This question has come up yet again…Is it true that people can exit more quickly through a pair of doors with a mullion than through a pair without a mullion? I have been hearing about this since my early days in the hardware industry, and apparently, many of you have heard it too. Dozens of people have contacted me over the years to ask where to find data corroborating this, and many mentioned that they heard it in “hardware school.” Despite my exhaustive attempts to find the original source of the information, I have not located it.
I’m not sure why it is so important to have concrete evidence of egress movement through a pair of doors, as the model codes don’t really differentiate. If I encountered an architect who didn’t want me to specify a removable mullion, I might try to use the egress theory as support for a mullion, along with the benefits of using rim panic hardware with a removable mullion – increased security, decreased maintenance. But otherwise I’m not sure what the compelling need for the data is…it’s not like adding a mullion will allow the number of exits to be reduced. Maybe everyone continues to search for it because it’s seemingly impossible to find.
I know that some of you use software for egress modeling – I even found a professor who had run this particular test before and said that it was indeed true that you can move people through an opening with a mullion more quickly. But he had not officially documented the data or saved a video of the test that was run using their software. I’d love to end this quest once and for all…is there anyone out there who has the ability (or knows someone who has the ability) to run a comparison using software and save it in a format that I could share it? Video would be ideal.
This is an interesting video about egress when there is an obstacle in front of the door, but I’m not sure what the impact would be if the “obstacle” (the mullion) was right in the door opening. I’d also be interested to see if having the doors NOT held open would have an impact.
You need to login or register to bookmark/favorite this content.
I believe and was taught that there are 2 vastly different scenarios that affect the outcome of this “Urban Legend”.
1. under a mass exit, as at the end of a sporting event or any large gathering, a mullion allows people to que up easier and exit faster because it allows people to merge into one or the other door.
2. under an emergency exit where the rules of etiquette are generally abandoned, no mullion allows people to exit faster as there is less of a bottle neck as everyone is fighting to exit first.
I don’t believe there is any hard evidence one way or the other but that is the way it was explained to me. If you do find someone to run this experiment it should be run under both scenarios. I’d be very interested in the results.
I too have heard this tale for many years and the only proof I have been able to come up with is on this national geographic experiment that blocks an exit. Somewhat similar except on a bigger scale. View this: http://channel.nationalgeographic.com/crowd-control/videos/emergency-exit-experiment/
This December, when your son/daughter/grand child has a holiday performance at the local high school auditorium, pay attention to the stream of people going out the main entry that has a pair of doors. You’ll notice that everyone is going through the center of the pair, holding both leafs partially open for those exiting behind them. If there was a mullion between the door leafs, there would be two lines of people exiting through each leaf separately. I can say, anecdotally, that would be faster than the typical scenario.
When I get to the doors in that situation, I open one of the leafs all the way, and hopefully it has hold open ability.
I have run across this with various architects. Most Architects end up telling me that they don’t believe that a mullion is better for egress.
It’s possible that a group of people might treat a double-door opening with no mullion as if it were one single door opening, slowing egress through that opening, whereas the same group of people might treat a double-door opening with a mullion as if it were two single door openings, increasing traffic through that opening.
Yes, that’s what I think too, Chuck.
I have no data, but it seems like having the mullion delineates that there are indeed two lanes, and therefore someone egressing is not rude for going out the other leaf. Watch people at a double door without a mullion – because there is one opening, people will be hesitant to go thru at the same time as someone else, especially in non-emergency situations or in two-way traffic. A mullion tells you that you will not hit the other person’s shoulder because both shoulders of both individuals will have to pass the mullion on the same side as their body. Without a mullion, the possibility of collision is greater.
I do think that two lanes are preferable to one. Hopefully we can find some evidence.
Thanks for posting. There is good reason for single doors over pairs of doors and that is that single doors perform better than pairs of doors. They are more reliable and less prone to maintenance. I don’t like movable mullions, but i use them as a last resort before dealing with CVR devices. And clearly architects do not like mullions for aesthetic reasons. Other than aesthetics, architects will commonly use the loading and unloading and fire evacuation reasons as why pairs of doors are preferred. The first argument is bogus because you don’t need every door in the place to load and unload; also, that is what a loading dock is for!. I ended up using Pathfinder to model a room 100×150 containing 200 people; and while i want to hold on the final results so far my preliminary sims indicate there is little or no difference. I will post the results on my web site when I finish and get the results. Here are the simulations i intend to run (i am still learning the software application).
Simulation 1: Room with single 32″ exit
Simulation 2: Room with a single 36″ exit
Simulation 3: Room with a single 72″ exit
Simulation 4: Room with two 36″ single exit doors separated by a mullion
Simulation 5: Room with Two 36″ single exit doors separated by a fixed lite 1′ in length.
I know there are differences in exit times depending upon occupant makeup, locations of doors, speed of discharge, etc. Those are all variables that may change the overall exit time. I keep all those constant and the only variable I change is the exit configuration. Assuming all goes well, i should have the results this weekend.
Any other simulations you would want me to run. As long as they are reasonably simple I can add them to the list.
Thanks for looking into this, Jim! I wonder if the results would change with a larger group of people?
Oh, one other thing. The third argument used for door pairs is that they allow better accommodation width for handicapped persons. That too is bogus (unless you are talking hospitals and other patient care facilities). I used to be a member of the ADA committee for a large university as was told more than once by the experts, that a single door is in-fact easier to negotiate. Moreover, if we put an operator on the door pair, it is usually only on one leaf.
Why would a comparison using software for egress modeling be helpful? The issue with any software tool to approximate real world conditions is that the real world conditions must be known, or assumed, and then input into the software (either through parameters that can be manipulated by the end user, or written into the code by the developer). So even if you could see that a software package supported, or debunked the claim, where did that come from? It would either be based on a real world test that supports that pattern, or based on an assumption of that pattern. If the former, where is the test? If the latter, aren’t we back where we began?
I was hoping that the software would incorporate whatever data there may be about how people behave when they’re faced with one paired opening vs. 2 single openings (or a pair with a mullion). And I thought I’d have better luck asking for help using a simulation model, than inviting several hundred iDH readers to come over for a test run. I’d gladly take either one.
I say a video regarding this test in the past 12 month. I am not sure where I say it. I was very interesting. I think they did it on a college campus with people standing around in a gym and then they issued an emergency command for all to exit the building and they timed it. They did it both was with and without a mullion. The mullion won.
Think, Fred! Where did you see it??
For what it is worth I heard that egress width came from studies during WWII done with military in ships, think of aircraft exiting drill done at the factory, everyone knowing what to do and then have done it several times, not very real world.
I think for practical proposes 2 people wide on a 44 inch stair going the same direction is possible holding on to the hand rail, 2 people wide through a 36 in door doubtful, 3 people wide through a 72 in opening without a mullion possible, will people do it or will they gravitate toward the door to hold them open?
Should be an interesting discussion.
I have never heard the legend, and it doesn’t make sense. People flowing through an exit are just like any other particles passing through a restricted opening: The smaller the opening, the more likely it is that particles will jam.
Hi Michael –
I think the theory is that people do behave in a different manner from other particles, but it’s interesting that when the Massachusetts State Building Code was changed after the Station Nightclub fire in Rhode Island, it requires the entrance to a nightclub to be 72 inches wide without a mullion.
You might contact Jake Paul’s to see if he has studies
I would say true mullion equals more people through the doors
I did not hear about this phenomenon in hardware school. Instead, several years ago, I read an article about a study of pedestrian flow thru a doorway. The article contained a figure showing a column placed in a doorway and how its presence improved traffic flow. As I recall, without the column, people didn’t queue well. But having the pole (mullion) made people queue better and pass through the doorway quicker. I cannot find that study, though I’m sure I “bookmarked” it at the time. I performed a new search and found this 2009 article in Scientific American: https://www.scientificamerican.com/article/obstacle-exit-pedestrian/
Here is another article: https://www.insidescience.org/news/quick-exit-just-block-fire-door
Here is an article with video: http://nautil.us/issue/13/symmetry/want-to-get-out-alive-follow-the-ants
And here is an ABC News article: http://abcnews.go.com/Technology/story?id=8384464
BTW, I found all of the above using this search phrase in Google: study of obstruction in exit
Interestingly, I searched the hyperlink for the professor referenced in the article, Andreas Schadschneider, and I found this URL: http://www.thp.uni-koeln.de/~as/as_engl.html
One of the hyperlinked pages is this study: http://www.springer.com/us/book/9783642045035
There are companies that place cameras overhead at ceiling level and utilize the cameras and tracking software to track people (actually, the cameras are IR thermal and track warm heads) as they enter a store, thus tracking pedestrian movement through any given store. The data from this tracking is then utilized to better place displays. I always thought an interesting study – one that could influence fire protection engineering – would be to have one of these firms team up with a graduate fire protection engineer to actually map and study the dynamics of people passing through a doorway with and without a mullion. I think a study like this could influence the codes, leading to better evacuation methods.
Maybe you can use this research to revise future building codes!
Thanks for all the info David!
I heard this from one of the now retired teacher at a recent FDAI class. He cites it’s a psycological thing where people automatically go thru the narrow opening much like a horse with blind flaps as to only see straight ahead.
I deal with a lot of doors at a college where movable mullions are used to have large deliveries come thru the opening.
Let me know what the official citation on this .
Not sure if there is anything in this or if NIST might have something??
My thought would be that having the mullion in would make more sense. Ever try to drive on a road when they haven’t painted the stripes on it yet? Most wouldn’t end up where the lane would be. Also people would be more likely to move faster through the opening knowing they are not going to get smacked by the door on the other side. It would almost turn the paired opening into more of a single. Especially in an emergency situation I feel most people would head straight up the middle.
I found the video I had seen. It was on the National Geographic Channel. Its call Emergency Exit Experiment.
Very interesting studies may be time for the codes to embrace some science.
The thing I find maddening is I try and exit somewhere when faced with double doors is the tendency for people to follow the person in front of them which leads to everyone exiting through a single door, while the 2nd door of the set remains unused.
The other thing that is maddening is the amount of times people exit a space and then stopped to decide what they’re going to do next choking the exit pathway to the rest of the facility
There’s also a psychological feeling that without a mullion a typical person would exit through the center of both doors causing less people per second exiting the door. With a mullion, people will exit through each door thus maximizing the space of use in a mass crowd exiting in an emergency situation. The removable mullion is there so that if there was a fire or a medical emergency, rescue workers can remove it to bring in or take out wide objects ie: fire hoses, smoking furniture, a person stuck in a vending machine, etc.