Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
Transportation United States

How a Series of Air Traffic Control Lapses Nearly Killed 131 People (deccanherald.com) 58

Due to an air traffic control mistake in February, a FedEx cargo plane flew within 100 feet of a Southwest Airlines flight in February. The New York Times reports that the flight's 128 passengers "were unaware that they had nearly died." In a year filled with close calls involving US airlines, this was the one that most unnerved federal aviation officials: A disaster had barely been averted, and multiple layers of the vaunted US air-safety system had failed... But the errors by the controller — who has continued to direct some plane traffic in Austin, Texas — were far from the whole story, according to 10 current and former controllers there, as well as internal Federal Aviation Administration documents reviewed by the Times. Austin-Bergstrom, like the vast majority of US airports, lacks technology that allows controllers to track planes on the ground and that warns of imminent collisions. The result is that on foggy days, controllers can't always see what is happening on runways and taxiways. Some have even resorted to using a public flight-tracking website in lieu of radar.

In addition, for years Austin has had a shortage of experienced controllers, even as traffic at the airport has surged to record levels. Nearly three-quarters of shifts have been understaffed. Managers and rank-and-file controllers have repeatedly warned that staffing levels pose a public danger. The controller on that February morning was working an overtime shift. In June, Stephen B. Martin, then Austin's top manager, and a local union representative wrote a memo pleading for more controllers. "Drastic steps are needed to allow the facility to adequately staff for existing traffic," they wrote to FAA and union officials.

Austin is a microcosm of a systemic crisis. The safety net that underpins air travel in America is fraying, exposing passengers to potential tragedies like the episode in February.

And yet the chair of America's National Transportation Safety Board calls the February incident "just one of seven serious close calls and near misses involving commercial airlines that we have initiated investigations on this year."

Thanks to long-time Slashdot reader schwit1 for sharing the article.
This discussion has been archived. No new comments can be posted.

How a Series of Air Traffic Control Lapses Nearly Killed 131 People

Comments Filter:
  • by banbeans ( 122547 ) on Saturday October 14, 2023 @05:52PM (#63925467)

    The air traffic control system has been broken since the 90s.
    They hire big contractors to fix it and they just pocket the free money and nothing gets fixed.

    • Re: (Score:3, Insightful)

      by jhoegl ( 638955 )
      Nope, its been broken since the 80s when Reagan destroyed the Union.
      • Shit you're right...Things were much more dangerous before Reagan destroyed PATCO in 1981.

        https://aviation-safety.net/st... [aviation-safety.net]

        I think you're on to something here...Maybe Ford and GM will stop producing shitty cars if the UAW is destroyed too.

    • They hire big contractors to fix it and they just pocket the free money and nothing gets fixed.

      Maybe I'm missing something, but why is it a hard problem?

      A couple college students should be able to slap something together in a weekend.

      • Submit a bid and git r done.

      • by RightwingNutjob ( 1302813 ) on Saturday October 14, 2023 @07:10PM (#63925587)

        It's not hard if you know what you're doing in every applicable domain.

        Software, radar, human factors, aviation, weather, probable some more obvious ones I'm missing, and a couple non-obvious ones too.

        Plenty of college students can get smart about one or two of these "in a weekend" but not many organizations can get together enough people competent across everything necessary *and* make sure they're working together vs in parallel silos.

        Same as any complex system that works most of the time. Someone did a lot of hard work to make it look easy.

        • by GumphMaster ( 772693 ) on Sunday October 15, 2023 @03:05AM (#63926033)
          I have been dealing with A-SMGCS [skybrary.aero] data streams from "big boy" systems for the past few months. Identifying everything moving in an airport precinct is a difficult problem. You have radars with all their weaknesses, active transponders with their weaknesses, numerous other potential sensors, a range of motions from stopped to ~200 knots on the ground and in the air, large numbers of obstructions (buildings, other aircraft, terrain, other transmissions...) that reduce raw data quality, human inputs (e.g. notification that an aircraft has been cleared to push back), disagreements about "facts" like what constitutes "airborne" or "landed", unavoidable measurement errors, ... and you have to synthesise a single, cohesive picture that is fit for its purposes (the one's you designed for and the others that have crept in). Above all, the system must improve safety. It cannot fail (usually at all) and certainly not in a way the causes safety issues.
      • by AmiMoJo ( 196126 )

        It's a mixture of ancient systems with proprietary and bizarre interfaces. Lots of rules and regulations that must be accounted for. There are established ways of doing things, created in reaction to various to accidents and failures. The industry and regulators are very resistant to change for that reason.

  • by RightwingNutjob ( 1302813 ) on Saturday October 14, 2023 @06:27PM (#63925499)

    How many more where noone involved even realized a mistake occurred?

  • Too many planes (Score:5, Interesting)

    by akw0088 ( 7073305 ) on Saturday October 14, 2023 @06:58PM (#63925559)
    Think there is an issue with the regular practice of allowing planes on a runway while there is a plane a few miles out on approach. If they didnâ(TM)t try to squeeze so many planes onto the runway there would be less chances of accidents. Overworked controllers with low pay, and the rules surrounding not wanting to train people over age 25 or something ridiculous like that only helps to contribute to the problem
    • I was at a local evening event after sunset that is not really near any major airport and there was never a time that a plane was not visible in the sky - in fact there were always at least two and usually three or four visible from the same vantage point. There are simply too many planes. At least at that event you mostly never heard them fly over unlike my residence where there is always one roaring overhead every minute or two.

      -Get off my sky.
      • Every now and then LAX gets loaded and there are a lot of planes coming in to land at the dual runways. At night we call them a "string of pearls".

        • Living in north seattle, when the weather was clear, looking south towards the SeaTac airport in the evening it looked like a "Stairway To Heaven".
          Always amazed me that it worked so well, and that on the cloudy/foggy days there weren't mass collisions.
          Got to trust all those instruments like a mofo.
          I avoid flying now.

      • by CAIMLAS ( 41445 )

        Nothing new. You must live in a city.

    • rules surrounding not wanting to train people over age 25 or something ridiculous like that only helps to contribute to the problem

      Long ago I had a friend that was an ATC. He told me that the cutoff age for new ATCs was 25(?) because they had 100% failure rate of everyone older than that who entered the training program. He was just under the cutoff age and scary driven and focused. So not so much a ridiculous rule and more of a pragmatic realization.

  • fix the february dup typo
  • These near misses have been increasing in the past few years. Start with a 50Â cess on all tickets for a few years to fund research and development for making things better.

    Why not feed all voice and traffic logs to AI and train it to find patterns leading to problems? But this should only be a small part of the initiative. The main problem should be addressed by fixing staffing issues and improving rule-based (non-AI) software systems.

  • With all these issues I am surprised airports are trying to reduce the number of flights and just have airlines use larger planes.

    Iâ(TM)d even argue the FAA needs to provide an updated formula on the number of planes that can pass through at any point.

    Beyond planes, the US really needs to starting reevaluating how high speed trains address future transportation.

    • ...or how about thinking in joined-up government & thinking about infrastructure in general as a whole system? Every country benefits from a planned, cohesive transport system which takes into consideration all the ways that people get from A to B & then consider how much that costs in people's hours (that could be spent doing something more productive or enjoyable) & in resources (energy/fuel, roads, rails, airports, etc.). If you can reduce the impact of transport on your population & your
      • I agree. Transportation needs to be seen as a complete solution made up of the different types of solutions. The US was built with the help of the railway, then the government favoured the car and the aeroplane to its disadvantage.

        The problem today in the US is partisan politics that gets in the way of actually improving the country.

        While the EU does have its own issues, at least it has a coherent plan to improve travel across its member states.

  • As of writing, I can see 7 aircraft on approach to AUS. Traffic increased greatly in 2022, almost doubling from 350 to 700 flights per day. Also when multiple flights arrive within a short time-frame, there have to be sufficient controllers so they're not overworked. The FAA has to get their shit together, pay controllers more, and ensure adequate staffing levels.
    • The explanation for the increase in flights in Austin was a very hot summer and locals leaving for vacation. Tourism also fell off, so it wasn't that.
  • I can't believe we allow this. It's perfectly accepted for aircraft big or small to takeoff and land simultaneously directly next to each other on close parallel runways. A mechanical failure, communication failures, flight plan mistake, simultaneous go around, or who knows what kind of series of unfortunate events could end in real disaster. This isn't only about the planes but also people on the ground in dense urbanized surroundings. Remember, most aviation accidents happen during takeoff and landing. Ma
  • money wasted (Score:5, Insightful)

    by bugs2squash ( 1132591 ) on Saturday October 14, 2023 @08:13PM (#63925641)
    Disband the TSA and put the money saved into making travel safer.
  • I read TFA.

    At the end: "It isn’t clear whether he [the controller] will face repercussions after the NTSB investigation is complete."

    One thing that *is* clear though: If he does face repercussions, it'll be even more difficult to hire competent ATC staff in the future.

  • incompetent article (Score:4, Interesting)

    by groobly ( 6155920 ) on Sunday October 15, 2023 @11:25AM (#63926573)

    These fine ladies who wrote this article know nothing about aviation, but it's the NYT, so why be surprised.

    The problem that they are unaware of, in their in-depth analysis of everything else other than what the problem is with this event, is that under FAA regulations controllers are allowed to put planes on the runway while another plane is on final approach. This is allowed for only 1 simple reason: to increase throughput because airports in the US don't have the capacity to handle all the traffic that wants to fly at peak times.

    As I understand it from comments from European pilots on aviation channels, Europe, which also has all the fancy doodads that these ladies complain about, PROHIBITS this action. No plane may be on the runway when there is another plane on final.

    This whole event, and numerous similar ones, are the result of bad FAA rulemaking.

  • I am a commercial pilot.

    The ATC system in the US is woefully outdated and obsolete. Most of the controllers are great professionals who are good at their jobs, but they have been handed a shit sandwich and are expected to chew and swallow it.

    I would say at least HALF of ATCs function could be automated. The need for human interaction in half to 2/3rd of the course of normal operations is a major flow choke point in how things moving metal. Couple this with simplex radio communications and their only option

Top Ten Things Overheard At The ANSI C Draft Committee Meetings: (7) Well, it's an excellent idea, but it would make the compilers too hard to write.

Working...