3 ms·
The mythbusters segment alluded to in the article is quite interesting to watch. I think they did a quite good attempt at creating a real-life scenario for boa
by brc 12y ago
The mythbusters segment alluded to in the article is quite interesting to watch. I think they did a quite good attempt at creating a real-life scenario for boarding a plane.
While the article suggests that part of the issue is that the airlines create inefficiency in order to be able to offer perks to get around the inefficiency. That's probably a minor consideration (I can easily see the 'rewards program' suit arguing in a meeting to keep it).
But I think there would be a lot of inertia in changing a program like boarding allocations. There are a lot of procedures around boarding - we don't know exactly what 'hooks' the airlines have around their procedures. We also don't know how more complicated systems test with the general public - including first time flyers, people who don't speak the same language as the airline, and other edge cases. The current methods are quite resilient against these problem-cases, and the more-efficient ones may collapse faster as a result.
Essentially I think that it's easy for outsiders to suggest changes - but we should reflect on that like user requests like 'why don't they just change the UI on this software to make it more like facebook?'. It would be good to get an airline insiders knowledge and comments.
- pessimizer 12y agoI don't like the 5% problem people that Mythbusters added to the mix. Boarding back to front is sensitive to morons, and they literally and arbitrarily guaranteed that 1 out of 20 people would delay the line, causing everyone to have to wait. If they did it at an even increment, that would be the worst situation possible for back-to-fronters. Couldn't you just integrate over all possible distributions of that 5% (assuming that number came from somewhere) in the line? Seems like this problem could be easily attacked with math instead of experiment.
- brc 12y agoThe moron-boarder is a non-trivial issue. I don't think you can adequately model that type of behaviour. Any method has to be robust to problem boarders - people who go for the bathroom, can't find their seat, bring on too much luggage, don't speak the language, etc. If anything I think the 'morons' in the Mythbusters ep were moronic enough. Certainly anyone who travels frequently wil l be able to cite behviour far worse than they modelled.
- ars 12y ago> The mythbusters segment I don't think that was mythbusters.