The Fill and Flush deplaning method
fillandflush.com
fillandflush.com
- Will people even obey? What if half the people from the wrong side get up too? After a long tiring flight, will that anger people on the correct side and lead to fights? Having to show your ticket to prove you were in the right column to get off is not very customer-friendly
- What about families with children, couples, and groups in general? People want to stay together. It doesn't matter if this is rational or not, it's what people want to feel safe and not anxious
- Is it really even going to make much of a difference? In my experience, people are already mostly grabbing their bags quickly, simultaneously while the rows immediately ahead are starting to exit
And of course, while this can in theory save perhaps a couple minutes for domestic passengers with only a carry-on, even in theory it saves no time at all for anyone who still has to go through a crowded passport control or wait for checked baggage.
I think it would help with getting onto the plane. Filling the plane can easily take 20 minutes - which in a half hour turnaround is often the critical path.
The challenge here is you need to get people to enter the plane door in a very precise order. Just 10% of people ignoring instructions probably eliminates most/all of the benefits. Groups who don't want to sit apart or board the plane apart have to go either first or last.
also have a better chance of not having to risk getting separated from your carry-on baggage because the plane's already full
Usually I find I can get a seat I don't hate unless I am running so late that I am in the latter half of the 'B' group or anywhere in the 'C' group.
One time I was traveling with a blind girl who had a guide dog and the boomer patrol had occupied all the bulkhead space. They seemed to think she should sit with a German Shepherd in her lap. Needless to say I aggressively advocated and reminded the flight crew that FAA regulations entitle persons with guide dogs to a bulkhead seat. Let me tell you, we got some stink eyes from the displaced boomers who didn't want to have to walk an extra six feet to the potty.
Google didn't find those regulations for me.
[1] https://upgradedpoints.com/travel/airlines/ultimate-guide-to...
can you really reduce it that much, given that the plane needs to be refueled, cargo unloaded and reloaded, etc? like it may be the case that while it's inconvenient for passengers, the airline doesn't really care because it takes them a half hour to turn the plane around on the tarmac side anyway, so there's no benefit from reducing the passenger loading time.
fuel and luggage crews can be quick. I've seen a luggage crew unload and reload a plane inside 8 minutes, including loading 2 airfreight containers.
As with de-boarding most time is wasted/spent by people standing in the aisle and waiting for someone who is putting their luggage in. The optimal solution has as many passengers loading their luggage in the overhead compartment in parallel as possible. The ideal way to do this would be that we first have passengers with window seats board with one person for each row. We should only use every other row to ensure people can maneuver. So we'd first board seats NA, N-2A, N-4A, etc. down to row 1 with N being the last row. Then NF, N-2F, N-4F etc. And then N-1A, N-3A and so on. Then we work our way through the middle seats and then aisle seats.
The current approach which commonly boards people starting from the front or back FORCES congestion and minimizes probability of two people reaching their seat at the same time and being able to put up their luggage in parallel.
Even if the approach I describe is broken because groups board together, it becomes at worst a random boarding order which is still much more efficient than the current nonsense.
The reason we cannot board more efficiently is because it's unintuitive, people don't like it and behave like wild animals in groups.
I think it can quickly become worse than random.
Imagine you have the perfect sequence for max parallelism, but right in the middle is one 6 person family group who are going to take 10 minutes to get seated while they arrange toys and stuff for each child, blocking the aisle the whole time.
With the perfect sequence, this halts all boarding for those 10 mins, since people need to pass the blockage for the scheme to work. Whereas with random boarding, there will actually be many people who manage to get seated during those 10 mins, because there will be a bunch of people who, by chance, are seated between the door and the blockage and can get to their seats.
such a group is slower, certainly.
they do not take 10 minutes.
It only roughly approximates though because the first few groups are usually not outside-in (eg. first class, premium/rewards customers, veterans, families with small children).
I guess it's just that people would feel bad going *last*? Psychology is weird - but powerful.
That really changes the dynamics.
Small amounts of money are often very motivating.
You could, for example, say that anyone can enter or leave the plane in any order they like. However, if they do so before their turn in the system, they don't get a $10 refund on their ticket. If they don't present their ticket, they are assumed to have gone early.
The question then, is whether the benefit would remain with proportional instead of total compliance, and whether the benefit outweighs the costs of enforcement.
There's this 2015 article on doing essentially what he/she is suggesting: https://www.vox.com/2014/7/8/5877863/it-takes-forever-to-get...
There's a paper on solving the problem by assigning passengers to seats so that their luggage is evenly spread throughout the overhead bins: https://www.sciencedirect.com/science/article/pii/S096969971...
There's a paper here on other methods tried, including creating "pre boarding" areas for passengers: https://www.researchgate.net/publication/269466040_Speeding_...
The first step in any scientific endeavor should always be a literature search.. see what's been tried, and what worked and what didn't.
One solution might be to have lights that light up when it's time for your column to deplane.
A more evil solution would be to use physical barriers that only open when it's a given column's turn. Maybe replace the seatbelts with rollercoaster-style restraints for extra fun.
> What about families with children, couples, and groups in general?
Have them deplane either first or last. The page seems to suggest that waiting for the next "wave" ain't a huge deal, so having groups wait until their last column is ready to go (or worst case, until all individuals have deplaned) could probably work.
> Is it really even going to make much of a difference?
Maybe not in isolation, but maybe the aggregate effects (e.g. more flights per day) would justify it?
Most people flying are not traveling alone. It's not just the oft quoted "families with small children". What about couples/friends/colleagues. People traveling together largely sit in the same rows, not the same columns, and they want to exit together.
These kind of analyses (especially the "if everyone saved 13 seconds we'd size the equivalent of 20 lives!" part) always make me a bit said because, in a lot of cases, it's clear the author understands math but not humans.
I wonder how rapidly this column wise deplaning breaks down in the face of random compliance and confusion.
I implore you to consider the group issue in more detail, and perhaps with a touch of human empathy this time.
This really says everything you need to know about this "analysis".
The notion that ALL sets of 2+ travelers now have to wait at the end of the gangway just inside the terminal, the disaster that would create at the gate, is just kicking the can off the plane in a way that doesn't actually help the next passengers get on any sooner. The gate -- and all gates -- would be a disaster.
Another commenter put it brilliantly: "My perpetual motion machine works brilliantly, just have that 'unresolved issue' of the second law of thermodynamics".
And more obviously, the level of coordination required to make it work is just not worth it. Families flying together, babies, people with disabilities, etc., none are considered in the model.
-Groups can wait for later "Flush" es - there is flexibility since I think you can fit a little more than one person in the aisle per row.
Domestic flights on single aisle aircrafts, have several classes (first, extra, etc.), with different price ranges. Even emergency exit seats come with some attached benefits, like priority boarding. Good luck telling customers that they paid extra for an emergency window seat, but have to wait until more than half of the plane have de-boarded to exit.
That's exactly what already happens (at least in Europe), you don't get priority to de-board if you bought seats with extra leg room on the emergency exits, you have to wait.
I haven't seen any of them fined 25k. Proportionality matters, and getting off the plane as fast as (in your mind) possible isn't going to be realistically enforced.
This also reminds me of an article I once read in an in-flight magazine about this. It was saying how an airplane’s brakes work really really well, not only at 350km/h when landing, but also at 50km/h when taxiing to gate. It described an incident where during taxiing to gate after landing, somehow some animal managed to jump right in front of the plane forcing the pilot to apply max break, causing a passenger who was handling the overhead bin to die from the injuries.
Actually, just found a picture I took of the article, so here’s the original text. I always think of it when people stand up before the seatbelt sign is off.
> … But ignoring these rules can have serious conseances. For instance: Do you really remain seated until the plane comes to a complete standstill? German travel journalist Heige Timmerberg likes to tell the story of the dog that ran across the runway in Casablanca, forcing the pilot to slam on the brakes. Brakes that work well at 350 km/h, work well at 50, too - and one of the passengers who had just emptied the overhead bin "flew from row 26 to the lavatory behind the cockpit, and was dead." > "We don't joke about security," says Fabienne Regilt, member of the lufthansa team.
Now you have to play with the airport itself. Most wouldn't have enough bridges to lend two per gate.
This is definitely not usual.
Admittedly, I haven't been to all airports across the US or Europe, but I have been to quite a few, and have yet to see people boarding from the front and back. Closest perhaps was a double bridge used on an Emirates A380.
Last time I flew to Berlin with EasyJet it also boarded from both doors, now that I'm listing it I can only remember a flight from Oslo-Stockholm with Norwegian where the boarding was done only through the front door. Most of the flights I've taken the past 2 years have boarded from both entrances, I feel it's quite usual.
In the US, I would say that it is not usual at all, or I have been to the wrong airports (IAD, DFW, AUS, IAH, SFO, LAX, JFK, LGA, CLT, ORD). I haven't seen this either in Asia.
Perhaps this is highly airport dependent? Still, I would think that this is far from common.
I'm very surprised that so many people are telling me this is uncommon, I've been through this setup countless times.
I used to fly much, much more but I do not remember that far in the past.
Loading and unloading from both ends is a lot faster though. United actually had a double ended gate that they were trying out years ago in Denver, which is the best of both worlds and saves a ton of time loading and unloading. It's this really long cantilever thing over the wing to reach the rear entrance. Unfortunately something broke on it one time and the jet bridge fell onto the wing, damaging the airplane, and they gave up on it.
As others have pointed out, this method works great in many other countries.
This is proven out already in other countries, the ones that want to leave through the jetway stay seated and still get out faster because even if they are near the back most people around them leave through the rear so they make it out the forward jetway much quicker.
Some airports will revert to passenger stairways when there's more planes to put passengers on, and the weather permits. Airports that often do that, may also sometimes use passenger stairways to accelerate deboarding, but passengers aren't very good at doing that; and when overhead bin storage is tight, you may have placed your items in a bin too far forward to retrieve it while passenger motion is towards aft stairs. I can't recall seeing boarding from the front and aft simultaneously.
Very very occasionally, I've seen or heard of airports set up to attach multiple jetbridges to a single aircraft, but that's going to be pretty complex geometry because distance between the front door and the aft door on different planes varies widely.
> How to accommodate the elderly, people with children, and groups.
Guys, you have to read the article at least. It's stated in the "further discussion" section so at least add something more than just repeating the topic.
Or, I suppose, rewritten in the vernacular of this site:
"This comment is obviously written by someone who didn't read the proposal properly".
Coming up with a better proposal than this is trivial. Here goes.
During flight, lock the overhead bins. When deplaneing, those without overhead luggage go first, from first to last row as now. This makes sense because they generally walk faster because they have less stuff. Those with overhead luggage have to wait for the steward or stewardess to unlock the overhead bins following the fast group, and then exit from last row to first. Therefore those who are slow to get things out of bins, do not slow others. Those needing assistance wait until the very end, as now.
Getting stuff out of bins is the gating factor here. And so there is no passenger conflict about the fast people leaving first, because the others are waiting on the bins being unlocked, then for people behind them to pass. And groups get to leave together.
There you go. A faster approach than now. With fewer problems.
I do not believe locking mechanisms in the overhead compartments are currently installed.
Controls to instantly lock/unlock such systems remotely by one attendant, thousands of times, reliably, is yet another challenge. Better to engineer a solution that requires no new equipment.
Engineering a solution that creates passenger conflict is not a good idea. I'd rather wait for slow people, than be around a loud argument at the end of a flight. I get enough stress already from hours of being cramped on a flight, I don't want more.
Also you are adding entirely unnecessary requirements to my solution. There is no need for the unlocking to be instant. Just to be doable one by one by someone going around the pace of people dragging luggage. The technology used for security tags in clothing stores would be more than adequate.
That said, I'm not seriously advocating for my solution. I'm just saying that it takes no more than 30 seconds to come up with something better than the one the blog post was about.
I've never had a problem going to the bathroom. People get up, sit down. And there is plenty of time to do this while waiting for the line to start moving.
Also, remember that this is not a proposal I'm seriously arguing for. Merely showing how easy it is to come up with a better proposal than the one made.
> Second Law Of Thermodynamics
Other than a few details to work out, my perpetual motion machine is going to revolutionize aviation.
It also means that families who are travelling together would need to deplane separately, which can screw up things in terms of helping each other with bags, and also wouldn't work with families who have multiple kids of an age where they need to walk with their parents. It feels like the kind of fragile system where you'd end up with so many exceptions that it becomes potentially even more messy as you'd need to communicate those exceptions and people would need to understand them.
Perhaps the procedure should be, if everyone in your group is holding all their belongings - either in the aisle or standing in the footspace of an aisle seat (so rows C and D in a 3-3 configuration), then they can join the deplaning wave. But if you're standing in the aisle but one from your group is still crouched in a window seat and will need to grab their bag, and you don't want to abandon them, then you don't occupy the aisle go back to your (presumably aisle-) seat and wait for the next wave/go into an empty row so the crouched passenger can stand in row C/D, and wait for the wave to finish so they can grab their bag.
> How to educate passengers
The author vastly overestimates peoples willingness to do literally anything that benefits the group rather than themselves.
It does not matter if every single person goes through a six week course on why this system is better: at least 20% will still ignore it to get themselves off the plane sooner. "I have a connection! I'm in a hurry! My Mom is waiting for me!".
You want a better system? Design one that works best when everyone is greedy and maximizing their own benefit.
https://www.youtube.com/watch?v=oAHbLRjF0vo
This is basically https://arxiv.org/abs/0802.0733 but differs to the paper in that it doesn't consider: alternating odd-and-even row disembarkment, which would give passengers more aisle space to potentially speed-up luggage handling.
Suffice to say, this is not my experience of boarding and leaving aircraft.
https://github.com/josephecombs/fill-flush/blob/fc0fa73b5879...
Ryanair even seems to have a way to communicate this. "cabin secure" means everyone is following the rules. "cabin secure, kinda" means we're gonna break the rules and hope everyone sits down before takeoff.
In that model, would it faster or slower for someone in the middle of end of the plane?
There are 2 reasons people use carry-ons rather than checked bags. First, because its cheaper for people w/o status, and second because they're worried about the airline loosing their luggage, and/or they don't want to wait for their bags.
To negate the first reason for carry-ons using a carrot + stick approach, I'd offer 2 free checked bags to everyone regardless of status, along with charging $100 or more per carry-on.
The second reason might be harder. The airlines would need to re-negotiate deals with their credit card company partners (suddenly the free checked back with the Delta Amex plat. card doesn't mean anything) and change their frequent flier programs, and they'd need to get better at bag handling in general.
I wish there was an airline simulator game.
There are probably solutions to the problem of it taking too long for people to pull their bags from the overhead bins.
Also you cannot exit with the people you sit next too (like your toddler who will have to make their own way off?).
And that's assuming anyone actually wants to minimise average time. Airlines don't. Neither do most passengers.
If you have ever taken a commercial flight, you have seen it: the individual that needs two people to team up (or one highly capable one) and lift up or down their 40-50 lbs (~20 kg) roller from the overhead. Half the time, their help is in their party, but it slows things down.
I would estimate that this happens at least once every 4 rows on a fully occupied 6 seat wide aircraft.
Given this, forbidding passengers to bring carry ons aboard that they can't overhead lift themselves would be the only way to speed things up regardless of boarding/deboarding strategy.
Is that practical? Probably not.
After a return flight, I'm usually at home by the time my bags would start coming out if I checked them.
An obvious easy improvement would be to move the doors to just above or before the wings, so you could drain from two sides at once.
First class, golden carpet diamond supreme customers won't tolerate waiting for someone from the back of the plane to exit. Their head would explode probably and they'll vow to never fly that airline.
But overall the problem is that people fly in groups and they like to sit together. Sending the kid out while the parent waits for the next column to deplane will never happen. Should the people in groups wait then, but a good part of the plane will just stay waiting and it's back to the existing algorithm anyway.
I would much rather if all planes had a second staircase at the back, jet bridges aren't needed. That, or let people check luggage for free again so they don't need to manhandle carry-ons.
My guess is that airports also prefer a slower release of passengers -- it means customs lines and bathrooms aren't immediately saturated.
I've never seen deplaning be the hold up, it's boarding that takes forever. But getting people to follow complicated high-speed boarding plans is like herding cats.
You could even up-charge for above head storage, but only allow seats at the tail of the plane to choose that upgrade. They board first, and they deplane last, ideally holding up no one.
This sounds like hell but hey… if it saves a few bucks a seat on a really short flight because they can deplane in 5 minutes… I’d be for it.
It's easy to improve over a fake strawman
It seems fairly realistic to me. People in rows toward the front step into the aisle as soon as possible to gather their carry-ons. This blocks everyone rearward that already has their belongings and is in the aisle from deplaning.
Its really not that realistic.
But I also think they should put healthy 25 year olds in the emergency exit row for… emergencies, but they’d rather sell those seats to someone that’s often the poorest candidate to execute in an emergency.
I also end up near the back of the plane all the time, because those are the available seats that don't have a higher price tag.
The delay is just a waste, but unfortunately the airlines aren't incentivized to make boarding or deplaning quick, they're trying to upcharge for more expensive seats. All the engineers are sitting at the gate for 15 minutes wondering why everyone's in a rush and paying extra to get "priority boarding" onto a cramped plane, then at the end of the flight, they're sitting in the back of the plane wondering why it takes 15 minutes for everyone to get off the plane, and how this wasteful process could be made more efficient.
Now simulate it with humans who really want to get off that flying toilet that has been their home for 12 hours. No amount of education will suffice.
Sorry for being negative, I like that you came up with something clever. I've seen some similar attempts to make boarding more efficient. I just fly way too frequently and I think the world is not ready for this yet.
Stop nickel-and-diming for luggage and focus on efficiency gains like gate-checked small luggage. Add front and back exits to every gate ramp to improve boarding and unboarding.
That's by far the easiest issue to handle: don't change a thing... they get off first exactly how they do now.
However efficient loading doesn't make passengers happy.
Priority boarding is worth an extra fee from some passengers and adds several minutes to boarding time.
https://youtu.be/oAHbLRjF0vo for a bit more about various options and the reason it is all moot.
Unfortunately, having flown many times, I just don't see humans ever being able to self organize to this degree, especially at the end of a flight!
If someone in row 3 finishes their bags, the other people in row 3 don’t always hold up the line for the rest of the plane.
- some people prefer to sit and wait
- some people let others through before blocking the aisle
Anyway, I think “let some people through before you get your bags” is also way easier for people to understand.
Note that this is from a boarding, not deplaning viewpoint, but the concepts are similar.
If they care about it then allocating seats closer to exit for people with connected flights would make more difference.
Easier to herd cats.
source code at: www.github.com/josephecombs/fill-flush
Field studies watching actual passengers loading actual planes (as opposed to computer simulations of theoretical "people" loading theoretical planes) frequently show the fastest loading when passengers are lined up randomly, ahead of any of the more seemingly "intelligent" approaches that work great in computer simulations. If the "best practice" algorithms can't even beat the random algorithm in a field study, that's a sign this is a much harder and much messier problem than it appears to be.
Just about any google search, like "airplane passenger loading research", will return more than a dozen theoretical and field studies on the first page of hits, including:
[0]The Role of Computer Simulation in Reducing Airplane Turn Time, https://www.boeing.com/commercial/aeromagazine/aero_01/texto...
[1]A Simulation Study Regarding Different Aircraft Boarding Strategies, https://upcommons.upc.edu/bitstream/handle/2117/19651/145014...
[2]Analysis of Airplane Boarding Times, https://www.researchgate.net/publication/220244330_Analysis_...
[3]Efficient Aircraft Boarding Strategies, https://labs.engineering.asu.edu/ilpil/research/efficient-ai...
[4]Passengers boarding airplanes: we’re doing it wrong, https://theconversation.com/passengers-boarding-airplanes-we...
[5]The Best Way to Board a Plane, https://phys.org/news/2008-02-board-plane.html
[6]Loading an Airliner Is Rocket Science, https://www.nytimes.com/2006/11/14/business/loading-an-airli...
[7]Comparative Study of Aircraft Boarding Strategies Using Cellular Discrete Event Simulation, https://www.mdpi.com/2226-4310/4/4/57
[8]Intelligent Boarding Modelling and Evaluation: A Simulation-Based Approach, https://www.hindawi.com/journals/jat/2021/9973336/
[9]Improving airplane boarding boarding time: a review, a field study and an experiment with a new way of hand luggage stowing, https://commons.erau.edu/cgi/viewcontent.cgi?article=1200&co...
[10]A Comparison of Algorithms That Estimate the Effectiveness of Commercial Airline Boarding Strategies, https://commons.erau.edu/cgi/viewcontent.cgi?article=1072&co...
[11]The Mathematics of Aircraft Boardingg, https://www.maa.org/external_archive/devlin/devlin_05_06.htm...
(among many others)
1. ANYONE fills the aisle
2. Entire aisle is flushed (aisle obstructors are somehow severely punished)
3. repeat until done
This would probably be even faster than what I propose but is a lot harder to model.
The last 10% or so of boarders are often left with no overhead storage near their seats. The last 5% or so of boarders are often left with no overhead storage available at all.
"Put it in the first space you see" is often explicitly an instruction that the cabin crew gives to boarders as overhead space starts dwindling. Passengers aren't doing it out of preference.
people default to the "f you, I've got mine" mentality so there's no way this will work.