence in support of a claim about safety.

While I agree that the design of multi-lane roundabouts is likely a contributing factor to many of the crashes at them (and this very thread has convinced me that it's likely), the specific example referred to was not particularly valuable to the conversation.  I might call it "clutter."

License does little to improve driving skills.
There is probably some fraction of unlicensed and/or drunk drivers on each intersection in US, they - and some properly licensed sober drivers - sometimes perform unsafe maneuvers, and probably some vehicles are not fully legal (inspected, registered, insured..).
This all gets enveloped into some numbers by statistics.
Now big question - I asked it before - if we're OK with Darwin approach towards non-topnotch drivers (drunk, sick, poorly trained, eldery, inexperienced, distracted... ), or design has to take into account that some drivers are less than perfect?