Airline Close Calls Happen Far More Often in the US Than Previously Known (yahoo.com) 36
The New York Times explores harrowing stories about recent airplane near-miss "close calls" on U.S. runways:
The incidents — highlighted in preliminary F.A.A. safety reports but not publicly disclosed — were among a flurry of at least 46 close calls involving commercial airlines last month alone... While there have been no major U.S. plane crashes in more than a decade, potentially dangerous incidents are occurring far more frequently than almost anyone realizes — a sign of what many insiders describe as a safety net under mounting stress. So far this year, close calls involving commercial airlines have been happening, on average, multiple times a week, according to a Times analysis of internal F.A.A. records, as well as thousands of pages of federal safety reports and interviews with more than 50 current and former pilots, air traffic controllers and federal officials.
The incidents often occur at or near airports and are the result of human error, the agency's internal records show... The close calls have involved all major U.S. airlines and have happened nationwide... In addition to the F.A.A. records, The Times analyzed a database maintained by NASA that contains confidential safety reports filed by pilots, air traffic controllers and others in aviation. The analysis identified a similar phenomenon: In the most recent 12-month period for which data was available, there were about 300 accounts of near collisions involving commercial airlines... One problem is that despite repeated recommendations from safety authorities, the vast majority of U.S. airports have not installed warning systems to help prevent collisions on runways.
But the most acute challenge, The Times found, is that the nation's air traffic control facilities are chronically understaffed. While the lack of controllers is no secret — the Biden administration is seeking funding to hire and train more — the shortages are more severe and are leading to more dangerous situations than previously known. As of May, only three of the 313 air traffic facilities nationwide had enough controllers to meet targets set by the F.A.A. and the union representing controllers, The Times found. Many controllers are required to work six-day weeks and a schedule so fatiguing that multiple federal agencies have warned that it can impede controllers' abilities to do their jobs properly.
The incidents often occur at or near airports and are the result of human error, the agency's internal records show... The close calls have involved all major U.S. airlines and have happened nationwide... In addition to the F.A.A. records, The Times analyzed a database maintained by NASA that contains confidential safety reports filed by pilots, air traffic controllers and others in aviation. The analysis identified a similar phenomenon: In the most recent 12-month period for which data was available, there were about 300 accounts of near collisions involving commercial airlines... One problem is that despite repeated recommendations from safety authorities, the vast majority of U.S. airports have not installed warning systems to help prevent collisions on runways.
But the most acute challenge, The Times found, is that the nation's air traffic control facilities are chronically understaffed. While the lack of controllers is no secret — the Biden administration is seeking funding to hire and train more — the shortages are more severe and are leading to more dangerous situations than previously known. As of May, only three of the 313 air traffic facilities nationwide had enough controllers to meet targets set by the F.A.A. and the union representing controllers, The Times found. Many controllers are required to work six-day weeks and a schedule so fatiguing that multiple federal agencies have warned that it can impede controllers' abilities to do their jobs properly.
Not really meaningful (Score:3)
Re:Not really meaningful (Score:5, Informative)
There are several examples in the article itself.
There are a few channels on Youtube that provide the ATC transmission audio for events like these, along with maps of varying degrees of quality. IANAP, but for the most part, the vast majority of incidents seem to fall into two categories:
Runway incursions - When an aircraft that isn't supposed to be on an active runway enters one, especially when another aircraft is on a takeoff roll or landing.
Separation failures - When two airborne aircraft get too close together, especially if the TCAS in one or both aircraft goes off.
What's interesting is that, of the incidents I've heard about, most of the time it's a controller making a mistake that causes the situation, not a pilot. I've always read that air traffic control can be an extremely stressful job. It seems like it may be taking it's toll.
Re: (Score:2)
An obvious solution is to replace the controllers with software. Get the humans out of the loop.
It is a simple problem. You have the GPS position of each plane. Direct them to where they need to go without crossing active runways.
Once they're in the air, it is even easier since there's a 3rd dimension. Easy for a computer, but too complex and overwhelming for a human.
I'm not busy this weekend, so if the FAA is interested I can whip up an ATC program in Rust and have it ready on Monday.
Re:Not really meaningful (Score:5, Insightful)
I think the problem here is that bold claims like this generate bold liability. How perfect do you feel your Rust ATC code will be? Any loss of life due to failure has previously been attributed to human error, but once you claim you have software that does this job "perfectly", well then everyone expects perfection for all scenarios. If your software fails, your company (ie, you) will be the scapegoat. Heavy fines are inevitable, and jail for negligence is always possible - after all, how come you didn't account for that previously unknown situation that happened only once but killed 400 otherwise healthy people in the process.
I don't disagree with your sentiment though. We should be using software to perform most of the ATC minutia and allowing humans to simply be observers, only engaged for the few unhandled emergency conditions. I just think that getting to that spot is rife with challenges beyond the software itself.
Re:Not really meaningful (Score:5, Funny)
How perfect do you feel your Rust ATC code will be?
It's Rust. So if it compiles, it works. Duh.
Re:Not really meaningful (Score:5, Informative)
That's a very strong statement that's ignoring a lot of critical aspects of computing engineering, the most glaring being human error by the programmer or the application designer. Just because it compiles does not imply that it does the desired work and handles all the edge cases. Not crashing isn't a useful success metric for code that is expected to make life/death decisions -- it needs to reliably not make poor decisions, and by "reliably" I mean "every time". Java posted a "don't use me to run nuclear reactors or really anything else that matters, ever" on every Windows install up through at least Windows 2k. When shit matters as a core property of function, it matters a lot.
But, hey, I wish you and your new startup a lot of luck. Though, I'd really appreciate it if you'd do me a solid and let me know what you name this new company so that I can ensure that I never personally interact with it during my travels. That'd be great. Thanks!
Re: (Score:3)
Re: (Score:2)
Possibly that. Or it's just not always that simple to detect, particularly when you don't know the sender very well.
https://duckduckgo.com/?q=dete... [duckduckgo.com]
For reference, this is why we have ambassadors to build relationships with touch points of foreign nations. If nations simply just emailed or texted back and forth, we'd all be dead from WW3 ages ago. But yes, apparently I missed that one. Good eye. Haha
Re: (Score:2)
No no, not Rust, but Haskell. You can write a mathematical proof that the code is correct!
Re: Not really meaningful (Score:2)
You'd think so, and if all aircraft and ground vehicles were run by computers, you might even be close to right.
But since there's humans in on the other side of the loop, figuring out how to deal with human input will consume all of your effort, and you'll have an appreciation fir why a weekend hackathon isn't a real solution here.
Re: (Score:2)
The problem is you have humans flying the plane, and they want to talk to ATC. A lot of effort has gone into standardizing the language used to avoid ambiguity, but even so. And when things go wrong and emergency situations arise, pilots don't always obey ATC instructions.
The real issue is simply the volume of traffic. There are too many aircraft trying to use too few airports and runways at the same time. They keep pushing for ATC to handle more traffic at once, so they can run more flights.
Re: (Score:2)
I've long suspected that ATC systems could be automated to a much greater extent than they are currently. At least in the U.S. (whose systems are quite old), but probably elsewhere as well.
You do need a human there for many, many reasons, but most of the routine work should be able to be automated by this point in our state of technological development, leaving most of the human workload reserved for difficult and/or unusual circumstances. That's more or less how commercial flying works now. Autopilot ta
Re: Not really meaningful (Score:5, Informative)
In aviation, unlike in most other industries, not just incident reports but also near incidents are reported. Not just internally, but those reports get forwarded to the authorities (in the U.S., that would be the FAA).
Such a near incident could be, for example, that ATC gave clearance to execute a certain request but crew autonomously just in time decided not to, that they deviated from the projected action on their own consideration. Or that a malfunction could have caused a flight-safety issue but that the flight ended without incident. Or that a tool was found where it could have wreaked serious havoc but for some reason it was discovered before any incident happened. Any (near) incident between 'blocks off' and 'blocks on' is a (near) flight incident, but the same goes for ground incidents reported by maintenance, for example.
So these near incidents are indeed not known to the public (what would be the use of that?) but they are commonly reported and thus known within the industry and studied and if needed procedures or whatever are adjusted to prevent actual incidents by similar causes.
Aviation tries to map the whole iceberg, not just the tip that's visible from afar. So maybe some in the public are shocked to realize there isn't just incidents, but lots more near incidents, too. Do they *need* to know about near incidents? Idk. Long as the industry learns and takes action when needed, I'm cool with that and there's nothing new there.
Re: Not really meaningful (Score:2)
Hi, your 250g drone will require special licensing fees and courses. Oh, did you mean "under 250g drone"? (To cover all basis, it better be under 249.476 grams, since the FAA also mentions under 0.55 pounds.)
So, of course you mean your 249.475 gram drone, right???
Is this an increase, or about the same? (Score:2)
Are we just now discovering "close calls," or are they actually becoming more common?
Re:Is this an increase, or about the same? (Score:5, Interesting)
Every once in a while they make this point that airline close calls are becoming more common. I barely watch the news, and Id say I see it once every 4 years or so.
Its weird, because nothing further ever gets reported on it, and even though the FAA in the USA is struggling to hire people (pay sucks and they run people ragged, ever since 1982 when Reagan fired all the Gov employees when they went on strike), the reports stay the same vague, monotonous report every time.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Literally the FAA and their policies are their own fault. They wont increase wages, and they wont support their workers. Weird how it got worse after the union was essentially fired/dissolved under Reagan (who was a member of SAG btw...)
Like crime statistics (Score:4, Insightful)
If you live in a very safe neighborhood like I do, you might think crime doesn't happen in your area. People in my neighborhood are not afraid to walk outside, even at night. Neighbors know each other and exchange housekeys so they can look out for each other when traveling. But the reality is that crime happens even in neighborhoods like this, far more often than we would think. You might not want to check the police crime blotter for your area, your bubble of security just might be burst.
These "near misses" are similar. We live in a world of near misses. Once in a while, the near miss becomes an actual collision. This is not new, it's just that we are now looking at the "police blotter."
Re: (Score:3)
These "near misses" are similar. We live in a world of near misses. Once in a while, the near miss becomes an actual collision. This is not new, it's just that we are now looking at the "police blotter."
I always preferred George Carlin's take: when they "nearly miss"... they've actually crashed into each other. We really ought to call them "near hits".
Re: (Score:3)
Indeed, "health and safety" teachings do treat them as equivalent. As in an action plan for remediation is activated on the near-miss. Cost is not spared when done right.
In this case, flights per day would be severely curtailed until more staff gets hired.
Re: (Score:2)
Right! But I could care less... :-)
Labor Laws. (Score:2)
Re:Labor Laws. (Score:5, Insightful)
Re: (Score:3, Funny)
Another reason to piss on Reagan's grave (Score:2)
Air traffic controllers wanted a shorter work week because humans just aren't built to have maximum concentration for 40 hours five days out of seven. US is such a failed state it is literally possible for your barista handing you a latte will also be the copilot on you flight.
But we could end homelessness in the US for what we give Ukraine in just one month. Where US taxpayers are funding their health care and pensions.
Re: (Score:2)
Nix hype, show me the stats and safety culture (Score:2)
I train pilots (Score:4, Interesting)
I am a flight instructor.
Near misses, and near fatal accidents, happen a lot more often than you think they do. They shouldn't, but they do. Generally, the hope is that people wash themselves out, before they become dangerous. It doesn't always happen that way, though.
EVERY person who tries to learn how to fly has weaknesses. Some are more aware of those weaknesses than others. Usually this results in a few near misses at your local flight school that you never hear about.
The problem is that this cavalier attitude extends beyond flight training into commercial aviation - that mistakes are okay. In many vocations, mistakes are quite acceptable. In aviation, they are not. Blame is unimportant and fault is unimportant, but each and every single mistake must be investigated.
In flight training, mistakes are common. In commercial aviation, they are not. A whole industry has evolved to ensure that mistakes are not common, but the primacy of flight training, where mistakes are okay and accidents are survivable, stays with pilots long before its time is over.
Re: (Score:1)
...Generally, the hope is that people wash themselves out, before they become dangerous. It doesn't always happen that way, though.
I would suggest the very notion of "washing out" as a trainee pilot these days doesn't really apply to a vast number of the pilots working towards a professional airline career.
The "washout" approach was founded in military pilot training, where you either meet the standards or you were out. After all if you have a large pool of interested candidates you don't need to waste the time on someone without the aptitude for the job. (Air traffic control training follows a similar regimen - unless things have cha
Diversity (Score:2, Funny)
Re: (Score:2)
You may joke, but based on numerous youtube comments I have seen from airline pilots, SFO is the worst, and the reason appears to be DEI. They even complain about individual ATC controllers as reliably incompetent.
previously known by whom? (Score:2)
Oh, ok, by NYT reporters. I'm pretty confident that anyone in the industry knows quite well. Perils of passive voice.
ATC from 1945 (Score:3)
The entire ATC system is essentially unchanged from 1945. Airports are much improved with various glide slope technology, airplanes are much improved, but ATC still functions by voice communication over noisy channels with the possibility of interference. There is no digital communication, and pilots literally have to write down instructions on paper, or on an ipad, by hand.
In addition, many of the recent incidents occur because unlike Europe, the FAA allows some airports to put an airplane on a runway while another airplane is already on final approach, in the hopes that the former will be able to take off before the latter crashes into it.
Easy solution (Score:2)
Start fining airlines and airports that are involved and continue to increase the fines for each subsequent event in a calendar year with no cap.
The movie "Pushing Tin." (Score:2)