Crash prone 'modern roundabouts'

[jakeroot]

I'm not sure whether or not this study has been linked here, or not (I don't believe it has); for the sake of the following comment, let's presume it has not been.

Back in 2002, several FHWA employees visited Sweden, Germany, the Netherlands, and the UK to study different traffic signal/control schemes practised in each of those countries. One of the biggest takeaways from this visit was the roundabout. Though roundabouts had already been built in the US, there hadn't been much official documentation on them (hence why it took until 2009 to develop marking and design standards). Throughout the document, the FHWA employees note that roundabouts are effective at reducing the severity of collisions, but not the number of collisions:

4. Promote roundabouts as alternatives to signalized intersections as a way to manage the consequences of collisions (severity versus frequency).

...The primary focus is on crash severity, not frequency....On the basis of this principle, SNRA has converted signalized intersections to roundabouts, recognizing that the frequency of total accidents may increase, but the severity of those accidents may be greatly reduced.

When high-accident locations have been identified, progressive solutions are developed...In some cases, where accident severity is high, Sweden has removed traffic signals and replaced them with roundabouts, recognizing that overall accident rates may increase and line-of-sight may be degraded, but the rate of severe (fatal and injurious) accidents will decrease...In urban areas, unsignalized roundabouts can have a negative systemwide effect because it is difficult to control and manage platooning and traffic progression.

And, on severity alone:

The turbo roundabout...is used in the Netherlands to eliminate weaving conflicts found in standard multilane roundabouts...This low-speed configuration also has been found to allow a higher capacity than the standard two-lane roundabout. While a standard roundabout has 16 potential conflict points, the turbo roundabout has 10. In either case, the conflict points are low speed, and the resulting accidents are typically less severe than those at signalized intersections (Figure 5-5).

There are additional comments -- you can read the study here: http://goo.gl/6MQq2Z

I do wonder why initial studies were so focused on demonstrating the reduction of collisions, versus the severity. I know both were part of the studies, but at least as far as I can tell, the biggest plus of roundabouts is the reduction in severity, not any sort of reduction of overall collisions.

I don't doubt the benefits of the roundabout. It's pretty obvious that roundabouts reduce the severity of collisions, but, the sheer number of collisions at a given intersection doesn't seem to have any connection to the geometric design of said intersection. There are some signals with a lot of collisions, and some roundabouts with a lot of collisions. Equally, there are some signals with very few collisions, and some roundabouts with very few collisions.

tl;dr: the primary benefit of the roundabout is reduced severity.

[kalvado]

I do wonder why initial studies were so focused on demonstrating the reduction of collisions, versus the severity. I know both were part of the studies, but at least as far as I can tell, the biggest plus of roundabouts is the reduction in severity, not any sort of reduction of overall collisions.
[...]
tl;dr: the primary benefit of the roundabout is reduced severity.

My strong impression is that primary advantage of roundabouts in US is ability to get more money into construction contracts, and who cares what bullshit is used to justify that to the crowd.

[jakeroot]

I do wonder why initial studies were so focused on demonstrating the reduction of collisions, versus the severity. I know both were part of the studies, but at least as far as I can tell, the biggest plus of roundabouts is the reduction in severity, not any sort of reduction of overall collisions.
[...]
tl;dr: the primary benefit of the roundabout is reduced severity.

My strong impression is that primary advantage of roundabouts in US is ability to get more money into construction contracts, and who cares what bullshit is used to justify that to the crowd.

I'm not usually convinced by conspiracies....this is not an exception.

[kalvado]

I do wonder why initial studies were so focused on demonstrating the reduction of collisions, versus the severity. I know both were part of the studies, but at least as far as I can tell, the biggest plus of roundabouts is the reduction in severity, not any sort of reduction of overall collisions.
[...]
tl;dr: the primary benefit of the roundabout is reduced severity.

My strong impression is that primary advantage of roundabouts in US is ability to get more money into construction contracts, and who cares what bullshit is used to justify that to the crowd.

I'm not usually convinced by conspiracies....this is not an exception.

Not a conspiracy, just plain old corruption. Which is just business as usual.

[jakeroot]

I do wonder why initial studies were so focused on demonstrating the reduction of collisions, versus the severity. I know both were part of the studies, but at least as far as I can tell, the biggest plus of roundabouts is the reduction in severity, not any sort of reduction of overall collisions.
[...]
tl;dr: the primary benefit of the roundabout is reduced severity.

My strong impression is that primary advantage of roundabouts in US is ability to get more money into construction contracts, and who cares what bullshit is used to justify that to the crowd.

I'm not usually convinced by conspiracies....this is not an exception.

Not a conspiracy, just plain old corruption. Which is just business as usual.

Right...corruption. Got it m8.

Anyone here interested in discussing roundabouts?

[kalvado]

Right...corruption. Got it m8.

Anyone here interested in discussing roundabouts?

Sure. Are there any estimates on demolition costs for those?

[jakeroot]

Right...corruption. Got it m8.

Anyone here interested in discussing roundabouts?

Sure. Are there any estimates on demolition costs for those?

Never heard of such a thing. Maybe tradephoric has some idea.

[kalvado]

Right...corruption. Got it m8.

Anyone here interested in discussing roundabouts?

Sure. Are there any estimates on demolition costs for those?

Never heard of such a thing. Maybe tradephoric has some idea.

Which is actually a big problem of those roundabouts. There is no clear understanding why they should be built; once studies showing how great those are turn out irrelevant, there is a different argument. There is no understanding of when roundabouts actually work best; I didn't see realistic estimates of throughput. There is no understanding of future costs - such as demolition. 
Two things are known for sure: that they must be built, and how much they cost. I can see two reasons for that:  corruption in financing construction and old good stupidity.
If I have to choose which one is more realistic, I definitely choose both.

[jakeroot]

Right...corruption. Got it m8.

Anyone here interested in discussing roundabouts?

Sure. Are there any estimates on demolition costs for those?

Never heard of such a thing. Maybe tradephoric has some idea.

Which is actually a big problem of those roundabouts. There is no clear understanding why they should be built; once studies showing how great those are turn out irrelevant, there is a different argument. There is no understanding of when roundabouts actually work best; I didn't see realistic estimates of throughput. There is no understanding of future costs - such as demolition. 
Two things are known for sure: that they must be built, and how much they cost. I can see two reasons for that:  corruption in financing construction and old good stupidity.
If I have to choose which one is more realistic, I definitely choose both.

Whoa whoa whoa, slow down partner. You asked me about demolition costs ... I told you I didn't know ... then you respond with some rambling, incoherent paragraph, darting from point to point, apparently at random, only to arrive at an indecipherable conclusion ("If I have to choose which one...I definitely choose both.").

I've never had to ask someone to do this before, but...please re-write your post. I want to be able to reply with a decent response, but I simply don't understand what you're saying.

[english si]

I've never had to ask someone to do this before, but...please re-write your post. I want to be able to reply with a decent response, but I simply don't understand what you're saying.

Seems to be saying that the crab people thrust roundabouts on us, without telling us the costs of demolishing them when we inevitably (and kalvado does think it's inevitable for some reason) see through the crab people's arguments that we need them and so that's a hidden cost of building them - the cost of demolishing the blighters when we find them shite.

[kalvado]

Whoa whoa whoa, slow down partner. You asked me about demolition costs ... I told you I didn't know ... then you respond with some rambling, incoherent paragraph, darting from point to point, apparently at random, only to arrive at an indecipherable conclusion ("If I have to choose which one...I definitely choose both.").

I've never had to ask someone to do this before, but...please re-write your post. I want to be able to reply with a decent response, but I simply don't understand what you're saying.
[/quote]

OK, same thing in a more formalized basis.

1. Current altitude to roundabouts (RA) among traffic engineers is based on faith, not on fact.
1a. When initial research on how great RA are proved wrong, a new set of "they are still great!" texts emerged.
1b. There are still no serious research regarding applicability of RA in terms of handleable traffic volumes, types of roads where RA can work best (2x2 lane discussion in this thread)  etc.
1c. Some official recommendation pushing RA as first choice are reinforcing religious altitude
1d. No understanding of full lifecycle costs; despite declared low-maintenance, things like re-training snow plow drivers and improvising and relocating pedestrian crosswalks after construction "because we're on learning curve", as well as costs of demolition of messy structure are not taken into account on decision making stage.
1e. I really doubt that organization, which is extremely busy with defining types of fonts to be used on signs, has enough technical expertise to develop understanding such complex system.

2. Cost of building roundabout is significant and mostly out of public control since there is little basis for comparison.
2a. That attracts construction companies, since MONEY. Visibly same structure can cost $0.5M and $2M within 20 miles radius (local example)
2b. That attracts politicians, since that allows to flow money in a non-transparent way.

Harsh? definitely. True? At least partially.   

[jeffandnicole]

2. Cost of building roundabout is significant and mostly out of public control since there is little basis for comparison.
2a. That attracts construction companies, since MONEY. Visibly same structure can cost $0.5M and $2M within 20 miles radius (local example)
2b. That attracts politicians, since that allows to flow money in a non-transparent way.

None of this is unique to roundabouts.  An intersection replacement costs can vary widely, depending on what needs to be done.  And there's a ton of basis of comparison...they know how much land they need, they know the various costs.  They can do an analysis of the cost of an intersection vs. the cost of a roundabout, along with ongoing maintenance and electricity costs.  All of this is pretty typical stuff.

Many states puts these projects out to bid, and there's plenty of contractors out there willing to bid on the projects.  Again, much of this can be found online. 

[kalvado]

2. Cost of building roundabout is significant and mostly out of public control since there is little basis for comparison.
2a. That attracts construction companies, since MONEY. Visibly same structure can cost $0.5M and $2M within 20 miles radius (local example)
2b. That attracts politicians, since that allows to flow money in a non-transparent way.

None of this is unique to roundabouts.  An intersection replacement costs can vary widely, depending on what needs to be done.  And there's a ton of basis of comparison...they know how much land they need, they know the various costs.  They can do an analysis of the cost of an intersection vs. the cost of a roundabout, along with ongoing maintenance and electricity costs.  All of this is pretty typical stuff.

Many states puts these projects out to bid, and there's plenty of contractors out there willing to bid on the projects.  Again, much of this can be found online.

NYSDOT tend to publish lump sum costs and nothing more than that. As for bidding... This is something what is likely to lend our governor either in jail or in the White house in near future. You just have to do it right!

[jeffandnicole]

2. Cost of building roundabout is significant and mostly out of public control since there is little basis for comparison.
2a. That attracts construction companies, since MONEY. Visibly same structure can cost $0.5M and $2M within 20 miles radius (local example)
2b. That attracts politicians, since that allows to flow money in a non-transparent way.

None of this is unique to roundabouts.  An intersection replacement costs can vary widely, depending on what needs to be done.  And there's a ton of basis of comparison...they know how much land they need, they know the various costs.  They can do an analysis of the cost of an intersection vs. the cost of a roundabout, along with ongoing maintenance and electricity costs.  All of this is pretty typical stuff.

Many states puts these projects out to bid, and there's plenty of contractors out there willing to bid on the projects.  Again, much of this can be found online.

NYSDOT tend to publish lump sum costs and nothing more than that. As for bidding... This is something what is likely to lend our governor either in jail or in the White house in near future. You just have to do it right!

BidX is your friend when it comes to all things bidded on about roads.  If you're maintain a thought process that includes everything the state does is evil and corrupt though, it's not going to be of any help.  There are rules that are generally followed, with few exceptions.

For NY: https://www.bidx.com/ny/main

[tradephoric]

It's pretty obvious that roundabouts reduce the severity of collisions

The severity index is simply a weighted average of fatal, injury, and PDO crashes.   Here is the formula from the FHWA:

Severity Index = (12 * F + 3 * I + 1 * PDO) / N
Where: 
F=total number of fatal crashes
I =total number of injury crashes
N = total number of crashes

Let's figure out the severity index of a multi-lane roundabout with the following stats:
AADT=40,000
Fatal crashes = 0
Injury crashes = 10
PDO crashes = 90
Total crashes = 100
Severity Index = (12 * 0 + 3 * 10 + 1 * 90) / 100  =  1.2

Now what is the severity index of a signalized intersection with the following stats:
AADT=40,000
Fatal crashes = 0
Injury crashes = 10
PDO crashes = 30
Total crashes = 40
Severity Index = (12 * 0 + 3 * 10 + 1 * 30) / 40 = 1.5

Sure, the roundabout had a lower severity of crashes but does that make it safer?  You have to consider the frequency of crashes to determine if the reduction in severity is actually leading to fewer injury crashes.  Stating that "roundabouts reduce the severity of crashes"  is just a clever way to mask the fact that roundabouts aren't always safer (IE. some 2x2 multi-lane roundabouts). 

[kalvado]

BidX is your friend when it comes to all things bidded on about roads.  If you're maintain a thought process that includes everything the state does is evil and corrupt though, it's not going to be of any help.  There are rules that are generally followed, with few exceptions.

For NY: https://www.bidx.com/ny/main

Thank you for the reference, will look at it later...
As about  state=evil.. not difficult in a state which lost couple of top leaders to federal prisons in a past year and governor shedding all his friends in attempt not to join the gang. But if you will, the root cause is unclear technical justification strongly supported at political level. I don't remember that level of support for SPUI, for example - probably because advantages become clear after single-digit drives through one of those.

[jakeroot]

Stating that "roundabouts reduce the severity of crashes"  is just a clever way to mask the fact that roundabouts aren't always safer (IE. some 2x2 multi-lane roundabouts).

There is no perfect, winner-take-all design. There are examples of every type of intersection with both a lot, and very few collisions. I specified this in my original post:

I don't doubt the benefits of the roundabout. It's pretty obvious that roundabouts reduce the severity of collisions, but the sheer number of collisions at a given intersection doesn't seem to have any connection to the geometric design of said intersection. There are some signals with a lot of collisions, and some roundabouts with a lot of collisions. Equally, there are some signals with very few collisions, and some roundabouts with very few collisions.

My point was simply that roundabouts, as a whole, have a better severity record. There are inexplicable exceptions, of course, but, they're exceptions.

And to be clear, to tie this back into your original post way back yonder, there are certainly some discrepancies in the FHWA's original studies. But, roundabouts are still pretty damn good at protecting from severe, life-threatening injuries, simply because their design makes such collisions rather difficult (unless of course you drive straight over the middle, like in Carmel, IN).

My post, at the top of this page, was to bring up what I thought was an interesting discrepancy (the FHWA knew full well that roundabouts don't always decrease collisions, but they still heavily marketed roundabouts with that tagline anyway).

[kalvado]

My post, at the top of this page, was to bring up what I thought was an interesting discrepancy (the FHWA knew full well that roundabouts don't always decrease collisions, but they still heavily marketed roundabouts with that tagline anyway).

And even reduced severity is a poor metrics. If that was the ultimate one, reducing speed limit on interstates to 25 MPH should be a no-brainer.
\

[jakeroot]

My post, at the top of this page, was to bring up what I thought was an interesting discrepancy (the FHWA knew full well that roundabouts don't always decrease collisions, but they still heavily marketed roundabouts with that tagline anyway).

And even reduced severity is a poor metrics. If that was the ultimate one, reducing speed limit on interstates to 25 MPH should be a no-brainer.

Not in my opinion. Two cars bumping into each other at 80 is safer than a car going 50 rear-ending a car going 25. Which is to say, set speed limits closest to where the majority of cars are going the same speed (lack of differential = less severe collisions).

[kalvado]

My post, at the top of this page, was to bring up what I thought was an interesting discrepancy (the FHWA knew full well that roundabouts don't always decrease collisions, but they still heavily marketed roundabouts with that tagline anyway).

And even reduced severity is a poor metrics. If that was the ultimate one, reducing speed limit on interstates to 25 MPH should be a no-brainer.

Not in my opinion. Two cars bumping into each other at 80 is safer than a car going 50 rear-ending a car going 25. Which is to say, set speed limits closest to where the majority of cars are going the same speed (lack of differential = less severe collisions).

And you realize that 56% of fatalities occur in single-vehicle accidents?
(putting on traffic engineer hat) to reduce severity of such accidents on freeways, speed of free flowing traffic must be reduced to match that with the new speed limit. traffic calming measures may include lane narrowing and possibly chicanes. While increasing travel times and severely hurting US economy, these measures would allow increased ticket revenue and would reduce crash severity for single vehicle accidents and number of fatalities. Safety first!

[tradephoric]

My point was simply that roundabouts, as a whole, have a better severity record. There are inexplicable exceptions, of course, but, they're exceptions.

Prince Fielder of the Texas Rangers is batting .197 and having a pretty abysmal year.  Brett Eibner of the Kansas City Royals is batting .471 and is lighting it up.  Can you predict who has more hits?  Prince Fielder is having a bad year but he still has 39 more hits than Eibner (since Eibner only has 17 at bats vs. Fielder's 238). 

You can have a lot more hits (ie. injury crashes) when you get more at bats (ie. frequency of crashes).  You admit that there is discrepancies in the FHWA's studies yet continue to regurgitate the misleading stats that they cite.  Sure, we can be clapping seals and say how great roundabouts are because they have low severity crashes.  But there should be some focus that at some of these 2x2 multi-lane roundabouts, the frequency of crashes has been abysmally high (leading to injury crashes).

Aren't roundabouts great!

[kalvado]

  You admit that there is discrepancies in the FHWA's studies yet continue to regurgitate the misleading stats that they cite.  Sure, we can be clapping seals and say how great roundabouts are because they have low severity crashes. 

That is actually what I call indoctrination. Studies may be flawed, but we know the Truth!

[jakeroot]

You can have a lot more hits (ie. injury crashes) when you get more at bats (ie. frequency of crashes).

So, let me get this straight: roundabouts that had more collisions after installation, also had an increase in injuries (in every case)?? ......

You admit that there is discrepancies in the FHWA's studies yet continue to regurgitate the misleading stats that they cite.  Sure, we can be clapping seals and say how great roundabouts are because they have low severity crashes.  But there should be some focus that at some of these 2x2 multi-lane roundabouts, the frequency of crashes has been abysmally high (leading to injury crashes)

The problem is that there has been very little formal follow-up to the FHWA's initials studies. Some DOT's have investigated before-and-after situations, sometimes finding that the roundabouts resulted in far more collisions than anticipated, but there hasn't been, at least to my knowledge, any sort of legit follow-up, covering the whole of the nation's roundabouts. We've cherry-picked some roundabouts in this thread, but we both know this thread isn't any sort of official study (although I will admit, as I already have dozens of times, that the FHWA's initial studies might be a little off).

[tradephoric]

The problem is that there has been very little formal follow-up to the FHWA's initials studies. Some DOT's have investigated before-and-after situations, sometimes finding that the roundabouts resulted in far more collisions than anticipated, but there hasn't been, at least to my knowledge, any sort of legit follow-up, covering the whole of the nation's roundabouts. We've cherry-picked some roundabouts in this thread, but we both know this thread isn't any sort of official study (although I will admit, as I already have dozens of times, that the FHWA's initial studies might be a little off).

I agree that more studies need to be done.  I already suggested a study that should be published...

This would be an eye opening study if it were published:

"The Safety Performance of Roundabouts with Average Daily Traffic exceeding 30,000"

The 30,000 ADT condition would weed out single-lane roundabouts and most 2x1 multi-lane roundabouts.  The focus would be nearly entirely on the safety performance of 2x2 (or higher) roundabouts in America. Ideally, the study would analyze interchange and non-interchange roundabouts separately and come up with two different sets of safety numbers.

[kalvado]

The problem is that there has been very little formal follow-up to the FHWA's initials studies. Some DOT's have investigated before-and-after situations, sometimes finding that the roundabouts resulted in far more collisions than anticipated, but there hasn't been, at least to my knowledge, any sort of legit follow-up, covering the whole of the nation's roundabouts. We've cherry-picked some roundabouts in this thread, but we both know this thread isn't any sort of official study (although I will admit, as I already have dozens of times, that the FHWA's initial studies might be a little off).

I agree that more studies need to be done.  I already suggested a study that should be published...

This would be an eye opening study if it were published:

"The Safety Performance of Roundabouts with Average Daily Traffic exceeding 30,000"

The 30,000 ADT condition would weed out single-lane roundabouts and most 2x1 multi-lane roundabouts.  The focus would be nearly entirely on the safety performance of 2x2 (or higher) roundabouts in America. Ideally, the study would analyze interchange and non-interchange roundabouts separately and come up with two different sets of safety numbers.

I suspect hourly traffic may be even more important. There are efforts on stretching commute hours - staged shifts, etc - to reduce hourly traffic while maintaining total number of vehicles.