Slashdot Mirror


A Worry For Some Pilots: Their Hands-On Flying Skills Are Lacking (nytimes.com)

An anonymous reader shares a report: Pilots now spend more time learning automated systems than practicing hands-on flying, so newer pilots are less comfortable with taking manual control when the computer steers them wrong, according to interviews with a dozen pilots and pilot instructors at major airlines and aviation universities around the world. "The automation in the aircraft, whether it's a Boeing or an Airbus, has lulled us into a sense of security and safety," said Kevin Hiatt, a former Delta Air Lines pilot who later ran flight safety for JetBlue. Pilots now rely on autopilot so often, "they become a systems operator rather than a stick-and-rudder pilot."

As a result, he said, "they may not exactly know or recognize quickly enough what is happening to the aircraft, and by the time they figure it out, it may be too late." [...] While automation has contributed to the airline industry's stellar safety record in recent years, it has also been a factor in many of the crashes that have still occurred around the world. A 2011 study by a federal task force found that in about 60 percent of 46 recent accidents, pilots had trouble manually flying the plane or handling the automated controls. Complicated automation systems can also confuse pilots and potentially cause them to take action they shouldn't, pilots said.

27 of 173 comments (clear)

  1. Nothing new... by sabri · · Score: 4, Insightful

    This is nothing new. It is a recurring topic, especially after Asiana in SF. Hand flying is a no-no these days so skills go away.

    Children of the magenta line...

    --
    I'm not a complete idiot... Some parts are missing.
    1. Re:Nothing new... by cdsparrow · · Score: 2

      You would think that reasonable training would be required before you hand someone the keys to a 150mil airliner... But I guess not.

      Seriously, if pilots can't actually fly the planes during an emergency anymore, why have the pilots at all? Completely automate it and I bet the planes would do just fine.

    2. Re:Nothing new... by thereddaikon · · Score: 4, Interesting

      Because even if they have it on autopilot they don't just get to check out. They have to stay on the radio, check radar, monitor instrumentation and be able to take control back from autopilot at a moment's notice. That is mentally taxing because you are forcing yourself to keep focus.

    3. Re:Nothing new... by Kjella · · Score: 2

      When the computer partially or completely hands over control it's often an instrument problem like a faulty sensor not a core logic problem, it's reached some kind of absurd/impossible state. Commercial airlines often fly in the dark, across oceans, through storms and under other conditions where it's impossible or at least extremely hard to navigate visually, so you got little choice but to try to make sense of it. Very often the problem is that you end up ignoring or fighting what you think is faulty when something else is the problem.

      For example one case I read about it was the altitude/airspeed/angle having a glitch, in any case the crew ended up convinced the plane was diving. So they pulled up, got stall warnings, ignored them and the plane trying to cancel the stall and pulled up more until the plane stopped giving alerts because the angle of attack was so off it considered it faulty. When they still lost altitude they tried to level out the plane a bit, which brought the angle back into a sane range and the warnings began again so they pulled up. Then stalled all the way down until they crashed in the ocean.

      Was is that they didn't know how to fly a plane as such? No, it's that they thought the stall warnings were false. They thought no warnings meant they'd regained control. Maybe it didn't really make sense, but a lot of these will be once-per-career situations. Not that it's the only time something will break but maybe the only time that thing will break in that way under those conditions. And you're supposed to figure it out on the first try. It's pretty easy for the crash investigators to sit there in their calm office sipping a cup of coffee to say what you could have done or should have done.

      --
      Live today, because you never know what tomorrow brings
  2. Computers and planes by sycodon · · Score: 2

    I would never pilot a plane without a physical connection, wires or hydraulics.

    But I'm not an airline pilot nor can I afford any plane that doesn't have wires, let alone a fly by wire system.

    But at some point a decision will need to be made...do we have pilots or do we have cockpit managers?

      Fly by wire is probably unavoidable in the future, but I think that every airliner will need a mode that provides a direct link to the control surfaces without a computer making decisions and overriding input.

    --
    When Fascism comes to America, it will call itself Anti-Fascism, and tell you to give up your guns.
    1. Re:Computers and planes by b0s0z0ku · · Score: 4, Interesting

      The 737 is not primarily fly-by-wire (not even the MAX). Some secondary flight controls are fly-by-wire (electric tail stabilizer trim, spoilers). The problem is that a pilot needs to know how to recover from runaway stabilizer trim and know that the situation is happening. Since the system was poorly documented, this was a problem.

    2. Re:Computers and planes by b0s0z0ku · · Score: 2

      99% of cars still have a mechanical connection between the steering wheel and the road wheels. "Electric power steering" doesn't mean "steer by wire" -- it mans that an electric motor has taken the place of the hydraulic ram in a conventional power steering system. The only car with "steer by wire" is the Infiniti Q50 -- even that car has a connection between the wheels and the steering wheel that's engaged if power to the compuker is lost.

  3. Cockpit voice recorder transcript by olsmeister · · Score: 2

    "Alexa, pull up. Alexa? Alexa! Alexa, pull up! Alexa, PULL UP ALEXA ALEXA PULL UP ALEXA OH MY G--------"

  4. This is not a new concern by bobbied · · Score: 5, Informative

    This issue has been a concern for more than a decade. The more pilots use automation, the more their manual flying skills languish. Manual flying is a skill that one must practice to stay current. It may be like riding a bicycle, you never forget, but the fine skills required to fly accurately without the automation is something you can loose. It takes practice to stay current and proficient. It takes practice to be smooth and accurate, like playing a musical instrument it takes regular playing to keep your skills sharp.

    There have been a couple of instances where the pilots where faced with the loss of automation and made mistakes with their manual flying. Or situations like Asiana Airlines Flight 214 where the automation wasn't set properly and the manual flying skill and experience wasn't enough to notice and avoid the accident. Pilots and airlines LOVE automation. Pilots like it because it makes their job easier (when it works). Even an unskilled pilot can fly like a pro using automation. I've been in simulators, and although I've never flown anything more complicated than a Cessna 172, I can get the simulator on the ground without balling it up, usually. Airlines love it because it allows the aircraft to be operated in the most efficient way, saving them fuel and maintenance costs.

    But button pushing and turning dials isn't flying. Pilots are spending lots of their time managing and monitoring the automation and very little flying the aircraft. It's hard to keep your manual flying skills in top form, when you do mostly button pushing and turning dials

    When the automation fails, and you are forced to grab the controls and fly while trying to diagnose what's wrong with your aircraft, Navigate, Communicate with ATC under high stress, you really want those "stick and rudder" skills sharp and current. It's one less thing to think about while you are trying to wrap your head around what is wrong with the aircraft.

    --
    "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
    1. Re:This is not a new concern by edtice1559 · · Score: 2

      They don't let us blind people fly, but I do believe that the sequence is aviate, navigate, communicate. I don't think you are supposed to diagnose until you are in controlled flight. But that's problematic if you can't stick and rudder fly so you'll never get to diagnosis.

  5. Re:This seems easy enough to remedy. by bobbied · · Score: 4, Informative

    How about training pilots how to fly?

    Also, does anyone else read this story as, "I don't know how to do my job, but I want to keep getting paid to do it."?

    Fucking stupid.

    Stick and Rudder skills are like playing a musical instrument. They require constant practice and exercise to stay "current".

    This is not a new realization, but one that's been enshrined in the FAA regulations for decades longer than I've been alive. Pilots are required to "be current" meaning they have flown a minimum number of landings as PIC in the last few months before they can fly. I am also are required to "be current" with a flight instructor every year. Airline pilots must be "current" in their type ratings, which includes both training, check rides and flight time.

    So it actually makes sense.. We are taking away the On the Job training opportunity in Stick and Rudder operation, replacing it with button pushing and dial turning. What's a pilot to do? Company regulations likely REQUIRE he/she use the automation for safety, efficiency and passenger comfort so it's hands off the controls while "Auto Pilot" flies for you.

    --
    "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
  6. Makes sense by Sqreater · · Score: 2

    I bet this is a general problem though and occurs across all automated systems and AI systems - the degradation of human knowledge, skills, and abilities. And from those things come the next advances. Automated systems and AI will kill advances by freezing systems into the current state of knowledge. Advanced Technology-caused Stasis.

    --
    E Proelio Veritas.
  7. Routes by JBMcB · · Score: 2

    Since when is going to Caribbean countries "crap"? It seems like a fun (and relatively inexpensive, depending on country) trip.

    Because, when flying from the southeast US, most of the Caribbean is *just* short enough to be considered a "day trip" so you don't get to stay overnight. You fly there, deplane, load up, then fly back. Also, you are mostly flying over water, which is boring. The quality of the airports is pretty variable, as well. Anywhere from a typical modern airport to the aforementioned landing strip. For the more "rustic" airports, the pilots tend to have to do more mechanical work as the ground crews can be unreliable or nonexistent. This means crawling around the plane checking everything in the Caribbean heat.

    A "good" route would be flying from Chicago to Dallas, hang out in the nice pilot's lounge for an hour, do a quick check of the plane, fly back to Chicago, hang out in the lounge... repeat...

    --
    My Other Computer Is A Data General Nova III.
  8. Re:A preview of self driving cars ... by Nidi62 · · Score: 2

    Autopilots in aircraft are closer to cruise control than any level of self-driving in a car.

    There was a mythbuster's on self landing aircraft that would dispute your findings. Maybe you should watch it.

    Autopilot in aircraft is literally nothing more than setting the airspeed and programming in navpoints, so it basically is cruise control that knows what headings to take.. Autoland is a little more complicated though.

    --
    The only thing necessary for evil to triumph is for it to be pitted against a slightly greater evil
  9. they will blame the pilot by swell · · Score: 4, Insightful

    My friend the airline mechanic told me that accident reports are designed to protect profits. The pilot will always be blamed. Any suggestion of bad design, poor construction or mismanaged maintenance has dire consequences for manufacturers, airlines and others including politicians who depend upon political donations.

    My friend has often told me of his discoveries as he goes about his work. Inside a jumbo jet wing he may find beer cans, panties, drug related garbage, as well as tools and loose parts. He found this shocking so he took it upon himself to investigate every airline accident.

    Each accident results in a huge report after a multi year investigation. Those reports are available to anyone willing to study hundreds or thousands of pages. The summary will say the pilot was at fault, but if you read carefully you will discover many disturbing facts about the condition of the aircraft.

    It shouldn't be necessary to explain to jaded Slashdot readers that money is the motivator in most corporate and government behavior. Blaming the pilot is the way to reduce costs and bad publicity. If there is a flaw in the airplane, it will be fixed quietly.

    --
    ...omphaloskepsis often...
    1. Re:they will blame the pilot by DerekLyons · · Score: 2

      My friend the airline mechanic told me that accident reports are designed to protect profits. The pilot will always be blamed.
          Any suggestion of bad design, poor construction or mismanaged maintenance has dire consequences for manufacturers, airlines and others including politicians who depend upon political donations.

      Your friend the airline mechanic should actually read the accident reports. I do, and I regularly see not suggestions but flat out pronouncements of "bad design, poor construction or mismanaged maintenance".

  10. Not an actual airline pilot, but... by Anonymous Coward · · Score: 3, Interesting

    I did fly around in glider planes now and then. That's very visceral. Stall the thing, you *know* you're losing control. So regularly when I'm watching mayday / aircrash investigation episodes, I'm like "nose down! pick up speed! get control back! THEN try and get back on track", or something like that. A number of crashes might possibly have been avoided if the pilots had caught on to stalling and remedied it in the obvious way.

    Frankly I find it amazing, but intellectually I understand that big airline pilots need stick shakers for them to even notice a stall. Apparently this lack of viscerality is how air france 447 managed to fly into the sea and nobody noticed. That and a couple other things with pilots not being in full control, being kept by a smart aleck computer confined to "safe spaces" that weren't so safe after all. A glider plane will really just lose lift, drop its nose on its own, and you'll *know* when that happens. You can feel the air flow disengaging and so your wings lose lift and your control surfaces lose control. If that didn't wake you up, that violently dropping nose will, hopefully. At that point you steer with it until you get control back, then ease out of trouble. Of course, you lose a good bit of height while doing that, so doing it too close to the ground can end with you sitting on the ground in the remains of the airplane, and you can count yourself very lucky if all you got is a few scratches. Apparently big aircraft, even just motorised aircraft, work differently. They certainly crash differently.

    Anyhow, I'd be very uncomfortable flying if I knew the pilot didn't have hands-on experience with stalls and the like, or was even the slightest bit uncomfortable with taking manual control. Who's the pilot here, hm?

    It also makes me a little jealous. Back when, prospects for pilots were dim. Expensive training (and now I'm like "but what tf do they even teach the kids there?!?") and probably no job. So I skipped on that "opportunity". Fast forward a couple decades, and they can't find enough experienced pilots so they'll just hire almost anyone? Why, maybe I should fire up the old flight sim for a bit again, eh. Haven't hit 50, so maybe. But no, the cut-off for pilot school was something like 29 and three quarters, and I'm way past that. Oh well. I'll take the train when I need to be somewhere, and plan accordingly. Only 80 days around the world, what.

    1. Re:Not an actual airline pilot, but... by Anonymous Coward · · Score: 3, Informative

      But to be fair, in the case of 447, they were flying in the coffin corner, meaning a small amount of airspeed between an aerodynamic stall due to flying to slow, and a stall due to flying too fast (yes, that exists). The first officer of that flight clearly did not recognize the state they were in and yes, essentially flew the plane to the ground.

      AF447 crashed because one of the pilots was holding the stick all the way back. I won't speculate as to why, but there was a lot going on in the cockpit. The other pilot was pushing the stick all the way forward, presumably because he recognized what the airplane was doing. In the A330, the control sticks are not physically coupled, so it was possible to have one stick all the way forward and one all the way back simultaneously. In the mode that the automation was running, it's resolution to the discrepancy was to average the two inputs, so the stick was effectively neutral. Since the place was already in a deep stall at that point, it did not recover. The captain was on a rest break at the start of the trouble. When he got into the cockpit, he quickly (and correctly) assessed the situation, but realized that it was too late.

      Go and read the cockpit recorder transcript (easily found online). It is chilling.

      Regarding stall recognition, I am a private pilot. My father was an airline pilot for about 30 years, most of it on the 737. He passed a way a while ago, so I can't talk to him about this incident. We did talk about flying frequently, and he assured me that it was possible to recognize an impending stall without the systems warning you about it. He loved to fly the 737. It was his favorite aircraft after flying everything from DC-3s, up to 737-300 and just about everything in between, including a large number of small planes.

    2. Re: Not an actual airline pilot, but... by bkmoore · · Score: 3, Informative

      Up until very recently airline training in stall recovery emphasized minimum altitude loss. If you allowed the nose to drop, by reducing the AOA, you might loose too much altitude and fail the training event. You were supposed to use thrust and preserve altitude. This procedure works in an approach to a stall, where the airplane isnât yet in a stalled condition, but is ineffective in a fully developed stall. After Colgan Air, the industry has changed how Stall recovery is trained to, emphasizing reducing the angle of attack then recovery from the unusual attitude. Itâ(TM)s not that airline pilots forgot how to recover from a stall, itâ(TM)s more that they were trained in the Simulator wrongly and then utilized an inappropriate procedure in an actual emergency.

    3. Re:Not an actual airline pilot, but... by ceoyoyo · · Score: 4, Interesting

      I fly hang gliders. One of the crazy things we do is tie ourselves to pickup trucks and let them tow us up into the sky like big kites. While you're under tow, it's like riding an elevator, and VERY nose up pitch. Hitting the clamp release and going airborne out of the bed of a moving truck is an experience.

      As a safety precaution, there's a weak link between the tow line and the tow bridle. It's designed to break if the force gets too high, such as if you lose control and lock out. If you've flown a kite, this is where, in strong winds, they sometimes just decide to flip over and dive straight at the ground.

      Anyway, sometimes the weak link breaks for other reasons. I broke one once, crossing a wind shear boundary. When you lose the tow during dynamic ascent you're instantly in a very severe stall. It feels like free fall. It was one of the scariest things that's ever happened to me. You desperately want to *not* be heading for the ground, but the only way to recover is to pull in and dive to pick up speed.

      Fortunately my instructor insisted that we practice stall recovery at high altitude and in calm conditions.

    4. Re:Not an actual airline pilot, but... by Falconhell · · Score: 4, Interesting

      In those incidents of total power failure, (the most famous example being the Gimli glider, where a steep side slip was required to get the approach right) the pilots with gliding experience have performed very well indeed. A glider never has the option of going around after a bad approach, so the pilots are very well trained in using angles to judge their circuits. During training in most countries, pilots even have to fly several flights with all instruments covered, to protect against failures. At least here in Australia stall and spin training is mandatory, and tested each year at the annual flight review.
      Most power pilots have never done a full blooded spin, which is a violent and disorienting manoeuvre, due to practice, glider pilots respond very quickly to an incipient spin, and usually take action well before the full spin develops.
      Even during takeoff, we are actively expecting problems and have a plan should the tow rope fail for each stage.
      I would back the stick and rudder skills and energy management of a well trained glider pilot against any other form of flight training.
      Gliding is relatively cheap too, and in competition, many of the pilots are airline pilots who like to use real flying skills, and can afford the latest very high performance gliders.
      If I could choose, I wouldn’t fly with a pilot without gliding experience.

  11. Which is why North America is great by tlhIngan · · Score: 3, Insightful

    The freedom to fly is greatest in two countries - the US and Canada, where anyone suitably trained and documented can get in and fly. And the little Cessnas and Pipers aren't the highly automated Boeings or Airbuses, they're quite manual. If you're lucky, you'll have a partial autopilot - one axis (elevator) usually, fancy pants ones have two (elevator and aileron).

    One could fly from one end of the country to the other, avoiding controlled airspace and thus not talking to anyone if anyone so desired.

    Some other countries notably Europe, imposes fees on flying. Lots of taxes. Sure they have some GA flights, but to do so requires a fair bit of perseverance and money.

    The vast majority of countries though, make such thought of flight impossible.

    Now tell me which pilot will likely have the best flying skills? The one who can on a spare day roll up to their little airport, get out and fly, or one who can only fly the bit iron and the simulator because the only other flying is military..

    It's why I worried less about the US and Canada being late to the grounding party for the 737 MAX - the pilots here simply have better access to flight on their spare time than pretty much everyone else.

    The only way to maintain stick and rudder skills is to fly stick and rudder, and really, that's stupidly easy to do in the US and Canada - you literally get to the airport and fly - no permits, no flight plans (within limits), and other than fuel, no taxes to just punch holes in the sky.

    In China, they noticed this and the Chinese military let civilian pilots have a narrow slice of sky to fly between two airports in China. But you tell me - who will be the better pilot - one who has to put down their wings at the end of the day, or one who has access to the same sky they were just in?

    And no, you don't have to own your own airplane Renting is popular, as is renting an instructor.

  12. Re: I imagine this is going to get worse... by mark-t · · Score: 2

    Try doing that with an aircraft that weighs 40+ tons, not including the mass of the fuel.

    There's this thing called "inertia", y'see... and it's going to put an upper limit on the pilot's ability to subject the plane's occupants to sudden changes that might knock a person off balance while still actually having real control of the plane.

  13. Re:Fear of manual by john.r.strohm · · Score: 4, Interesting

    Actually, this is blatantly untrue.

    It has been well over twenty years, and probably closer to thirty years or more, since enough data had been amassed to prove conclusively that women were SAFER pilots than men.

    At the time, no one knew why. I suspect no one knows now. I saw one writer speculate, very carefully, that women MIGHT be a bit more risk-averse than men, to the point that they MIGHT make the determination that "this is getting just a bit too interesting for my taste" a bit sooner than a man might, exercise the Captain's prerogative to choose the alternate plan, a bit sooner than a man might.

    A lot of pilot training is teaching judgment and caution and knowing how to recognize that a situation is getting a bit too "interesting", for a suitable definition of "interesting".

  14. Re:US Airways Flight 1549 by dryeo · · Score: 2

    I also can think of a couple of accidents where pilots saved the day. I guess the real question is number of lives saved by pilots saving the day vs number of lives saved by automation.

    --
    https://en.wikipedia.org/wiki/Inverted_totalitarianism
  15. Plane must *NOT* over-ride pilot by knorthern+knight · · Score: 2

    Case 1) Before "The Miracle on the Hudson" there was "The Miracle at Gottrora" https://www.youtube.com/watch?... Executive summary...

    * Scandanavian Airlines flight 751 (MD-80) took off
    * The wings were not properly de-iced by the ground crew.
    * Clear ice broke off from the wings and was ingested by the 2 engines, located at the tail end of the plane.
    * The left engine caught fire, which was extinguished, but it was basically dead
    * The crew reduced throttle to limit further damage to the one functioning engine
    * With a competent pilot, the plane should have been able to limp in back to the airport, on one engine, for an emergency landing
    * But at this point the ATR (Atomatic Thust Restoration) algorithm kicked in. It basically said, "thou shallst not reduce throttle whilst climbing during take off".
    * It tried to to get 2 engines worth of thrust out of the one remaining engine.
    * Not a good idea. Just like redlining the shit out of your car engine, this quickly destroyed the remaining functional engine.
    * The crew managed to dead-stick crash-land in a snow-covered field without killing anybody.
    * Of the 129 people on board, 25 were injured, 2 seriously, but no fatalities. That's why it was called the "Miracle at Gottrora".

    Case 2) As for the 737 MAX, at worst, MCAS should sound an alarm, it should *NOT* over-ride the pilot, and drive the plane nose-down into the ground. If the 737 MAX can't be easily flown by a competent pilot without the MCAS band-aid, the 737 MAX should have its airworthiness certificate revoked.

    --

    I'm not repeating myself
    I'm an X window user; I'm an ex-Windows user
  16. It's not the only profession so affected. by az-saguaro · · Score: 2

    I am not a pilot, but I recognize concerns that effect other life-and-death professions and skills. The wisdom of the comments above is that automated flying is safe and efficient, but it can go wrong, and when it does, somebody with skills needs to be at the helm. But, if automation robs pilots of hands on experience, their knowledge to handle the emergencies is compromised. The focus in pilot training should be to ensure that they get the hands on flying time.

    My anesthesiologist colleagues always describe their profession as hours upon hours of sheer boredom punctuated by moments of sheer terror. Automated ventilators, monitors, and iv fluid pumps regulate most of the mundane tasks, making modern anesthesia one of the safest things you can possibly do. But, when something goes wrong, no system of automated controls or AI powered technology can take over for the seasoned anesthesiologist. It's a matter of life and death, not 150 lives at a time as on a 737, but one life at a time, which can add up.

    As a surgery resident, I learned the tried and true statistics that hernias are mostly a natural condition, and that post-surgical incisional hernias were infrequent. When they did happen, they were usually for legitimate reasons, and later on, they could be fixed, reliably, by the same experienced surgeons who knew how to avoid them in the first place. Then, circa 1990, surgeons started to use the laparoscope to do abdominal surgery without incisions. Mostly, that has been a huge benefit to the public, reducing hospital lengths of stay, minimizing many conventional surgical complications, and making once difficult operations safe and effective. The downside though is that surgeons have since forgotten how to properly make an abdominal incision, and especially, and more important, how to repair it. Since 1990, there has been an exponential rise in the rate of incisional hernias, the rate of failed hernia repairs and recurrent hernias, and severe morbidity to accompany those bad results. Furthermore, instead of focusing on acquiring the lost skills, surgeons have turned to companies who manufacture a boatload of flawed plastic implants that have made the problems far worse. (Search if you will on terms such as "hernia mesh complications" or "hernia mesh lawsuit".) The problem has reached epidemic numbers, and the death rate from complications of that sort now almost certainly exceed the casualty rate from commercial air accidents. As someone whose career has become ever more focused on fixing those unnecessary messes, it is all too obvious that a new technology that is valuable for ordinary everyday simple affairs robs the professionals and practitioners of vital skills needed for the unusual and extreme problems.

    A while back I saw a short on TV, sad but true. A man and woman, professionals in business suits and brief cases, enter a huge high rise office building and start riding the long tall escalator from the lobby to the third floor. Half way up, there is a power outage. The escalator stops. They look around bewildered, and call out for "Help, is there anyone who can help us?" They knew not how to walk up or down the escalator, idiots of the technology era. Technology is wonderful. But when the lights go out or the software balks or nature doesn't cooperate, people need to have foundational skills. We as a society have made a mistake putting so much focus on teaching new exciting chi chi technologies, and forgetting the reasons for learning the ABC's, the 3-R's, and other essential skills. For a surgeon, sewing up an incision so it does not fall apart is a fundamental skill. People suffer when surgeons fail to do so. I am not a pilot, but I can appreciate that stick and rudder skills are of the same essential importance. Our educational systems need to be corrected to supply those essential skills.