Hacker News Reader

Comment by mbreese | Hacker News Reader

Another way to look at this is:

25M out of 4.3B is a success rate of 99.4%.

Or 1 in 172 bags, or about 1 bag on every other flight.

You're in line at the boarding gate, ready to get on the plane. You checked your bag before going through security, so you're happy you don't have to lug it around, lift the heavy monster up, and try to cram it into a small overhead-bin on the plane. However, you notice the gate next to you is boarding as well. Between your flight and the other, one checked bag will go missing. Do you feel lucky?

When you put it that way, it doesn’t square with lived experience. Are the 25M lost or just misrouted? Nearly everyone I know who flies has had a bag misrouted at some point.

1-in-172 seems surprisingly low to me.

Bag handling errors aren't going to be randomly distributed but heavily effected by airport and airline quality so it makes sense that the actual stats don't match any one person's anecdotal stats. If you live or fly out of an airport with a bad baggage system most people you know are probably using that same airport when they fly.

I'm pretty sure they mean misrouted not lost. Lost makes a better headline.

Sucks badly, when you're in the 0.6% group, though.

That said I was standing a number of times dumbfunded at the carousel after all bags were distributed.

Usually it was due to having been sent with an earlier flight, have fallen down (probably) on the way to the carousel, to then be delivered to my place of stay (arriving earlier than me) or something around those lines.

Really "lost" only happened once, which was on a BA flight connecting in Heathrow (of course). It was delivered the next day and sent from the airport by taxi.

Still, it sucks if you're part of a statistic and your luggage really gets lost. And it does happen.

> success rate of 99.4%.

So after 100 flights, your chances of never having had an issue with your bags is 0.994^100 = 0.547 aka coin toss.

Math checks out except the problem is not randomly distributed - certain airports have a very much higher rate of lost luggage (LHR, JFK‚ CDG, looking at you!) and some types of luggage will be more prone to disappear.

https://www.primelinkinsurance.ca/blog/worst-reviewed-airpor...

nitpick: the problem is randomly distributed, just not uniformly distributed.

Yeah, there are definitely correlations. This was more a thought experiment.

This assumes an even distribution of likelihood this will happen, which is far from the reality. You can proactively reduce your risk — don't check in late, watch to make sure the check-in agent actually puts the tag on until you see it go down the conveyer belt, don't book tight connections, book even more generous connections in complex airports when you have terminal changes across the airfield from each other or airports with a poor reputation for mislaying bags on connections, use transfer airports that have been explicitly designed for rapid transfers (e.g. MUC). A general rule of thumb is if you are going to have to uncomfortably rush to make a connection, your bag is even less likely to make it.

Plus, should your luggage get waylaid, there are other things you can do to hasten its recovery: have a fairly unique bag that is unlikely to be mistaken by another traveller as theirs, clearly label your contact details inside and out in case the paper tag gets ripped off.

The success rate of a plane not crashing is much higher, yet it seems like keeping track of bags would be the easier problem to solve than ensuring a plane lands safely.

The potential losses of a plane crash are orders of magnitude greater than a few lost bags, so obviously much more effort and resources have been put into solving that problem. No system is perfect, and these two issues aren't even comparable.

Which might be close to the theoretical limit. The probability of two suitcases on a given flight having the same appearance must be close to 99%.

> Nearly half of all bags that go astray do so because of problems with flight transfers, Sita says. Delayed flights have knock-on effects - your bag just doesn't make it to the connecting flight in time.

> And a significant proportion of missing bags are down to passengers or handlers simply picking up the wrong one.

The former does seem like a hard problem without restricting people's ability to switch flights or delaying flights further to allow for luggage transfers.

I wonder if there is any room to improve the latter on the airline's part by adding visual variations to the tag without impeding the function. Like randomly assigning colors and giant emoji to be printed prominently on the tag, so that it is easier to distinguish at a glance whether it is really your bag.

(I mean, you should already be checking your name tag when you collect your bag, and people already add their own flair to their bags, but obviously not everyone is.)

A few airports have bag checks at exit, where they validate the tag against your chit. This is surprisingly rare though.

You can also tilt the odds in your favor by choosing an unusual bag color/design though: nobody's going to mistakenly pick up my lovely lime green number.

Or simply customize it at no cost.

I've just put some large stickers at the bottom of my bag (normally invisible, but in plain sight when on the conveyor belt).

Honestly I didn't have any such problems in mind; did it mostly for my own convenience - to avoid the usual "erm, is it mine?" hesitation.

That's exactly what I (and many others, including my old parents) do. It really helps when you're waiting for your case, particularly for those big conveyor belts with tons of people.

I do wonder whether the rare fellow traveler with a similar bag would be less likely to verify the tag when pulling it off the carousel than those with the standard black bag.

I like the random color/giant emoji idea. The major downside I see is space on the tag and all the current printers are black and white so they all need to be replaced for this to be effective.

I wouldn't be so sure of this. Usually when you go to a luggage shop here they only carry one of each models and refill them every now and then.

This is basically a reformulation of the Birthday Paradox. Even with a large variety of makes and models of luggage (cf. 365+ days in a year), the chances of any two being the same increases superlinearly wrt passengers.

That's what I had in mind. I figure on average half the passengers on any given flight are returning home. People living in the same area tend to have the same retail choices. The number of variations among this group probably isn't in the hundreds.

My point is that the variety of luggage is so vast (i'd hazard tens of thousands of models) that the probability would be low even for a quite high number of passengers.

So if limited choice of suitcase colour available, get a can of spray paint and customise it so you can easily spot it - and others won't accidentally mistake it for their own ...

Appearance doesn't cause airlines to lose bags.

Which is terrible.

Yeah but that's not a fun headline.

Reply on news.ycombinator.com