News for hackers. Stuff that matters.
Re:Letting slower passengers board airplane first is faster,... by Reedx on Wednesday January 15, 2020 @03:23PM (#22058073)
CGP Grey has a great video examining different boarding methods and why some are slower or faster than others: https://www.youtube.com/watch?v=oAHbLRjF0vo

Also noting that the way we deplane (front to back) is the slowest method, heh.

Re:Letting slower passengers board airplane first is faster,... by virtuous_signal on Wednesday January 15, 2020 @03:37PM (#22058223)
Great video. This is one of those phenomena, like how to fix traffic jams, that everyone has an opinion on because of firsthand experience, but few people know about the actual research on. Despite there being known fast methods, I think 2 things stand in the way of implementing them:

1) Business and 1st class make a lot of money for airlines. Not letting them board first (although personally I've always wondered why this is desirable) would devalue their tickets and make less profit for the airline.

2) Airplanes have to undergo certain maintenance checks or procedures whenever they land. Those take some fixed amount of time. There is no incentive to speed up the boarding/deplaning process past that fixed amount of time, since it won't be the limiting factor.

Re:Letting slower passengers board airplane first is faster,... by jsjohnst on Wednesday January 15, 2020 @06:12PM (#22059866)
> Not letting them board first (although personally I've always wondered why this is desirable)

As someone who customarily boards early (due to fare class or airline status), I can tell you one solid reason, overhead bin space. If flight attendants would ensure folks put their bags on their row and the gate attendants ensured bags weren’t too large, this would be less an issue, but I routinely see folks fairly early in the boarding process putting their bags in rows in front of theirs, even though their row’s bin is empty or at least not full.

The worst example of this was my flight a few weeks ago. I was the first to board that flight, followed by another customer in first class and her family. The woman proceeded to put all her kid’s and husband’s luggage in the first class overhead despite her being the only one in that cabin (her seat was via an upgrade, not that it matters). The rest of the passengers in first class (besides me, thankfully) had to put their luggage in the first rows of economy due to her actions which had a ripple effect down the plane. Probably extended boarding and deboarding time by at least five minutes, all because of what I’d call selfish behavior.

Re:Letting slower passengers board airplane first is faster,... by sdoering on Thursday January 16, 2020 @12:26AM (#22061908)
I strongly despise people like that. If you add up lost time for all other passengers and compare that to won time for her and her family (less time waiting at the baggage claim) this ends up massively weighted against her behavior. non the less her behavior is actually rational and optimal from a standpoint of game theory. At least under current conditions (bag size not checked and enforced up front). This is in my books a classic example of 'tragedy of the commons'.

As long as egoistic behavior is the optimal strategy we will not see a significant change in behavior.

It can only be changed by changing the surrounding rules and conditions.

Re:Letting slower passengers board airplane first is faster,... by senordevnyc on Thursday January 16, 2020 @09:46AM (#22064957)
So you strongly despise someone for being a rational actor within a broken system?
Re:Letting slower passengers board airplane first is faster,... by HeWhoLurksLate on Thursday January 16, 2020 @10:15AM (#22065236)
My guess is that his dislike stems from the fact that this lady only made things better for herself and her family, and not GP, or anyone else on the flight.

If everyone can't do it (bringing more luggage than will fit above you) then doing that seem like you're doing it because you're better than everyone else, which is offensive to most people for good reason. If her whole family had flown first class, that'd be a different story. (They'd probably still be annoyed at the family for basically cancelling out some of the perks of paying extra for flights, though.)

Re:Letting slower passengers board airplane first is faster,... by MauranKilom on Thursday January 16, 2020 @02:35PM (#22068020)
Society doesn't operate only on rationality... https://xkcd.com/1499/
Re:Letting slower passengers board airplane first is faster,... by billatq on Wednesday January 15, 2020 @09:22PM (#22061129)
Pretty much this. When I’m flying business I actually would rather board at the end, but I also don’t want to check my bag because it’s a hassle to retrieve and the contents are thrown several meters. If I could guarantee there was space in the overhead for my bag or if it was loaded for me, I’d board last.
Re:Letting slower passengers board airplane first is faster,... by seanmcdirmid on Thursday January 16, 2020 @12:16AM (#22061863)
Business class overhead space is often reserved for business class, even on domestic flights. That space doesn't usually fill up completely, so you have a lot more leeway in boarding.
Re:Letting slower passengers board airplane first is faster,... by jsjohnst on Thursday January 16, 2020 @07:24AM (#22063841)
> That space doesn't usually fill up completely, so you have a lot more leeway in boarding.

Maybe the airlines I fly are just the worst examples, but I frequently fly business/first both domestic and international and the overheads are always full, usually due in part to passengers not in that cabin.

Re:Letting slower passengers board airplane first is faster,... by seanmcdirmid on Thursday January 16, 2020 @12:38PM (#22066818)
What airlines do you fly? Delta is pretty aggressive about keeping economy baggage out of business even on domestics. I can’t imagine anyone even trying on an international. More to the point, I rarely have overhead problems on international flights (actually, I don’t think it has ever been a problem). Only domestic legs are problematic, mainly smaller 737s and 320s.
Re:Letting slower passengers board airplane first is faster,... by jsjohnst on Thursday January 16, 2020 @03:36PM (#22068601)
United predominantly.
Re:Letting slower passengers board airplane first is faster,... by reaperducer on Wednesday January 15, 2020 @03:56PM (#22058453)
although personally I've always wondered why this is desirable

When I was fat, I paid extra for first class tickets so that I wouldn't be the slob jerk who overflowed into my neighbor's coach class seat. Because of that, I had a lot of coach-mindset first-class seats.

Boarding first in first class gives you an opportunity to get some work done, to relax a bit before the plane takes off, and almost always you get to have a drink and a snack before the plane leaves the ground. It gives you a chance to put your head on straight for the next leg of the journey, or catch up on stuff before the ascent.

If you're a busy person (as I was back then), boarding first really makes a difference.

Re:Letting slower passengers board airplane first is faster,... by ehnto on Wednesday January 15, 2020 @04:07PM (#22058586)
How much longer do you get typically? I never have to wait that long to board as economy so I am curious if my economy boarding is atypical. I could totally see how an hour sitting in first class might be nicer than poking about a terminal. Is it ever that long?
Re:Letting slower passengers board airplane first is faster,... by kshacker on Wednesday January 15, 2020 @04:33PM (#22058918)
It is not a question of how long after business does economy board. It is a question of how much after economy starts will business board if order was reversed.

I have done some business travel and some premium economy travel, and I have seen airlines take 40-60 minutes from "formal" boarding time to takeoff. When they require 60 minutes, it usually is a big plane and the difference between business and economy could be 20 minutes easily (assuming the seating was reversed).

I have wondered many a times if boarding last would not be better? But there is a joy to leaving the lounge and directly walking into the plane if you time it right. Otherwise you end up waiting for others to board first. Also, I could afford to be late and still get in, not so if my boarding group was the last.

These are minor luxuries, but each has some value. In the bigger scheme of things, and given that it is hard to afford business travel all the time on personal dime, I would say it does not matter much :)

Re:Letting slower passengers board airplane first is faster,... by trillic on Wednesday January 15, 2020 @04:32PM (#22058911)
It really depends on the size of the airplane. Just a couple weeks ago I was one of the first 10 people to board the plane and was in "Premium Economy" on a 777-300ER for a full flight from Eastern Europe to the Midwestern US. I'd guess I was on the plane for nearly an hour before we pushed back from the gate. But this is for a plane with 300 people that boarded through a single door.
Re:Letting slower passengers board airplane first is faster,... by rootusrootus on Wednesday January 15, 2020 @08:08PM (#22060701)
That is mindblowing. Even on an A380 that was only boarding through a single door I've never experienced an hour on the ground in the plane before pushing back. That would be super annoying.
Re:Letting slower passengers board airplane first is faster,... by kortilla on Wednesday January 15, 2020 @08:27PM (#22060805)
It sounds like you just normally board well into the boarding process. The standard for many domestic flights is to start 40 mins prior to departure. International is 1hr+.
Re:Letting slower passengers board airplane first is faster,... by maccard on Thursday January 16, 2020 @04:31AM (#22062943)
In the US. I fly Ryanair fairly regularly, and European flights are fully boarded within 20 minutes often.
Re:Letting slower passengers board airplane first is faster,... by dave5104 on Wednesday January 15, 2020 @04:38PM (#22058976)
I've typically had 20-25 extra minutes before they finally close the door. It's enough to get a head start on a podcast or watch a short TV episode.
Re:Letting slower passengers board airplane first is faster,... by Invictus0 on Wednesday January 15, 2020 @05:00PM (#22059224)
There's nothing stopping you from watching that in the airport.
Re:Letting slower passengers board airplane first is faster,... by dave5104 on Wednesday January 15, 2020 @06:33PM (#22060047)
The only thing that comes to mind is worrying about listening for boarding info and making sure I don't miss it. Once my butt is in a seat, I'm no longer worried about making sure I get onto a plane at the right time.
Re:Letting slower passengers board airplane first is faster,... by s1artibartfast on Wednesday January 15, 2020 @05:55PM (#22059743)
It is still much more pleasant to do it sitting down with a drink, than breaking focus every 30 seconds to check your boarding status.
Re:Letting slower passengers board airplane first is faster,... by Invictus0 on Wednesday January 15, 2020 @06:55PM (#22060228)
I have lounge access through my credit card but even if you don't, it's way cheaper to pay the airport lounge day rate than to fly biz class.
Re:Letting slower passengers board airplane first is faster,... by scarejunba on Wednesday January 15, 2020 @09:14PM (#22061081)
No, there is one thing. Once I'm in my seat I don't have to pay attention any more. This is worth a lot because they won't know whether you're the guy who's missing. And if you're in the lounge there's no way you make it to the gate on time unless you time it yourself.
Re:Letting slower passengers board airplane first is faster,... by s1artibartfast on Wednesday January 15, 2020 @04:34PM (#22058924)
>although personally I've always wondered why this is desirable

Most people finding waiting to board unpleasant and want it over ASAP. For me personally, it is often less comfortable in the terminal for a number of reasons. The climate and noise is worse, I must keep an eye on luggage, and I have a mild anxiety about inexplicably losing my documentation or becoming too engrossed in whatever activity I am doing while I wait.

Alternatively, I could be sitting, stretching, taking off my shoes, ordering a drink, and focusing my full attention work/entertainment/sleep for that 20 minutes.

Re:Letting slower passengers board airplane first is faster,... by billatq on Wednesday January 15, 2020 @09:24PM (#22061141)
Many airports have half decent lounges. My personal preference is to hang out in the lounge until the boarding is nearly over, stroll onto the plane and sit down to go, but this tends to be impractical due to luggage.
Re:Letting slower passengers board airplane first is faster,... by davinic on Wednesday January 15, 2020 @07:30PM (#22060460)
Think of it as first class seats giving the optionality to board at any time (not to mention those free drinks and snacks).
Re:Letting slower passengers board airplane first is faster,... by adrianmonk on Wednesday January 15, 2020 @07:43PM (#22060552)
Also, because first class is smaller, you spend less time standing in the jetway and airplane aisle.

Part of the annoyance of boarding isn't just the total time it takes from when they open the gate door until when it's done. It's the amount of standing up, holding your luggage, shuffling forward a bit, and wondering when the person in front of you will move.

Use driving as an analogy. Suppose you get out of a doctor's appointment at 3pm and are meeting someone to see a movie at 6pm. During those 3 hours, the only two activities you can do are kill time on your phone (either lingering in the doctor's office lobby or sitting in the movie theater lobby) and drive. Do you leave at 3pm and avoid rush hour, or do you wait until the last minute and get in traffic? Either way, you're going to wait 3 hours, so you're not saving time, but one is less stressful.

Re:Letting slower passengers board airplane first is faster,... by virtuous_signal on Wednesday January 15, 2020 @10:56PM (#22061559)
Yes that does sound nice and I agree that the first option would feel better. However if someone offered me the choice between being able to do the first, and say, a new Macbook Pro 16", I would take the computer every time. I think even if I were a billionaire I would rather keep my few thousand dollars and suffer the indiginities of economy, but of course I'm not, nor can I know how my values would change if I were a billionaire.
Re:Letting slower passengers board airplane first is faster,... by seanmcdirmid on Thursday January 16, 2020 @12:21AM (#22061889)
If you fly a lot on long hauls, you'll quickly succumb to economy class syndrome if you spend all your time there. Once or twice a year, it is now problem, but not being able to sleep very well for 11 hours straight gets old quickly, and there is only so much content to go there in their entertainment system (if you are a frequent flyer, you've seen all of it)...not to mention your butt and legs just start getting sore after awhile.

Even though I'm not really a frequent flyer, I generally buy business class when they are on sale. E.g. Beijing to Bali on Hong Kong Express for $900 roundtrip (vs. $500 for economy), or Beijing to LA on ANA for $2000 last minute one way (vs. $1500 for economy, and since I was moving, it was tax deductible at the time). You can find some good deals, especially in Asia, and the experience can be worth it.

Re:Letting slower passengers board airplane first is faster,... by emiliobumachar on Thursday January 16, 2020 @05:40AM (#22063311)
Boarding first doesn't need to justify the full first class price difference. It's a small peel in a pricey bundle deal that includes much better perks.
Re:Letting slower passengers board airplane first is faster,... by adrianmonk on Wednesday January 15, 2020 @07:33PM (#22060480)
> Business and 1st class make a lot of money for airlines. Not letting them board first (although personally I've always wondered why this is desirable) would devalue their tickets and make less profit for the airline.

Seems like you can model this as equivalent to boarding two airplanes. Boarding first class is like boarding one airplane, and when you're done with that you start the clock ticking on boarding coach, which is like the second airplane.

In which case, you have two airplane-boarding sub-problems, and the number you care about is the sum of the times. You can apply whatever strategy to both subproblems, then you can reduce both numbers that you're adding, and you still get a smaller sum.

Maybe it's not as much of an efficiency gain as doing away with the first class and coach separation, but it should still be an efficiency gain.

Re:Letting slower passengers board airplane first is faster,... by dunmalg on Wednesday January 15, 2020 @04:51PM (#22059137)
>I've always wondered why this is desirable

Others have noted other reasons, but I think the biggest reason is that 1st/business class seats are more comfortable than the crappy crowded boarding area. Economy seats, it's the other way around. And as an added bonus, on something like a 787 where the boarding doors are behind business class, boarding first means you're lounging around in the front of the plane being served drinks and addressed by name while behind you the rabble are are being herded like cattle into their tiny seats at the back.

For 3x the price, 1st/business class want luxury, and sitting elbow to elbow with the riffraff in beat up terminal seating isn't that.

Re:Letting slower passengers board airplane first is faster,... by United857 on Wednesday January 15, 2020 @04:56PM (#22059190)
Business/1st class passengers (at least on international flights) almost always have access to lounges, rather than waiting in the gate area. As good as the plane seats are, I'd still rather spend as much lounge time as possible.
Re:Letting slower passengers board airplane first is faster,... by s1artibartfast on Wednesday January 15, 2020 @06:03PM (#22059793)
I would never consider waiting in a lounge past the start of boarding.

The risk of having to reschedule a $5K flight is to great to justify the reward. If I'm not going to be in the lounge, I'm back to minimizing the time at the gate.

Re:Letting slower passengers board airplane first is faster,... by matthewowen on Wednesday January 15, 2020 @10:26PM (#22061432)
Ok, but on domestic they generally don't.

Also, most lounges really aren't very good. AmEx centurion lounge is nice. But like, the United lounge in your average airport is really a few steps down from your typical airport bar.

Re:Letting slower passengers board airplane first is faster,... by barrkel on Wednesday January 15, 2020 @05:04PM (#22059274)
To be clear, 1st class is about 10x to 35x more expensive than economy for a transatlantic flight (I wouldn't bother with upgrading on any flight I couldn't catch some sleep on).

(BA: ~£200 Economy for LHR-JFK, £2300 to £7300 depending on day 1st Class)

Re:Letting slower passengers board airplane first is faster,... by mehrdadn on Wednesday January 15, 2020 @04:37PM (#22058958)
> personally I've always wondered why this is desirable

Maybe peace of mind? One you're on the plane you know you're good to go. When you're in the terminal you have to keep diverting your attention to announcements, changes, distractions, etc.

Re:Letting slower passengers board airplane first is faster,... by sokoloff on Wednesday January 15, 2020 @07:10PM (#22060330)
It also has optionality. If you want to board first, you're welcome to. If you want to board last, you're also welcome to do that with a first-class ticket.

I'm in the "board first, get it over with, get settled in" camp. My wife wants the boarding door to brush her backside as they close it. When we travel together, we don't always board together.

Re:Letting slower passengers board airplane first is faster,... by wtvanhest on Wednesday January 15, 2020 @03:44PM (#22058304)
One option that isnt discussed is having the overhead bins be seperately ticketed and a box. So if you want a bin, you pay $40 and one gets asigned to you. It would streamline boarding because people would put their stuff in the bin and be done.
Re:Letting slower passengers board airplane first is faster,... by reaperducer on Wednesday January 15, 2020 @03:57PM (#22058464)
Until they don't, and then the air hostesses who are just trying to get everyone seated have to become policemen as well. They have enough to do.
Re:Letting slower passengers board airplane first is faster,... by njarboe on Wednesday January 15, 2020 @04:53PM (#22059156)
Not really any different than having assigned seats for people.
Re:Letting slower passengers board airplane first is faster,... by novok on Wednesday January 15, 2020 @05:29PM (#22059518)
Overhead bins tend to be under provisioned for the amount of passengers boarding.
Re:Letting slower passengers board airplane first is faster,... by gefh on Wednesday January 15, 2020 @08:35PM (#22060855)
Because airlines starting charging for checked bags.
Re:Letting slower passengers board airplane first is faster,... by billatq on Wednesday January 15, 2020 @09:26PM (#22061154)
Aside from it being difficult to implement, I like this idea, especially if it’s less to check. I have noticed that some airlines, Alaska for example, will give free gate checks to folks.
Re:Letting slower passengers board airplane first is faster,... by TomVDB on Wednesday January 15, 2020 @04:17PM (#22058701)
although personally I've always wondered why this is desirable

The hustle and bustle at the gate stresses me out. (There's no good reason why and it shouldn't, but it does.) Once I'm sitting in my window seat, I close my eyes and relax.

Re:Letting slower passengers board airplane first is faster,... by Macha on Wednesday January 15, 2020 @05:19PM (#22059415)
Boarding first on budget Airlines like Ryanair guarantees you a space in the overhead for your carry on, as opposed to it taking all your legroom or being forcibly checked in at the gate ("free of charge" as if it's a service).

In some airports it also reduces total queueing time, though in others it just moves you from queuing at the boarding desk to queueing at a bus stop/stairwell/skybridge.

Re:Letting slower passengers board airplane first is faster,... by seanmcdirmid on Thursday January 16, 2020 @12:14AM (#22061854)
> although personally I've always wondered why this is desirable

Business class seats are nice and you get a swanky drink/treat to enjoy after boarding. When in economy, I always hang back during boarding unless I'm worried about overhead, but in biz it is really different. And they often board through their own door anyways (if international).

Re:Letting slower passengers board airplane first is faster,... by Domenic_S on Wednesday January 15, 2020 @07:31PM (#22060469)
> although personally I've always wondered why this is desirable

Carry-on space. I have yet to see a truly full flight where at least some carry ons aren't gate-checked.

Re:Letting slower passengers board airplane first is faster,... by groby_b on Wednesday January 15, 2020 @05:54PM (#22059722)
Boarding first is desirable because overhead bin space is v. limited. If you board last, it will be gone.
Re:Letting slower passengers board airplane first is faster,... by sokoloff on Wednesday January 15, 2020 @07:12PM (#22060349)
This is much, much less true in first or business class in my experience. (Same number of bins for fewer seats for a given aisle length.)
Re:Letting slower passengers board airplane first is faster,... by billatq on Wednesday January 15, 2020 @09:27PM (#22061166)
I’ve personally found that it depends on the aircraft. It’s especially true in a widebody where economy doesn’t walk through the same cabin, and not so true on a narrowbody with few seats.
Re:Letting slower passengers board airplane first is faster,... by mamon on Wednesday January 15, 2020 @05:15PM (#22059366)
One thing airlines should pay closer attention to is amount/size of carry-on luggage to better match overhead bins capacities. I've seen people bringing clearly oversized roller bag, AND big laptop bag, AND big purse, and then spending a lot of time trying to place all of that in bins, or under preceeding seat, blocking the aisle while doing so.

That's also the main reason why I try to board as fast as possible - to ensure that there's still space for my luggage.

Re:Letting slower passengers board airplane first is faster,... by reaperducer on Wednesday January 15, 2020 @03:53PM (#22058416)
I love CGP Grey's animations, and you can clearly tell this one involved a lot of work.

The problem with his analyses is that they're frequently wrong because he look at problems in a very academic way and severely oversimplifies them.

For example, his video about cars going through intersections makes perfect sense in a simulated world on a computer. But it falls down instantly in real life because it presumes that every car is identical, with the same mechanical characteristics, the same acceleration capability, the same connectivity (for the magic fictional traffic control signal), the same latency, they never break down, there are never any external factors at an intersection, and on and on and on.

Similarly, his plane boarding simulation falls down as soon as someone shows up late, or in a wheelchair, or with a bum leg, drunk, a baby, luggage that the wheel has fallen off of, etc...

He seems to live in a world where there is no difference between theory and reality. I still watch his videos, but I consume them as entertainment, not as documentaries.

Re:Letting slower passengers board airplane first is faster,... by NoodleIncident on Wednesday January 15, 2020 @04:43PM (#22059057)
I was under the impression that his conclusions come from the same sort of papers we're discussing in the OP, and that the animation is just a demonstration of the effect. Some real-life issues should be addressed in future studies, for sure; for airplanes, it's probably important to model "families" that sit, board, and deplane together. But just listing the variables not accounted for in the study or video isn't scientific criticism, you have to do your own work to actually demonstrate that it makes a difference.
Re:Letting slower passengers board airplane first is faster,... by reaperducer on Wednesday January 15, 2020 @05:38PM (#22059591)
You are correct. But his videos aren't presented as "this is possibly the best way, and it should be further investigated." They're presented as "This is the best way because I say so, even though I haven't done any testing but I'm a college professor and on YouTube so you should believe me."
Re:Letting slower passengers board airplane first is faster,... by emiliobumachar on Thursday January 16, 2020 @05:49AM (#22063366)
They're presented more like "This is the best way", the rest of your characterization is unfair.

This doesn't invalidate your core criticism, but there's no need to accuse him of trying to impose authority.

I've watched a lot of CGP grey and I just learned now, from you, that he's a college professor. I may have missed or forgotten it being mentioned a couple times, but I'm sure it's not front and center as you imply.

Being on YouTube is, tautologically, obvious from being on YouTube, but it's also necessarily obvious from being on YouTube, and a similar accusation of appeal to authority could be constructed for any other medium, e.g."it's best because I say so and I have a blog".

Re:Letting slower passengers board airplane first is faster,... by ntsplnkv2 on Wednesday January 15, 2020 @04:10PM (#22058620)
> Similarly, his plane boarding simulation falls down as soon as someone shows up late, or in a wheelchair, or with a bum leg, drunk, a baby, luggage that the wheel has fallen off of, etc...

All of these things occur in non-ideal boarding scenarios as well. I don't see how they affect his solution more so than what we currently have.

Re:Letting slower passengers board airplane first is faster,... by cgriswald on Wednesday January 15, 2020 @04:35PM (#22058947)
It’s sort of like comparing two algorithms and using the one that has the simplest best case time complexity without considering the average case or that your particular real-world use might actually be a worst case.
Re:Letting slower passengers board airplane first is faster,... by ntsplnkv2 on Wednesday January 15, 2020 @05:10PM (#22059330)
But ultimately the same problems affect both algorithms in the same way. So the best case algorithm will be slower in the real world, but should still be faster than the other cases.
Re:Letting slower passengers board airplane first is faster,... by reaperducer on Wednesday January 15, 2020 @05:36PM (#22059578)
Only if all algorithms are affected the same way by the same external factors, which simply doesn't happen.

It's like when a Toyota and a Freightliner crash into the same wall. They both crashed at the same speed, why is one damaged more than the other?

Re:Letting slower passengers board airplane first is faster,... by ntsplnkv2 on Wednesday January 15, 2020 @09:44PM (#22061248)
You can come up with endless scenarios to justify it, but the fact is you can't design something that will deal with all of them. So you design for the typical scenario.

Maybe there is an algorithm that boards a plane faster when there are 2 handicapped people, 3 drunks, and 1 person being late. But how typical is that? And is it realistic to design the whole algorithm around that? I would argue it isn't.

Re:Letting slower passengers board airplane first is faster,... by cgriswald on Wednesday January 15, 2020 @11:05PM (#22061600)
Real flights have those things in some numbers and a model probably ought to attempt to account for them, since, as you noted in a separate post I can’t seem to find now, they’ll affect the model negatively. The goal isn’t to find the best number of late people and wheelchairs. It’s to be able to board real flights as quickly as possible and every real flight I’ve ever seen, except the flight I took shortly after 9/11, has had some number of strollers, wheelchairs, or what have you.
Re:Letting slower passengers board airplane first is faster,... by s1artibartfast on Wednesday January 15, 2020 @06:07PM (#22059825)
I don't think we can make that claim. There is no reason to believe the problems will affect both algorithms the same way. Some may be more fragile or failure resistant to different failure modes.
Re:Letting slower passengers board airplane first is faster,... by tedunangst on Wednesday January 15, 2020 @08:10PM (#22060713)
Except this is not true. Quicksort is typically faster than mergesort, but worst case it can be very much slower. The worst case does not affect both algorithms the same way.
Re:Letting slower passengers board airplane first is faster,... by ntsplnkv2 on Wednesday January 15, 2020 @09:42PM (#22061239)
But we are not talking about quicksort or mergesort, either way it helps prove my point. Typically faster is an interesting way of describing something. What is typical? 80% of the time?
Re:Letting slower passengers board airplane first is faster,... by tedunangst on Wednesday January 15, 2020 @10:10PM (#22061355)
Your point is that all algorithms respond the same way to inputs, and a counterexample helps prove it?
Re:Letting slower passengers board airplane first is faster,... by ntsplnkv2 on Thursday January 16, 2020 @08:55AM (#22064498)
That wasn't my point - we are talking about a specific scenario here, not "all"
Re:Letting slower passengers board airplane first is faster,... by cgriswald on Thursday January 16, 2020 @11:22AM (#22065976)
An algorithm sees its best time complexity over some percentage of the input space. Over some other percentage it sees its worst time complexity. And, on average over the entire input space, it sees its average time complexity. The algorithms are necessarily different in what those best, average and worst are, and importantly, over what parts of the input space those categories apply. If they weren't, we'd call the algorithms mathematically equivalent and it wouldn't matter which one you chose. And if they all handled 'problem' data equally, we'd see that they all have the same performance penalty between best and average and average and worst. They don't. Sometimes best and average are equal. Sometimes worst and average are equal. Sometimes all three are equal.

If your data is truly random, you should pick the algorithm with the best average time complexity. Is your data random? Almost certainly not. It might be 'random enough', or it might not be. A real world data set may fall completely within the best case of algorithm A and the worst case of algorithm B, even if the average case of B is better than the average case of A.

In this case, the data points that are most likely to negatively affect an algorithm are specifically being excluded from the model. Far from comparing even the average case, we're comparing best cases. We aren't using a 'real world' data set. So it's an interesting simulation, but it isn't enough to reach any conclusions at all.