Crash prone 'modern roundabouts'

[kphoger]

People who blow through the middle of a roundabout..... How do these people manage to make it to Wal-Mart and back?  I just don't get it.

[jakeroot]

Video of an RV blowing through the middle of a roundabout in Worthington Minnesota.  At least their wasn't a retaining wall in the central island for the driver to hit.  Nobody came to a "dead" stop, like the fate of several people at the 96th and Westfield roundabout in Carmel.  Three people have died at that Carmel roundabout already, that can be considered several right?  Luckily there wasn't a family in an RV to add to the 96th and Westfield roundabout death list.

https://www.youtube.com/watch?v=u4oltdv0i9w

I think that video demonstrates exactly why those retaining walls don't make any sense. The only reason that crash wasn't any worse, was because there weren't any retaining walls in the center. Retaining walls inside the center of roundabouts are about as smart as putting concrete barriers in all of the negative space inside of an intersection. They don't allow for error.

[tradephoric]

^Apart from the RV blowing through the middle of that Worthington Minnesota roundabout, there has been two other pretty major events at the roundabout over the past couple weeks.

Tanker hauling cream overturns in Worthington roundabout

http://www.dglobe.com/news/accidents/4356792-tanker-hauling-cream-overturns-worthington-roundabout


These little piggies almost didn't make it to market when a Minn. hog truck rolled

http://www.agweek.com/news/4350581-these-little-piggies-almost-didnt-make-it-market-when-minn-hog-truck-rolled

Cream trucks rolling over, hogs running around the highway, some RV driver thinking they are Bo Duke... these roundabouts really give the good people of Worthington Minnesota something to talk about!

[kphoger]

Worthington Minnesota roundabout

It was only due to my attentiveness that I didn't get side-swiped at two of Worthington's roundabouts last month.  In fact, the experience made me think of this thread at the time.  But, since they are 2x1 rather than full 2x2, I didn't bring it up.

The main road (MN-60) is striped as two lanes in each direction through the roundabouts.  The driver next to me didn't seem to be the most capable, and seemed like the kind who drives as if she's the only one on the road.  So I decided to lag behind just a little bit as we drove through–which was fortunate, because she took both lanes and would have run into me if I hadn't lagged behind.  At the next roundabout, it was the same thing.  Good thing for me, she was getting off the highway at that second roundabout–turning left, I might add, from the right entry lane.  I can understand trucks and buses taking two lanes through a roundabout, but pure disregard for other drivers is just really irritating to me.

[jakeroot]

I seem to recall that, when the MUTCD was starting to draw up some "modern roundabout" standards, that all roundabouts would have truck aprons, instead of wider-than-normal circulating lanes, because drivers have a tendency to take the fastest line. Since most drivers aren't too keen on driving on curbs, the apron came into play to ensure that drivers, assuming they were paying attention, would not attempt to take the fast line through the roundabout, but still allow trucks room to navigate through the circle.

Unfortunately, when a roundabout has two lanes, it's much easier for an inattentive driver to simply ignore all the markings, and take the fastest line anyway. I can understand doing this when no one else is around. I drive over lane lines when I'm "rushing" all the time. But people do seem to forget that they aren't the only ones driving, so you get some close calls like ^above^.

[tradephoric]

Wouldn't it make more sense to have depressed truck aprons as opposed to raised truck aprons at roundabouts?  Raising one side of the trailer up 5 or 6 inches as they ride the apron curbing in the middle of a sweeping turn sounds like it's a recipe for a tip-over.  If the apron was depressed, it would  be counteracting the centrifugal force of the trailer wanting to tip.  I'm sure someone who has researched truck aprons will tell me why a depressed apron would be a horrible idea, but i do know there are a lot of trucks tipping over at roundabouts (that Worthington roundabout has seen 2 in the past 2 weeks alone).  Just look at the picture below and you see the trailer leaning to one side as it climbs the curbing.

[kphoger]

Wouldn't it make more sense to have depressed truck aprons as opposed to raised truck aprons at roundabouts?  Raising one side of the trailer up 5 or 6 inches as they ride the apron curbing in the middle of a sweeping turn sounds like it's a recipe for a tip-over.  If the apron was depressed, it would  be counteracting the centrifugal force of the trailer wanting to tip.  I'm sure someone who has researched truck aprons will tell me why a depressed apron would be a horrible idea, but i do know there are a lot of trucks tipping over at roundabouts (that Worthington roundabout has seen 2 in the past 2 weeks alone).  Just look at the picture below and you see the trailer leaning to one side as it climbs the curbing.

[img

Drainage, for one thing.

[jakeroot]

Pretty sure I recognised that roundabout...near Vernon, BC (if anyone's curious): https://goo.gl/zAuqC9

The city of Vernon re-worked a few of their right-turn slip lanes at a few of their intersections to include truck aprons, and are raised in much the same as those you see at roundabouts: https://goo.gl/rvPMMy (note the red stamped concrete). Not aware of any rollovers at these intersections.

I'm sure someone who has researched truck aprons will tell me why a depressed apron would be a horrible idea

Drainage, for one thing.

I don't think that would present too much of an issue, as long the grade was kept between like 1-3%. You'd need to have a part of the apron be slightly deeper than the rest, and the water would drain to that point. Just hookup a sewer, and you're golden.

[kphoger]

Pretty sure I recognised that roundabout...near Vernon, BC (if anyone's curious): https://goo.gl/zAuqC9

The city of Vernon re-worked a few of their right-turn slip lanes at a few of their intersections to include truck aprons, and are raised in much the same as those you see at roundabouts: https://goo.gl/rvPMMy (note the red stamped concrete). Not aware of any rollovers at these intersections.

Aren't roundabouts already constructed such that water drains away from the center island?  Assuming that's correct, then the truck apron at a roundabout exists on a slope–which also happens to slope in the same direction of centrifugal force.  Alongside a typical intersection, any such slope would be counteracting the centrifugal force of a turn.  FWIW, I think the apron implementation you linked to is a great idea.

[tradephoric]

There was a presentation about geometric parameters that affect truck stability at the international roundabout conference this year in Green Bay.  They simulated the Rollover Index (RI) which is the relative difference in loading on left versus right side of truck at the rear axles.  Here are some of the key findings (most of these seem to be common sense):

-RI decreased with increase in ICD.
-RI lower in right lane than left.
-RI decreases most significantly when lowering truck speed from 15 mph to 10 mph.
-RI lower with crowned section than with constant outward slope.
-Higher rollover risk when encountering a modeled 3-inch vertical face truck apron. 

They also compared different truck types and loading to determine which were more susceptible to rollovers.  The key findings of truck type and loads are the following:

-WB-67 trucks less stable than SU-30 and B-train at smaller, single-lane roundabouts
-Empty trucks (WB-67 and B-train) at higher risk of rollover in small roundabouts.
-Fully loaded trucks higher risk in two-lane roundabouts due to higher center of gravity
-Non-static loads (unsecured or liquid loads) not tested.


Review of Fatal and Severe Injury Crashes and Evaluation of Geometric Parameters that Affect Truck Stability
http://teachamerica.com/RAB17/RAB172A_Steyn/index.htm

[kalvado]

There was a presentation about geometric parameters that affect truck stability at the international roundabout conference this year in Green Bay.  They simulated the Rollover Index (RI) which is the relative difference in loading on left versus right side of truck at the rear axles.  Here are some of the key findings (most of these seem to be common sense):

-RI decreased with increase in ICD.
-RI lower in right lane than left.
-RI decreases most significantly when lowering truck speed from 15 mph to 10 mph.
-RI lower with crowned section than with constant outward slope.
-Higher rollover risk when encountering a modeled 3-inch vertical face truck apron. 

They also compared different truck types and loading to determine which were more susceptible to rollovers.  The key findings of truck type and loads are the following:

-WB-67 trucks less stable than SU-30 and B-train at smaller, single-lane roundabouts
-Empty trucks (WB-67 and B-train) at higher risk of rollover in small roundabouts.
-Fully loaded trucks higher risk in two-lane roundabouts due to higher center of gravity
-Non-static loads (unsecured or liquid loads) not tested.


Review of Fatal and Severe Injury Crashes and Evaluation of Geometric Parameters that Affect Truck Stability
http://teachamerica.com/RAB17/RAB172A_Steyn/index.htm

So rollover is the dominant problem for trucks... Interesting... and slowing down to pedestrian pace is the solution? 

[kphoger]

What pedestrian walks at 10mph?

[jakeroot]

This PDF from Teach America covers the roundabout design guidelines in Germany. As it relates to the topic at hand, Germany does not use truck aprons at all for larger roundabouts, and uses a much shorter variation at smaller single-lane roundabouts. The apron, rather than the raised type seen in the US, is more of a slight bump, composed of asphalt, stamped concrete, or bricks (judging by the pictures on page 6). Larger (rural) single-lane roundabouts do not have truck aprons because the diameter does not require them, and multi-lane roundabouts also do not require them because of the additional pavement width. Something tells me Germany doesn't have quite the same rollover issues.

Page 11 also covers the landscaping of roundabouts. Roundabouts on state highways used to have quite a few large structures built in the middle, but after a very severe collision that killed two people, the government ordered a review, and many of the structures were removed. For the most part, the middle of roundabouts no longer have any vertical elements, such as columns, trees, or rocks. Roundabouts in low-speed urban environments do not have the same restrictions (although all decorations must be approved).

[kalvado]

This PDF from Teach America covers the rounbabout design guidelines in Germany. As it relates to the topic at hand, Germany does not use truck aprons at all for larger roundabouts, and uses a much shorter variation at smaller single-lane roundabouts. The apron, rather than the raised type seen in the US, is more of a slight bump, composed of asphalt, stamped concrete, or bricks (judging by the pictures on page 6). Larger (rural) single-lane roundabouts do not have truck aprons because the diameter does not require them, and multi-lane roundabouts also do not require them because of the additional pavement width. Something tells me Germany doesn't have quite the same rollover issues.

Page 11 also covers the landscaping of roundabouts. Roundabouts on state highways used to have quite a few large structures built in the middle, but after a very severe collision that killed two people, the government ordered a review, and many of the structures were removed. For the most part, the middle of roundabouts no longer have any vertical elements, such as columns, trees, or rocks. Roundabouts in low-speed urban environments do not have the same restrictions (although all decorations must be approved).

Another nice chart is the one with roundabouts types and dimensions for different traffic volumes.
Once a 100 ft = 30 m roundabout is put into a 52k traffic intersection - while german standard calls for a signalized one with 60m=200' dia minimum, you do get one of most dangerous intersections in the area... I wonder if NYSDOT did that on purpose..
Overall, comparing those guidelines with actual NYS structures provides a nice demo of indoctrination that roundabouts are assumingly superior design - without understanding of limitations.

[tradephoric]

Here is that chart from Germany that looks at roundabouts based on ICD and ADT.  Ignoring signalized roundabouts, the cap for the applicability of roundabouts in Germany is about 30,000 ADT.   



Here is an equivalent chart taken from the FHWA website.  The cap for the likely acceptability of double-lane roundabouts in the US is about 30,000 AADT (the same as Germany).   Roundabouts may be sufficient at higher AADT but additional analysis would be needed. 




The M-5 and Pontiac Trail roundabout in Michigan has an AADT approaching 55,000.  That's almost double the cap that Germany and the FHWA set as an acceptable volume for roundabouts.  A complex triple-lane roundabout with daily traffic volumes approaching 55,000 AADT isn't going to perform the same as a simple single-lane roundabout with 15,000 AADT... the type of roundabout analyzed in the IIHS study back in 2000 that really started the roundabout indoctrination.  And maybe the roundabouts built back in the early 2000s are great, but the roundabouts being built in America today are a whole other beast!

[jakeroot]

The M-5 and Pontiac Trail roundabout in Michigan has an AADT approaching 55,000.

Fucking hell. Do you know what the AADT was before the roundabout was installed?

[tradephoric]

The M-5 and Pontiac Trail roundabout in Michigan has an AADT approaching 55,000.

Fucking hell. Do you know what the AADT was before the roundabout was installed?

Before it was T-intersection and now it's a 4-leg roundabout.  With one less leg of traffic, volumes before the roundabout were inevitably lower (can't find an exact number.. my guess would be 35-40k).  By the way the 55,000 AADT may be conservative.  According to this article the spokesperson for the Road Commission of Oakland County says the roundabout carries 60,000 trips per day.

http://www.wxyz.com/news/region/oakland-county/despite-166-accidents-officials-say-roundabout-makes-m-5pontiac-trail-intersection-safer

[kalvado]

The M-5 and Pontiac Trail roundabout in Michigan has an AADT approaching 55,000.

Fucking hell. Do you know what the AADT was before the roundabout was installed?

I don't think anyone cared. Roundabouts are seen as an ultimate high throughput solution and a good fit for any situation.

[tradephoric]

The M-5 and Pontiac Trail roundabout in Michigan has an AADT approaching 55,000.

Fucking hell. Do you know what the AADT was before the roundabout was installed?

I don't think anyone cared. Roundabouts are seen as an ultimate high throughput solution and a good fit for any situation.

Watch any PM rush at the M-5/Pontiac Trail roundabout and you see drivers on WB Pontiac Trail backed up to the Walmart just creeping along at a snail's pace.  It routinely takes 10-15 minutes to get through the roundabout.  That is equivalent to drivers waiting 5-8 cycles to get through a light that is running a 120 second cycle.  The problem is NB M-5 traffic is so dominate that it overtakes the roundabout and there are very few gaps for WB Pontiac Trail traffic to enter.  High AADT roundabouts can break down pretty badly during rush hour especially when a dominate leg is present. 

When another major roundabout was being proposed in the area, they put together a traffic model using estimated 2035 Peak hour volumes.  Now that the 14 Mile and Orchard Lake roundabout has been built, it's laughable at how inaccurate the model was at depicting real world conditions.  There is a traffic signal literally 175 feet upstream from the roundabout and we are to believe that traffic will never back up through the roundabout?  Somebody should really take some drone footage of the roundabout and compare it to this model.  By the way, this roundabout was the most crash prone intersection in Michigan in 2016 experiencing 163 crashes including 27 injury crashes. 

https://www.youtube.com/watch?v=EpBJ_g63Q70

Now that we know what we know... 163 crashes, 27 injury crashes, traffic that routinely backs up through the roundabout.... was this roundabout a good idea?  Even the most ardent supporters of roundabouts should question that.

14-Orchard roundabout is tops for accidents, but is it really dangerous?
http://www.freep.com/story/news/local/michigan/oakland/2017/04/04/orchard-roundabout-tops-accidents-really-dangerous/100046220/

[tradephoric]

1 person dies when car slams into tree in Summerlin roundabout
https://www.reviewjournal.com/local/summerlin/1-person-dies-when-car-slams-into-tree-in-summerlin-roundabout/

A fatal crash happened at the Village Center Circle and Hills Center Drive roundabout in Summerland, NV on Wednesday night.  The Village Center Circle roundabout is credited as being the first modern roundabout built in America.  The driver was able to get out of the vehicle but the passenger had to be freed by rescue personal and was pronounced dead at the hospital.  Here is a streetview of the central island:



Some brush and shrubs planted throughout the large central island would be effective at slowing down an out of control driver without the vehicle coming to a "dead stop".  Blowing through 50 yards of shrubbery would get a drunk driver's attention and probably disable their vehicle.  Then the driver can be placed in handcuffs instead of being placed on a gurney after slamming into a palm tree.

[jeffandnicole]

Some brush and shrubs planted throughout the large central island would be effective at slowing down an out of control driver without the vehicle coming to a "dead stop".  Blowing through 50 yards of shrubbery would get a drunk driver's attention and probably disable their vehicle.  Then the driver can be placed in handcuffs instead of being placed on a gurney after slamming into a palm tree.

The driver was placed in handcuffs for DUI.  It was the passenger who died.

Then again, there's something she could've been wearing that might've reduced her chances of dying as well.

His 56-year-old female passenger, who wasn't wearing a seat belt...

While the roundabout was where the fatal accident occurred, better them slamming into a tree than into another innocent vehicle.

And, as this is supposedly the 'first' modern roundabout, if there have been no other fatal accidents here, speaks to how well the roundabout has worked over the years.  Even the most safest roads will have a rare fatal on occasion; mostly due to other circumstances, which in this case was a drunk driver transporting an unbuckled passenger.

[kalvado]

Some brush and shrubs planted throughout the large central island would be effective at slowing down an out of control driver without the vehicle coming to a "dead stop".  Blowing through 50 yards of shrubbery would get a drunk driver's attention and probably disable their vehicle.  Then the driver can be placed in handcuffs instead of being placed on a gurney after slamming into a palm tree.

The driver was placed in handcuffs for DUI.  It was the passenger who died.

Then again, there's something she could've been wearing that might've reduced her chances of dying as well.

His 56-year-old female passenger, who wasn't wearing a seat belt...

While the roundabout was where the fatal accident occurred, better them slamming into a tree than into another innocent vehicle.

And, as this is supposedly the 'first' modern roundabout, if there have been no other fatal accidents here, speaks to how well the roundabout has worked over the years.  Even the most safest roads will have a rare fatal on occasion; mostly due to other circumstances, which in this case was a drunk driver transporting an unbuckled passenger.

In a grand scheme of things, most of the roads around are incredibly safe. We're talking about average crash rate of (round up to make my calc easier) of 10 in 1 million per intersection.
I am going through about 10 intersections a day - that means I (assuming I am an average driver) can expect 1 crash in 10 000 days - or 1 in 30 years,  give or take. And most of those crashes are property only..
With that - significant number of those accidents are due to driver being not fit to drive or making a mistake. Drunk, sick, old, under influence, stressed, tired, on the phone, unnecessary changing lanes, rushed through traffic light  etc. Alert and attentive driver is much safer than average.
And if you want to keep accidents in check - you need to take those drivers into account. Yes, we need to push for DUI awareness etc - but extermination of offenders is not the only approach

[kphoger]

Why is it even possible to drive a straight path into the center island?  Shouldn't deflection direct a driver away from that?  Guardrails (guiderails?) along the approaches might help with that.

[kalvado]

Why is it even possible to drive a straight path into the center island?  Shouldn't deflection direct a driver away from that?  Guardrails (guiderails?) along the approaches might help with that.

Guardrails look fairly ugly for urban locations. And with those curved approaches, number of expensive property-only accidents would go up as people would scrape those rails - while now it is just running the curb.
Some soft stopping items in central island - no trees no concrete - seem to be the least evil.

[tradephoric]

While the roundabout was where the fatal accident occurred, better them slamming into a tree than into another innocent vehicle.

As mentioned in another thread, retractable bollards should be added at signalized intersections.  The bollards would extend up when the signal turns red.  It is better for a red light driver to slam into a bollard than to T-bone somebody already in the intersection.  Would you be in favor of this idea?  I found a crash test on youtube that simulates a red light runner plowing into the bollards.

https://www.youtube.com/watch?v=9FfUlojxiqE