Talk:Continuous-flow intersection

Definition
What is a "turn conflict?"--Greatjones 00:45, 6 February 2006 (UTC)
 * When the path of one vehicle intersects with the path of another vehicle. --Thisisbossi 15:31, 4 October 2006 (UTC)

Good source for info
--SPUI (T - C) 03:06, 23 July 2006 (UTC)
 * Agreed. Added it into the External Links. --Thisisbossi 15:31, 4 October 2006 (UTC)

My recent revert
Perhaps I am nitpicking on semantics, but I just undid an edit that replaced "Motorists sometimes cite discomfort" with "Motorists sometimes cite safety issues". In the same sense as a square is a rectangle, but not vice-versa; safety concerns include discomfort, but discomfort does not necessarily indicate a safety issue. An adequately designed CFI can still violate driver expectancy due to the "keep left" configuration and therefore will cause discomfort to motorists, but such does not necessarily translate to a safety risk in excess of other typical intersections.

Also, included in the same revert, was the mention of using Jersey walls for delineation. These are not typically recommended due to the crash risk that they pose in the event of collision and also due to sight distance limitations. While their use is definitely feasible and I'm sure that there is or will be a CFI using such barrier walls, the phrasing used appeared to indicate that Jersey walls are a requirement for an adequately-designed CFI.

Lastly, length is not the sole trait of an adequately-designed acceleration lane, though I will agree it is an important trait. I feel that my phrasing provides a more vague approach rather than singling out only one trait. On a CFI involving lower speeds, length tends to become less relevant; rather the form of merge control becomes a higher priority. Turning radii and lane widths are other important factors.

--Thisisbossi 05:47, 2 February 2007 (UTC)

Agencies must pay royalites?
It seems U.S. agencies don't have to pay because this US patent expired October 15, 2003.

http://portal.uspto.gov/external/portal/pair enter 5049000 as the patent number.

Juxtapos99 10:51, 18 October 2007 (UTC)


 * I Checked this out, and it seems to be true. I'll add this to the article. &mdash; Val42 21:08, 19 October 2007 (UTC)

Concerns about clarity
I find the narrative and the monochrome diagram to be quite confusing. I have good spacial skills but am not a traffic professional or maven. Perhaps a color-coded diagram would be the single most helpful improvement. --Jdmstl 03:56, 31 October 2007 (UTC)
 * I'm planning on redoing the diagram because of other issues, but tell me about this color coding idea that you have. &mdash; Val42 05:19, 2 November 2007 (UTC)
 * I think the comment requests that each movement be color-coded. For example, the path of northbound lefts might be blue; northbound throughs red; eastbound lefts yellow; etc... -- Bossi  ( talk • gallery • contrib ) 21:12, 17 December 2007 (UTC)

"outside Mexico" vs "in the US"
I changed the section title "Locations outside Mexico" to "CFI Locations in the United States". I found the title confusing when reading the article. Though the article mentions the original development being in Mexico, there's nothing to indicate that all other CFI implementations are in Mexico, leaving it open whether examples exist other than in MX and US. If there really are no others, then it would make sense to clarify this both in the text and by further altering the section title as appropriate.Paleolith (talk) 01:35, 1 November 2010 (UTC)

US versus the world... Since Wikipedia is deliberately a global resource, the authors of this article should reference the fact that CFIs are an import from Mexico. That would certainly help clarify your "outside Mexico vs in the US" problem. Obviously, Wikipedia is user-based, but you should at least include one original reference. It's almost like claiming that Shakespeare was from Connecticut. —Preceding unsigned comment added by 99.254.16.135 (talk) 11:49, 16 March 2011 (UTC)

Description is useless and confusing
The top photo shows cars on a road and is completely useless for the purposes of the article. The second photo shows a couple of lanes that could be from any intersection of two roads anywhere and is completely useless for the purposes of the article. The small diagram, which I have studied in the "no higher resolution available" version, is confusing and therefore useless. As far as I can tell from the diagram currently in this article, three lanes of southbound traffic pass jaggedy-like through the intersection then smash into two lanes of oncoming traffic that intend to turn left, by trying to be the same place at the same time, causing fatal accidents every few minutes, and nobody with any sense would seriously consider building this type of intersection. Now, either this article needs to have a sentence at the top saying that this whole article is a hoax, or it needs a clear diagram showing how this intersection design could possibly be sensible. By contrast, see Diverging_diamond_interchange or Michigan_Left which use clear diagrams and aerial photos to show the traffic flow. Don't bother replying here; if the article can be fixed, fix it. 76.21.81.159 (talk) 05:16, 2 August 2011 (UTC)
 * The article cannot be fixed because the term "continuous-flow intersection" is misleading and invented for promotional purposes. There is no continuous flow in this sort of intersection; traffic must be controlled by traffic lights. The only true continuous flow intersection is the Cloverleaf. 173.174.85.204 (talk) 19:56, 1 August 2016 (UTC)Eric

Safety, capacity, delay reduction and design options (for continuous flow intersections)
I have analysed and sketched continuous flow options for 33 sites in my previous job and now my old employer has has been directed to examine 7 of them for Hoddle St, the major north-south arterial in Melbourne. In all cases, there were major capacity and delay benefits (generally 50% capacity increase and 50% delay reduction). Now I realise that there is a theoretical safety benefit that is also huge. This is supported by the one site in Utah for which there is impressive crash data. I have published sketches and theory on my web site at quicksafeintersections.com. Also there are simulations for these 7 intersections on you tube. There are many options in the design and I keep making improvements, many of which are at variance with the Utah examples (I think they are only removing turns from one road and not getting further capacity, safety & delay benefits from removing the other turn phases.) Further, the public have access to Victoria, Australia crash data and can test whether the Utah constraints of access control are out weighed by the benefits. At this point, I am thinking that contributing to the wiki might be a better way to go. Any advice on direction would be appreciated. Kindest regards Cleeland John (talk) 09:43, 4 September 2012 (UTC)Cleeland John (talk) 10:02, 6 July 2015 (UTC)

The disadvantages of the parallel flow intersection is that (a)it still requires four phases at the main intersection and so has double the delay. (b) It also interrupts both directions on the departures and so messes up the linking with all three intersections on each through route. — Preceding unsigned comment added by Cleeland John (talk • contribs) 03:54, 1 January 2013 (UTC)


 * @Cleeland John, sorry, we need to revert Your last edit. Wikipedia collects the scientific knowledge of the world. Sure, sometimes Wikipedia is written from lobbyists and activists. The link to (Your?) site is not what Wikipedia is used for, see WP:SPS. Giving Your site a visit: Those collector lanes needs to be protected by signal, if later queued traffic is faster, it does not yield its lanes while right turning traffic is following its green signal. A PFI is not that You described. PFI is similar to CFI, but just the parallel roadway is moved to the other leg of the intersection. The differences in performance may be caused by the last feet of the stop line or any other minor different. CFI and PFI are working the way they are designed due 120 meters nearly equals 400 feet and this is the compromise of roadway length from waiting vehicles to signal time to improve traffic flow. The samples (You?) created and uploaded on Youtube does not fulfill this. This might be a problem in performance and safety. Note: Some intersections collect traffic for queueing it on more lanes. CFI and PFI eliminate this extra lanes and use it for driving, not for waiting. As an engineer, calculate the worst case of estimeted traffic and have the result be a part of the signal timing. If this is not enough, see what a de:Geschwindigkeitssignal (translate it with Google) is and know what time it takes to make jammed traffic go fluently or just keeping it flowing. Some contries regulate the design of intersection types, some just regulate the design to eliminate traffic savety issues. -- Hans Haase (有问题吗) 14:54, 6 July 2015 (UTC)

Thanks for the response Hans. The collector lanes are proposed to be protected by give way signs and lines facing the earlier crossing vehicles. I still believe that parallel flow is hard to link in both directions and so needs more storage for little benefit but I have so much work to do on CFI's that I must put that aside. Have you seen the Utah document CFI Guidelines? I have some issues with it but its 3 pie diagram for 2, 3 or 4 phase intersections is basic to intersection design where capacity is critical (all urban sites). Cycle time should be optimized to reduce delay. Regards Cleeland John (talk) 02:47, 7 July 2015 (UTC)
 * To explain what a CFI is, I heard translated to english and seen fron the Australian point of view expecting lefthand traffic: "CFI is giving the right turning traffic the habits of the left turns". Now give a look on the 2, 3 or 4 phase signals:

Where the displaced turns are placed is a question of demand. When increasing the lanes lenght, its efficiency will drop the longer it is cause vehicles are speeding up. -- Hans Haase (有问题吗) 07:13, 7 July 2015 (UTC)

The PFI
I think that a new article should be added that details the Parallel Flow Intersection. It is like this one, but driver expectancy is higher due to the traditional turn pockets. Diagram here: http://www.fhwa.dot.gov/publications/research/safety/09060/images/09060_img_166.jpg as you can see, it moves the turn lanes and turns them into kind of like slip lanes. I don't have the time or the patience to make an article myself. 209.112.137.248 (talk) 22:53, 6 October 2012 (UTC)
 * Done. -- Hans Haase (有问题吗) 08:07, 4 August 2015 (UTC)

Use of coordinates
The coordinates moved inline by Hans Haase make it difficult to read the prose. As far as I can tell, most coordinates are added to Infoboxes. They seem very helpful positioned here, but is there some other way to implement them? Thanks. --Chaswmsday (talk) 14:22, 3 February 2013 (UTC)
 * Thx, move the °N, °W to better place or use the ##, refer the help on the coord template for alternative printout. I think the coordinates should be visible for offline use of the article. It would not fit the screen. Better now? --Hans Haase (talk) 20:27, 3 February 2013 (UTC)
 * From what I can tell, the inline use of °N, °W or ("short hand coordinates") is controversial (as is everything in Wikipedia!) Personally, I think the addition of inline coordinates is quite beneficial to the article, but I still can't get past the way it, IMO, diminishes readability. The icons wouldn't line up in as pretty a fashion, but I'm going to try moving them after each entry. Please tell me what you think. --Chaswmsday (talk) 22:05, 5 February 2013 (UTC)
 * You are right. I have seen you already changed it. Looks good now and provides exact references. Well done! --Hans Haase (talk) 19:06, 21 February 2013 (UTC)
 * Thanks, Hans! --Chaswmsday (talk) 00:22, 23 February 2013 (UTC)

Areal photo of CFI
FYI, on UDOT's Flickr page there are a few areal photos of a CFI, is one. These are licensed CC-NC-ND, meaning they can't be uploaded to commons, but could be used here under a claim of fair use.Dave (talk)

CFI / Continuous Green T hybrid in Accokeek, Maryland
Copied from Talk:3-way junction:

''... CFI is in a 3way-IS in Accokeek, MD. See the map! This has also the continous-green north to southbound. There's no XING PED. CFI and High-T increase the total VPH of the IS. In the DE-WP article of the CFI, I discovered it completely (in 4way cfi, 2 displaced left turn lanes, conficting way groups are from 4 to 3, (see most of the exsamples) in a 4way cfi with all 4 displaced left turn lanes coflicts are form 4 to 2, in a 3way cfi, only 1 displcd ltls reduces conflicts from 3 to 2). Afaik, this are most efficient methodes to increase VPH saving costs and space. All kinds of intersections should be compared. A full grade separated interchange provides free flowing traffic. The only bottleneck is waving traffic. Cloverleafs have it inside, Stack interchanges have the weaving before and behind the intersections, spread out over longer way. Here's an other one, using just 2 conflicting directions:'' ''49.463815,8.505955 3-way superstreet or at-grade trumpet ? In Mannheim, DE, EU'' 49.488973,8.512526 a superstreet in Mannheim, DE, EU (4way) --Hans Haase (talk) 11:10, 30 August 2013 (UTC)

End copy. --Chaswmsday (talk) 19:55, 9 September 2013 (UTC)


 * Hans, from the map, the Accokeek, Maryland CFI also has the same free-flow characteristics of a continuous green T/seagull intersection. If you don't object, I would propose mentioning this intersection in both the CFI & Seagull articles. --Chaswmsday (talk) 20:00, 9 September 2013 (UTC)


 * Yes, traffic light are green for it. --Hans Haase (talk) 17:47, 10 September 2013 (UTC)

False interpreatation of double quadrant roadway
This edit declared the signal animatet sketch as a "double quadrant roadway". No, the double quadrant roadway is signalized between the intersections. CFIs on 4-arm interections may have 2 or 4 displaced left turns (DLT), See Bangerter Highway (UT-154). --Hans Haase (talk) 03:08, 5 June 2014 (UTC)

Parallel flow intersection
Are there good sources describing this in more "lay" terms? I find the patent application hard to follow. Just looking at some examples, the CFI seems to cross early, run parallel to opposing lanes, then turns. The PFI on the other hand seems more like a regular left turn lane, crosses opposing traffic closer to the intersection itself then runs parallel and opposite to the destination road, before merging with the destination road at a traffic signal. Is there a clearer description of PFIs? --Chaswmsday (talk) 10:50, 28 December 2014 (UTC)
 * There's one known intersection. Yes, AIIR refers it. -- Hans Haase (有问题吗) 14:28, 30 December 2014 (UTC)

Video is just plain useless
The video shows someone going straight through an intersection after briefly stopping at a light and turning the camera left and right quickly as if at a stop sign. If you're going to go to the intersection with the intent of filming how it works....how about actually going through the part of the intersection that applies to the article? — Preceding unsigned comment added by Bpf007 (talk • contribs) 16:07, 17 March 2017 (UTC)


 * I also felt this way when viewing the video. David Spector (talk) 23:40, 26 January 2020 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 4 external links on Continuous-flow intersection. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20070930055756/http://www.urbanstreet.info/1st_symp_proceedings/Ec019_e3.pdf to http://www.urbanstreet.info/1st_symp_proceedings/Ec019_e3.pdf
 * Added archive https://web.archive.org/web/20070618062858/http://www.udot.utah.gov/cfi/ to http://www.udot.utah.gov/cfi/
 * Added archive https://web.archive.org/web/20090102062830/https://www.dot.state.oh.us/districts/D07/Projects/Pages/AustinPike.aspx to http://www.dot.state.oh.us/districts/D07/Projects/Pages/AustinPike.aspx
 * Added archive https://web.archive.org/web/20111029002140/http://i15core.utah.gov/flash/interchange_diagrams/universityparkway/ to http://i15core.utah.gov/flash/interchange_diagrams/universityparkway/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 16:20, 12 August 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on Continuous-flow intersection. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added tag to http://www.stltoday.com/stltoday/news/stories.nsf/stlouiscitycounty/story/98FA052BF09ED6468625738500198B0C?OpenDocument
 * Added archive https://web.archive.org/web/20100912145935/http://www.ci.loveland.co.us/publicworks/Streets/CapProjSupport/NewMadisonEisenhower.htm to http://www.ci.loveland.co.us/publicworks/Streets/CapProjSupport/NewMadisonEisenhower.htm
 * Added archive https://web.archive.org/web/20110720090118/http://attap.umd.edu/UAID_ags.php?UAIDType=5&iFeature=2 to http://attap.umd.edu/UAID_ags.php?UAIDType=5&iFeature=2
 * Added archive https://web.archive.org/web/20131230022522/http://attap.umd.edu/bbs/data/publications/Seonyeong_IEEE08_1.pdf to http://attap.umd.edu/bbs/data/publications/Seonyeong_IEEE08_1.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 04:39, 17 September 2017 (UTC)

Video no longer works
The video that's supposed to show a person going through an intersection is not playing. I've tried it on a couple browsers. Can anyone else confirm? Golfulus Shampoo (talk) 18:53, 22 August 2019 (UTC)
 * I'm guessing that the video was removed. David Spector (talk) 23:41, 26 January 2020 (UTC)