Are diverging diamonds a fad?

[jakeroot]

The thing with DDIs is that there isn't much point to using them unless space is limited. There are huge operational advantages over standard diamond interchanges and they're a heck of a lot cheaper than SPUIs while better for pedestrians because of the shorter crossing distances. I'm waiting for NYSDOT to install more.

Not to sound too much tradephoric, but DDI's should, really, only be used if ROW is limited. If it's not, a partial cloverleaf, of some form, likely has much better operating capacity.

[cl94]

The thing with DDIs is that there isn't much point to using them unless space is limited. There are huge operational advantages over standard diamond interchanges and they're a heck of a lot cheaper than SPUIs while better for pedestrians because of the shorter crossing distances. I'm waiting for NYSDOT to install more.

Not to sound too much tradephoric, but DDI's should, really, only be used if ROW is limited. If it's not, a partial cloverleaf, of some form, likely has much better operating capacity.

Precisely. Out east, I see great potential for them because there is rarely room for a partial cloverleaf. I could see quite a few interchanges near me getting the treatment due to the high turn volumes and very tight ROW. The only non-tight ROW scenario it might be good for is where there is little through traffic. In this case, the signals would operate as 2-phase synchronized with through traffic possibly stopping twice.

[tradephoric]

Since DDIs are being constructed with large footprints that could accomodate Parclos, it's fair to compare the operational performance of each.  CL94, I asked you why Canada prefers the Parclo A4 over the B4 and you gave me the following reasons. 

1. Exit ramp is long and straight, providing space for deceleration and reducing the chance of accidents.
2. Turning traffic typically doesn't have to turn cross the path of opposing traffic.
3. All entrances are on the right, providing consistency.
4. Fewer exits from the expressway. Standard is to keep the number of departure points as low as possible.

For each one of your points, strong counter-points can be made that supports the B4 (I listed these counter-points in a previous post...feel free to respond).  Secondly, none of your points actually address the operational performance for each type of Parclo.  How can you determine what type is Parclo is preferred when you don't know the MOEs for each?   If you care about average driver delay, total number of stops, or queue lengths... the Parclo B4 wins. 

There's a balance between delay and cost. Usually, the proposals at any extreme are unacceptable because the balance is not met. You need to be able to sell it to the public. Here, i think the DDI is easier to sell. Hell, being at an MPO, that's a huge part of my job-selling it to the public.

Since both Parclos require similar ROWs, the cost for each should be nearly identical.  The focus can shift towards which type of interchange is actually better at moving traffic through it.  Can you find any literature that indicates the Parclo A4 is more efficient?

[tradephoric]

Here is a summary of the results for both balanced and un-balanced traffic conditions.


http://fau.digital.flvc.org/islandora/object/fau%3A3779/datastream/OBJ/view/Comparative_analysis_between_the_diverging_diamond_interchange_and_partial_cloverleaf_interchange_using_microsimulation_modeling.pdf

The ParClo B4 is superior to the ParClo A4 in every simulation tested when analyzing delay time, stop time, and number of stops.  Under no circumstance was the A4 better.  The only MOE where the ParClo A4 outperformed the ParClo B4 was Max Queue at High 2 & 3.  The reason for this is explained in the report:

The maximum queue length on a ParClo B4 occurs at the off-ramps, where cars are trying to enter the crossroad, but are unable to do so since the crossroad is very congested and the vehicles wanting to go through the intersection do not have to stop since they have green throughout the whole cycle length.  For the High volume flows, the DDI designs had a much better performance than the ParClo designs.  The ParClo A4 had double the queue length than the DDIs, and the ParClo B4 had almost triple the maximum queue lengths at some instances compared to the DDIs.

Simply put, if the off-ramp signals of the ParClo B4 were signalized the max queue of the ParClo B4 would be comparable to the other interchange designs tested.  And there are many examples of ParClo B4s with signalized off ramps.  The ParClo B4 wins out over the Parclo A4 when analyzing common MOEs. 

[johndoe]

I must play devil's advocate on this thread

Saying DDI should only be considered on tight R/W corridors is inaccurate:
1. Take a look at those delay values trade posted.  All the DDI values, even the 4 lane geometry, are well within acceptable LOS.
2.  It's likely that FHWA approval will be easier to get for four ramps instead of six.  If a diamond interchange, for instance, is being upgraded it will take a lower-level report to the feds than adding two ramps.
3.  DOT around the country are strapped for cash.  Nearly all the parclo variations require more bridge area than DDI. 

Also, saying parclo b4 off-ramps can simply be signalized in order to reduce queue isn't fair.  What do you think this does to each vehicle's delay? Besides, giving those "left turners " their own lane to accelerate in is one of the parclo b4's best attributes.  I'm not saying signalizing them is a bad idea, just that it may diminish the very advantage the geometry has over others.

[cl94]

I must play devil's advocate on this thread

Saying DDI should only be considered on tight R/W corridors is inaccurate:
1. Take a look at those delay values trade posted.  All the DDI values, even the 4 lane geometry, are well within acceptable LOS.
2.  It's likely that FHWA approval will be easier to get for four ramps instead of six.  If a diamond interchange, for instance, is being upgraded it will take a lower-level report to the feds than adding two ramps.
3.  DOT around the country are strapped for cash.  Nearly all the parclo variations require more bridge area than DDI. 

Also, saying parclo b4 off-ramps can simply be signalized in order to reduce queue isn't fair.  What do you think this does to each vehicle's delay? Besides, giving those "left turners " their own lane to accelerate in is one of the parclo b4's best attributes.  I'm not saying signalizing them is a bad idea, just that it may diminish the very advantage the geometry has over others.

Glad I have someone to back me up. The main advantage of a DDI is that the bridge can be significantly narrower in most cases (2+ lanes less) and the bridge is by far the most expensive portion of interchange construction. In the event of an upgrade to an existing diamond, the existing bridge can stay. Good luck doing that with a parclo. I don't know what kind of experience tradephoric has with design and in the field, but from knowing what I do, delay time doesn't mean crap if (a) the LOS remains acceptable and (b) the reduction in delay doesn't offset the increase in cost. Parclos are expensive. DDIs cost no more than a simple diamond, maybe a little less because the bridge can be a lane or two narrower.

[tradephoric]

Yes cl94, I agree that cost is an important factor to consider but let's get back on point.  Previously you said there was a reason Canada prefers the Parclo A4 over the B4.    I'm trying to pin you down on why you believe this.  The cost for each type of parclo is relatively close (since they require roughly the same ROW and bridge deck widths).  While both types of parclos have similar costs, the Parclo B4 is superior when analyzing common measurements of effectiveness under varying traffic conditions (balanced and unbalanced).  If two interchanges have similar costs, wouldn't you pick the one that performs better?   

I'm not impressed when people talk about how smart they are.  It's great that you work in the field... a lot of us do.  But citing your resume is not a way to make a compelling argument.  Cite a study of your own that supports your side of the argument or at the very least try to poke holes in the study i cited.  Do something other than regurgitate how cost is important.  I'll ask you again, why does Canada prefer the Parclo A4 over the Parclo B4? 

[vdeane]

Because they don't like that android in Star Trek: Nemesis

[tradephoric]

Also, saying parclo b4 off-ramps can simply be signalized in order to reduce queue isn't fair.  What do you think this does to each vehicle's delay? Besides, giving those "left turners " their own lane to accelerate in is one of the parclo b4's best attributes.  I'm not saying signalizing them is a bad idea, just that it may diminish the very advantage the geometry has over others.

The reduction in delay at a Parclo B4 is due to the on/off ramp traffic signals only stopping one-direction of travel and they can easily be coordinated to each other (regardless of how close the signals are).  It essentially functions as one-way street pair.  With a Parclo A4 you have traffic signals that stop both directions of traffic, leading to excess delay because you can't optimize the signal timing for both directions of travel.  That's the meat and potatoes as to why the Parclo B4 is superior at reducing driver delay (quite simply, it's easier to time a one-way street than a two-way street).  But to your point adding additional on/off ramp signals at a B4 will likely increase total delay.  Will the Parclo A4 suddenly have better MOE's though?  No way.  And there are some major advantages to signalizing the ramps:

1.   It makes it easier for pedestrians to cross the street.
2.   Off-ramp drivers who need to make a left-turn at an upstream signal don't have to merge and fight their way over into the left-most lane.
3.   It reduces the required bridge deck widths (since acceleration lanes are no longer needed).

Consider the following Parclos that service a 6-lane arterial street.  The Parclo B4 has double left-turns that extend onto the bridge deck (but since the off-ramps are signalized there are no acceleration/deceleration lanes).  Keep in mind that this is a B4 from Florida.  Engineers in Florida have to design for extremely long storage lanes because they run incredibly long cycle lengths (since they have long pedestrian crossings along their major corridors).  The point is this is a worst case scenario and in other states the left-turns likely wouldn't need to extend onto the bridge deck.  OTOH, the Parclo A4 doesn't have any left-turns but it requires acceleration lanes for on-ramp drivers.  When comparing the bridge deck widths of both, it turns out the Parclo A4 has a wider deck (even though the A4 doesn't have any pedestrian facilities while the Parclo B4 does).   This is an example of how signalizing the off-ramps at the B4 can help reduce the total bridge deck width.

Parclo A4:

https://www.google.com/maps/@41.28607,-96.07538,657m/data=!3m1!1e3

Parclo B4:

https://www.google.com/maps/@26.00754,-80.34041,17z/data=!3m1!1e3

[tradephoric]

delay time doesn't mean crap if (a) the LOS remains acceptable and (b) the reduction in delay doesn't offset the increase in cost.

Does delay time mean anything to you if two interchange alternatives cost about the same?   If you admit the cost for each type of parclo is similar then you would be forced to look at which performs better.   For whatever reason you don't want to do this, so you will probably make some non-convincing arguments why the Parclo A4 is actually cheaper to construct.  You said Canada prefers the Parclo A4 over the Parclo B4.  All i'm asking is why.

[cl94]

delay time doesn't mean crap if (a) the LOS remains acceptable and (b) the reduction in delay doesn't offset the increase in cost.

Does delay time mean anything to you if two interchange alternatives cost about the same?   If you admit the cost for each type of parclo is similar then you would be forced to look at which performs better.   For whatever reason you don't want to do this, so you will probably make some non-convincing arguments why the Parclo A4 is actually cheaper to construct.  You said Canada prefers the Parclo A4 over the Parclo B4.  All i'm asking is why.

You're taking my comment out of context. I never said that in relation to why Canada uses A4s. That particular cost comment relates to why some places use DDIs. I listed what is published online by Canadian agencies and that is all I have. Maybe one of the Canadian members could answer your question. I cannot. All I know is that they avoid them at all costs and their few B4s have significant traffic issues.

[tradephoric]

^I have been asking you about the operational differences between the Parclo A4 and Parclo B4 for the past few posts.  You know that's what I've been asking.  Instead you deflected from the question and focused on DDI costs.  Let's be clear, the reason your comment is being taken out of context is because you refused to answer my question.  Now you finally admit that your previous list of reasons why Canada prefers the Parclo A4 is nothing more than generic talking points you read from some online publication.  And yet you still try to discredit the B4 by saying that "the few B4s have significant traffic issues" .   A lot of interchanges have significant traffic issues.  That's as vague as the "sky is blue" .   

Maybe Canada doesn't like exiting loop ramps and think it's a safety concern.  I personally feel that point is way overblown.  There are literally hundreds upon hundreds of interchanges in North America that have exiting freeway loop ramps.  Assuming the results of the study i cited are valid, Canada is willing to double or triple driver delay within an interchange just to avoid a few additional run off the road accidents?  If Canada is so concerned with the safety of exiting loop ramps, why don't they eliminate Parclo AB interchanges (which include exiting loop ramps)?  There are plenty of those in Canada yet they don't seem to have a problem with them.