back to article Aviation regulators push for more automation so flights can be run by a single pilot

Regulators are pushing the UN's International Civil Aviation Organization (ICAO) to examine ways of making single pilot operations the eventual norm in commercial flights. The area that I think is the most concerning is a pilot sitting on their own in the dark and tired at 3am body clock time for four hours with only text …

  1. Anonymous Coward
    Anonymous Coward

    Bank of mum and dad

    I know of someone who's parent paid for them to do the training to become an airline pilot. It's simply out of reach of someone who doesn't have a well paid job already or who has access to the bank of mum and dad.

    The airlines could try paying to train up pilots instead of expecting them to pay for their training themselves.... Then there wouldn't be a shortage of crew.

    1. Neil Barnes Silver badge
      Black Helicopters

      Re: Bank of mum and dad

      One problem is that recent(ish) US rule changes require a pilot to have 1500 hours *before* he can even sit in the co-pilot's chair, in spite of evidence suggesting that things have got worse, not better: in Europe you spend most of those 1500 hours actually doing the job, not e.g. crop dusting or even as pilot instructors. That means there simply aren't any pilots available, hence the airlines pushing the plane designers for more automation.

      Mentour did a recent Youtube in which he discussed this. https://www.youtube.com/watch?v=l83d_z3GPeo

      I agree; a better system would be one in which there is a training scheme within the airline - it would require something to keep the pilots once they are qualified, of course - but the US system makes it illegal.

      Personally, I look upon one-pilot schemes with disfavour and would be unlikely to fly if I knew it was happening; I would definitely not fly a no-pilot plane... I don't travel in anything if there isn't a driver with skin in the game (minor automated train systems excepted).

      1. An_Old_Dog Silver badge

        Air Force In-Cockpit Pilot Reductions

        Traditionally, pilots retiring from the Air Force were considered good hiring material by the airlines; they already had high in-cockpit hours. Many air forces are reducing the number of their in-cockpit pilots, as many previously-manned/womanned missions are being taken over by drones. Drone pilots do not receive the same, relevant, experiences which in-cockpit pilots received.

        1. MachDiamond Silver badge

          Re: Air Force In-Cockpit Pilot Reductions

          "Many air forces are reducing the number of their in-cockpit pilots"

          Getting as many hours is also more difficult in the military. If the pilot goes for a rating in the most advanced fighters, the flight time can be very limited. The F-35 costs something around $40,000/hour to fly. A full profile mission in a B-52 can run the Air Force a couple of million dollars in fuel. I'm not sure how much simulator time is worth to the commercial sector.

      2. Hairy Airey

        Re: Bank of mum and dad

        I came here to reference the very same Mentour Pilot video. As I said in the comments of that one I was surprised that he didn't say "1500 hours on the Xbox" because that movie seems to be the only possible explanation for this figure. Basically no-one wants to change it in America as they don't want to give the impression of reducing safety.

        Some of the European airlines are investing heavily in flight simulators (see the crossover Tom Scott/Mentour Pilot videos where Tom lands a 737 with no prior flight experience - "Good luck, we're all counting on you" was hilarious!)

        To misquote one of my favourite documentaries "never send a machine to do a human's job".

      3. MachDiamond Silver badge

        Re: Bank of mum and dad

        "One problem is that recent(ish) US rule changes require a pilot to have 1500 hours *before* he can even sit in the co-pilot's chair"

        I can see the airlines reluctance to take on trainees with no experience. The cost of the wash outs could be very steep. If they were open to taking on people that have a bit more than a basic rating such as instrument, jet or multi-engine, they would at least know the person enjoys being a pilot and has some background. Since indentured servitude is not allowed in the US and flight training might not be eligible for government guaranteed student loans, people other than mum and dad might be hesitant to put up the money for training.

        Mentor Pilot has a good YouTube channel and I have seen several others that talk about commercial pilot positions that aren't with the passenger airlines. Steve Kinevo flys single pilot in a TBM single engine executive turbo-prop and seems to make out ok. He also was (maybe still does) fly small cargo flights from Florida to the Bahamas in a Cessna Caravan. It seems like getting rated and flying small scale cargo can be a good way to build hours. It can be thought of as serving one's time in an apprenticeship. Who knows, going through the ranks might lead to an executive pilot position that's even better than driving a bus for the major airlines.

    2. NoneSuch Silver badge
      Mushroom

      FFS

      Pilot redundancy is one of the basic flight safety rules. FOR GOOD REASON.

      The number of planes that have fallen from the sky due to system malfunction is staggering. Removing a trained human and replacing it with a machine is moronic. It does not improve safety.

      They can't make robot cars work on roads yet and they want to try it at 30,000 feet with 300+ souls on board.

      https://aviation.globalincidentmap.com/

      1. Lars Silver badge
        Coat

        Re: FFS

        Yes there are questions.

        But I wonder about the relation between "human made errors" and "technical errors" causing accidents in the air.

        There is one thing to remember regarding automation, it will actually get better and better and learn, while we, meat bags, all start from scratch, and that goes for every pilot too.

        My choice would be to keep both pilots but to keep on advancing the automation.

        And that of course is where we are today.

        1. An_Old_Dog Silver badge

          AI to the Rescue?

          I presume you're talking about AI-based automation. Assuming it can learn, the more-important question is, "Is it learning the correct lessons, and not discarding (pruning) previous correctly-learned lessons?"

          Post-crash, we have boards of investigation, but I believe, unlike human pilots, AI systems have no way to input those conclusions and the lessons learned therefrom.

          1. Lars Silver badge
            Coat

            Re: AI to the Rescue?

            "I presume you're talking about AI-based automation".

            No, not really, but improving the software is like learning as it adds to the "previous" knowledge.

          2. This post has been deleted by its author

          3. SCP

            Re: AI to the Rescue?

            I presume you're talking about AI-based automation. Assuming it can learn, the more-important question is, "Is it learning the correct lessons, and not discarding (pruning) previous correctly-learned lessons?

            [Resubmitted due a grievous typo not spotted in the edit period]

            There has been quite a lot of research into adaptive and non-linear flight controls - it does not need to be based on AI. One approach (IIRC) was to seek to establish effective control by comparing the actual aircraft response to a high-fidelity model and using the discrepancies (along with other available information) to adjust how effectors were used - this gave a means of adapting to faults in the system.

            While many very interesting and useful things can be done a major difficulty is that such systems are very difficult to certify in the current regulatory framework (particularly that for high-assurance software).

            Caution is a prudent virtue when dealing with safety matters, but it should not totally stifle development. At one time digital flight control computers, and the software running them, were novel and a cause of concern - but are now fairly common and accepted technology. Generally the progress in flight automation has seen major improvements in flight safety; but that progress has often been at the behest of commercial aspects rather than altruism.

            1. Neiljohnuk

              Re: AI to the Rescue?

              If the reports of AI learning and from doing so wants to kill us are correct deliberately crashing a plane full of passengers into a metropolitan area for maximum ground/tower block kill as well potentially awaits. SKYNET exists, did the MoD name it that as a joke, or a warning?

            2. MachDiamond Silver badge

              Re: AI to the Rescue?

              An expert system of some kind can be a very good tool and might even prevent a pilot from committing a major error. The problem is that those systems aren't creative and if they make a bad decision due to poor data, everybody dies. If the tool complements human pilots to further reduce incidents, it's a good thing. It would be very bad to recover a black box and have the final data read +++Out of Cheese Error, redo from start+++

              My comparison is with automated systems driving a car and being claimed to be X many times safer than a human. Maybe that's true, but some of the mistakes these systems have made that have led to an accident or a driver was able to correct the fault are often errors that a human driver wouldn't make. I point to cars of a certain brand having a thing for crashing into emergency vehicles with flashing lights, HGV's that are blocking the road, etc.

        2. vogon00

          Re: FFS

          'it will actually get better and better and learn'

          Yes it will, but only for as long as humans are the ones doing the learning and expressing their knowledge and experience in the code. Even then, unless there is continuous transfer of knowledge and experience from the outgoing 'old guard' maintainers to the incoming 'new guys' us humans will forget *why* that bit of code does what it does in the wider system.

          Given the propensity for people to just have a go despite not understanding what they are about to do, IMO there has to be an assurance that the overall system and code-base are continuously, and thoroughly, understood despite the inevitable staff/dev turnover.

          AI is NOT the answer. AI is not, and never will be, capable of dealing with the 'edge cases' as they arise - you've only got to watch a few Mentour videos (Other providers are available, including CAA/FAA/NTSB 'wash ups') to realise that issues generally get solved by humans, not machines.

          If something has gone wrong, that's where I want a properly qualified pilot with some experience in charge. Also, two real pilots instead of one mean there are two heads involved, and there is the chance to talk something through with a peer, or at least someone different, first. The only thing AI has to converse with before taking action is...itself..or more correctly the programmer who created it, or the crap data it was trained on.

          If the stories about 'Olympic Airways 411' are true, the only thing that stopped a disaster was a bloke with experience - something that AI, and it's coders, cannot have (Especially as the human through the rules book out of the window..which AI would never consider doing!)

          When it comes to flying lots of people about, the best things to oversee that activity are other humans... assisted by automation, yes...replaced by it, no.

          OK, I guess supply/demand/costs will win and either single-or-zero pilot air travel will happen. When it does, I hope the CEOs of the airline and/or of the company doing the flight control software are forced to be on the certification flights and 'eat their own dog food'.

          1. SCP

            Re: FFS

            If the stories about 'Olympic Airways 411' are true, the only thing that stopped a disaster was a bloke with experience - something that AI, and it's coders, cannot have (Especially as the human through the rules book out of the window..which AI would never consider doing!)

            There are also tragedies like Kegworth where people got it wrong. Decisions on progress in the use of technology need to be based on all the data, not cherry picked cases.

            (In the Olympic case it is also claimed that human error led to the turning off of the water injectors)

      2. runt row raggy

        Re: FFS

        i'm not taking issue with the fact that two pilots in the cockpit is a good thing, but...

        autopilot and even autoland is significantly easier than self-driving cars. kids and dogs don't run out into the sky chasing balls. so i can't agree that "we can't make robot cars work" so it follows that planes are out of the question. autopilot has been a thing for decades, and full autoland since the late 60s. it's even trusted to land when pilots wouldn't otherwise be able to land.

        due to human factors, i would think we are closer to 0 pilots being realistically safe, than 1.

        1. Edward Ashford

          Re: FFS

          Trucks do occasionally drive onto the runway when they shouldn't. A friend was on an A320 landing at BFS when the pilot had to make an emergency go around.

          1. HenryCrun

            Re: FFS

            I can't see the A.I. getting out of this sad example from just a few days ago.

            https://avherald.com/h?article=5013c619&opt=0

            1. runt row raggy

              Re: FFS

              i don't think ai is the solution, but i agree there is a huge number of things that need to be addressed before certifying automated flights, or single-pilot flights. because a single pilot can be incapacitated, all the problems of 0 pilot flight need to be solved for single-pilot flight, and additionally, you need to solve for single pilot problems such as mental health. and that's my only point.

              btw, the failure in this case appears to lie with the meat computers on the ground who according to official reports are saying that fire trucks were not cleared for the runway, but clearly fire trucks were on the runway.

          2. MachDiamond Silver badge

            Re: FFS

            "Trucks do occasionally drive onto the runway when they shouldn't."

            There was just a recent accident like that. There was an emergency response training evolution going on when a vehicle drove onto an active runway and an aircraft crashed into it. It's not always possible to increase power and go around when the plane has slowed down and the remaining amount of space is too short.

      3. Ken G Silver badge

        Re: FFS

        Actually there's less to bump into at 30,000ft and systems don't fail that often, pilot error is more common. I still agree with having a human in control but automation should reduce their workfload as much as possible so they can be alert for problems as they arise.

        1. MachDiamond Silver badge

          Re: FFS

          "Actually there's less to bump into at 30,000ft and systems don't fail that often, pilot error is more common."

          The accidents that I can recall weren't during the cruise portion of a flight unless it was a mechanical failure except for one where the pilots were behaving badly with an empty plane. An expert system could be a good tool to help eliminate many pilot errors, but it's doubtful that it will be good to use all of the time. When I say "many errors" there aren't that many. Flying is far safer than driving but it's never a bad idea to keep improving.

    3. Anonymous Coward
      Anonymous Coward

      Re: Bank of mum and dad

      Most airlines will sponsor candidates once they have demonstrated a significant amount of proficiency, but it tends to be as a loan against future earnings with a future job in the airline. Getting a CPL is hundreds of thousands of dollars so not something most can do even with "the bank of mum and dad". Most candidates will have a bursary of some kind from a future employer (not always a passenger airline). Students can be forced to drop out for medical reasons outside their control) so it's an expensive risk for an airline to sponsor a student that then drops out.

      Flying a plane is not something that everyone can do with enough training, getting your PPL is a financial commitment similar to going to university for 3 years. Once you have a ppl you can get CPL sponsorship if your exam grades are good enough.

      1. werdsmith Silver badge

        Re: Bank of mum and dad

        A PPL on fixed wing is about $10k. Easily doable with a bit of savings.

        I did the self improver route to CPL/IR as it was known, many years ago. At the time there were people supporting training by driving taxis, remortgaging. I did it by instructing, glider towing and eventually taking a share in a plane. As I understand it the door is now closed on the self improver route. I didn’t continue to ATPL.

    4. Fonant

      Re: Bank of mum and dad

      As I understand it you can get an airline to pay for all your training. The problem then is that you're ~£100,000 in debt to them, AND they're your employer. Better to keep the debt separate from the employment, so you can switch jobs if needed.

      My son is looking seriously at commercial pilot training. 18 months of intensive training costs £100,000. But that gets you to the point of being able to earn £50,000 a year starting salary as a First Officer ("co-pilot") (at least in Europe) and much higher salaries as you progress.

      If you're keen enough to be a commercial pilot, you persuade someone to give you a Big Loan, which you they pay back as you earn. You have to be pretty sure it's the job you want to do!

      Mentour Pilot pointed out that the training-to-earning ratio was much better for a commercial pilot than for a surgeon...

      1. Anonymous Coward
        Anonymous Coward

        Re: Bank of mum and dad

        Yes and no.

        Most still want you to pay the training up front in case you fail it, and then you'll get it back over X years (about 10).

    5. usbac Silver badge

      Re: Bank of mum and dad

      Back many years ago, I was on track to become an airline pilot. I was paying for all of my own training. While I still lived at home, my parents could not afford to fund my training. So, since I had low living expenses, I worked a full time job, and spent most of my income on flight training.

      This was before the (moronic, in my opinion) 1500 hour rule change. This was a bunch of politicians needing to look like they would "do something", and they destroyed the future of the airline industry in the process.

      You need 250 hours to get a commercial license, which is what allows you to get paid to fly. In the old days, you could eventually get a right-seat job at some of the regionals with 300+ hours. You needed an ATP (which requires 1500 hours) to work for the major airlines (all airlines now).

      Now, there isn't nearly enough work to be found to fill in the gap between 250 and 1500 hours. Paying your way out of pocket for the full 1500 hours will cost you at least $275,000 at current aircraft rental rates. Flight instructing on the weekends will never get you there. Also, keep in mind that the regionals are notoriously low paying. Who is going to invest $275,000 to get an $22,000 per year job?

      I ran out of money just trying to get from the 155 hours I have, to the 250 mark. And, that was when I could have found a job at 250 hours (it's nearly impossible now).

      This is how I came to work in IT now.

      1. ElRegioLPL

        Re: Bank of mum and dad

        Under EASA, after the training you get a Frozen ATPL which many airlines will hire you as a cadet on. After 1500 hours it becomes unfrozen

    6. Anonymous Coward
      Anonymous Coward

      Re: Bank of mum and dad

      Some airlines do offer the cadet route where they bankroll the training but ... most want the value paid up front first and will pay it back over 10 years. So still a bit pointless

    7. Ken G Silver badge

      Re: Bank of mum and dad

      About 20 years ago, my flying instructor told me that a certain Irish based low fares airline paid £19,000 a year to a co-pilot and that they were expected to fund their multi crew training themselves. It's certainly not a way to get rich and it should pay better than a bus or train driver carrying the same number of passengers on a more predictable route.

    8. Binraider Silver badge

      Re: Bank of mum and dad

      Yep, it's just fantasy-money territory even for those with good/great jobs. Training to the point of being able to co-pilot an A319 is something of the order of 150k of costs (not withstanding living costs) - and then - you are still not likely to be able to land a job. It's too expensive to save up for even with a "great" job, and "drip-feeding it" doesn't work either.

      Military pilots are much rarer a thing. In the 70's and 80's the default for many was to do your tour of duty or two then a conversion course. Todays air forces are very much smaller, so that convenient pool to draw from is now much smaller. (Particularly, outside the US. The RAF is a tiny fraction of what it used to be).

      Demand for air travel is as high as ever, so, frankly, it's contingent on airlines to start offering training in meaningful numbers at levels that can be afforded by mere mortals. This is not a difficult thing to arrange - make the training cost repayable by years of service.

      Or just pay for it out of your profits / charge customers accordingly. It is, so to speak, the "cost of doing business".

  2. Furious Reg reader John

    Perhaps it is a common misconception, but I was under the impression that human pilot decisions had caused far more crashes and fatalities than machine pilot ones, despite the 737 Max catastrophes. Perhaps the safest option is no flight crew at all.

    1. Zolko Silver badge

      No, the safest option is to not fly at all. Which also improves the CO2 emissions. There, problem solved.

      And while we're at it, we could go living in caves.

      1. Stork Silver badge

        We should not have gone into the caves in the first place ( with due credit to HHGTTG)

        1. captain veg Silver badge

          Caves? The big mistake was coming down from the trees.

          -A.

          1. Lars Silver badge
            Happy

            I think we fell, if not we pushed each other down.

          2. Trigonoceps occipitalis

            Trees! The huge mistake was getting onto the beach.

            (Tips hat to DA.)

          3. Potemkine! Silver badge

            "Back to the trees!"

            == Bring us Dabbsy back! ==

      2. Anonymous Coward
        Anonymous Coward

        If God had meant us to fly we would never have been given railways ....

        1. Neil Barnes Silver badge
          Thumb Up

          Kudos for 'On the Slow Train' there.

          1. Tim99 Silver badge

            Link to Flanders and Swan with historical railway video here: YouTube - barleyarrish.

        2. Ken G Silver badge
          Windows

          As Edwin Fuller Torrey and Judy Miller wrote in "The Invisible Plague: The Rise of Mental Illness from 1750 to the Present", trains “injure the brain.” In particular, the jarring motion of the train unhinges the mind and either drive sane people mad or trigger violent outbursts from a latent “lunatic.” Mixed with the noise of the train car it could shatter nerves.

          It's the best explanation I've read for why there's always a nutter on public transport.

    2. Phones Sheridan Silver badge
      Mushroom

      Whilst that is true in normal operation, in the event of plane malfunction, the opposite is the case. Autopilot cannot cope, and simply offloads to the pilot to make the decisions, usually just with an alarm going off that it has just done so after the event.

      A month ago Mentour Pilot and Tom Scot did a collab video where Mentour Pilot Petter Hörnfeldt talked not-a-pilot Scot through landing a 737 twice, once using auto pilot and then manually with hands on stick. Hörnfeldt only had access to the radio, and the same data air traffic controllers on the ground see. The results were interesting, if predictable. https://www.youtube.com/watch?v=AbTDzPUDxqY

      In the air, a "Stop at all costs" policy in the event of malfunction like is seen (theoretically) in automated cars, wouldn't really work, unless you want planes to plummet from the sky. Quite how you would train an AI to solve the Trolly Problem in the air, or even to choose the least destructive method of getting down on the ground I wouldn't know.

    3. BOFH in Training

      Yeah, thats what I thought - that pilot error was a leading cause of incidents.

      I will be comfortable with a single pilot plane, especially if the backup is someone who can remotely control from the ground in case the pilot is unable to control the plane for whatever reasons.

      Maybe a few ground based control stations scattered around the globe with pilots on standby to take control in an emergency will be a much cheaper option and not negate the safety aspects much.

      1. ChoHag Silver badge
        Stop

        BOFH with Scars

        > I will be comfortable with a single pilot plane, especially if the backup [involves a computer]

        No.

      2. VoiceOfTruth Silver badge

        -> if the backup is someone who can remotely control from the ground

        With what kind of security? What kind of anti-jam system would it use?

        1. TheRealRoland
          Devil

          That's a problem to be sorted out in the second iteration, after this all has been put into place, with 'adequate' testing

          "What? Oh that? Wasn't a use case so we didn't test against that. Good, moving on."

      3. Androgynous Cupboard Silver badge

        "I want to die peacefully in my sleep like my father, not screaming in terror like his passengers" -- Bob Monkhouse

        As I recall there have been other cases historically like the German Wings horror. At least two in the cockpit at all times, thank you.

        1. Anonymous Coward
          Anonymous Coward

          peeing in a lucozade bottle?

      4. Scene it all

        You cannot fly an airplane in bad weather, with two engines out, remotely, with zero notice. Miracle on the Hudson had *seconds* to react to the bird strike. And I can't see anyone doing a "Gimli Glider" slipped approach in a big airliner remotely. The computer probably would not even allow such a thing, and the cockpit camera probably can't swivel to the side to deal with the fact that the plane is intentionally flying *sideways*.. In the Sioux City DC-10 crash in 1989 it took *three* pilots to land the aircraft when an engine explosion took out *all* of the hydraulic systems.

        Everyone who proposes this hair-brained scheme should be force to watch every episode of "Air Crash Investigation".

        1. Mike 137 Silver badge

          "The computer probably would not even allow such a thing"

          Although this detail was left out of the movie, Sullenberger actually commented that the automation fought his efforts to land safely on the water as he did. He memorably stated "There's simply no substitute for experience in terms of aviation safety".

        2. David Hicklin Bronze badge

          > Sioux City DC-10 crash in 1989

          And nobody has been able to do it in a simulator since

      5. Antron Argaiv Silver badge

        ground based control stations scattered around the globe with pilots on standby

        You know how I know you've never flown a commercial passenger plane before?

        Soon to be found at airline check-in kiosks:

        Select all aircraft you currently hold endorsements for: []

        1. Hairy Airey

          Airplane reference of course

          Really, I just thought they made the "Is there anyone on board who can fly this plane" announcement?

          1. captain veg Silver badge

            Re: Airplane reference of course

            Yes there is. The pilot. And the co-pilot.

            -A.

            1. Anonymous Coward
              Anonymous Coward

              Re: Airplane reference of course

              That's the point when the cabin crew hit the buzzer for "wrong".

      6. RPF

        Pilots are THE LAST LINE OF DEFENCE for when systems break down.

        Accidents very rarely just happen because the pilots mess up; there will be systemic (poor maintenance, Boeing MCAS...) and circumstantial factors such as weather, fatigue, etc as well.

        So while it may appear that pilot errors cause a lot of incidents, pilots head off HUGE numbers of potential incidents every day.

        This single-pilot thing is purely economically driven and as a very experienced military and civil aviator, I can tell you it's total bullshit safety-wise.

    4. Anonymous Coward
      Anonymous Coward

      Used to be human error, but lately the technology is getting so complex a frozen pitot tube will bring down a jet.

      Imagine a machine autopilot during the double engine flameout on US Airways 1549 that ended up in the Hudson River.

      Sully Sullenberger and Jeff Skiles saved everyone by thinking laterally and making choices no machine would ever make.

      1. John Sager

        The accident rate would go up from that type of accident, but they are very rare. However the accident rate due to pilot error would probably go down. What the benefit would be in saved lives is almost impossible to gauge, probably, and this also assumes that the introduced automation would deal with issues that pilots currently take in their stride.

      2. bazza Silver badge

        A frozen pitot tube will not bring down a modern airliner. The worst it can by itself do is confuse the autopilot which (at least on airbuses) will trip off and tell the pilots that it's confused. Pilots fixated with fixing confused computer systems instead of aviating (using the various other pitot tubes available to them) brings down airliners.

        This is probably a reference to the Air France Airbus that went down in the Atlantic. The conclusion of the excellent eventual report found that a pitot tube had frozen up, causing the autopilot to trip off, and the pilots then spent the rest of their lives trying to understand why the computer didn't work instead of flying the aircraft by the standby instruments on their dashboard.

        A number of human factors issues were identified related to an over reliance on and over use of automation. More practice at flying (something Sully and crew had in abundance) is what is needed.

        What is interesting about the Hudson miracle is how Sully knew he could rely on the Airbus alpha protection to automatically get the best slowest glide down to the river. Sully completely exploited that, and the gentle splash down they experienced certainly gave them all the best chances. For me this is an exemplar of safety engineering. Good design carefully thought out, highly skilled and knowledgeable operators (pilots) who knew what they had to do when the birds hit the fans, and more importantly also knew what they didn't have to do.

        1. VoiceOfTruth Silver badge

          -> The conclusion of the excellent eventual report

          That is called hindsight. With months of calm investigation, not the frantic minutes under extreme stress that the pilots experienced. It's very easy afterwards to say "they should have done this".

          1. ChoHag Silver badge

            That's why that is not what they do.

            They say "they should have done this; that is what contributed to them not being able to do this; we recommend urgent changes to all airlines' procedures to include less of that so that the pilots can do this more effectively".

            That's a lot harder to say, and they do an extremely good job of saying it.

            1. vogon00

              @ChoHag.Interesting handle you have there.

              Anything to do with this*?

              Had I not read Douglas Adams first, I'd be going by the name Belgarath rather than a vogon.

              (* Highly recommended escapist** reading. It also teaches you about people. You should also look at the other side of the coin by reading this, if for no other reason than 'equal opportunities'.)

              (** Actually entrapment - back in the day, I nearly got fired for not being being able to stop reading the entire series from start to finish.).

        2. Lars Silver badge

          @bazza

          That is a bit harsh. The least experienced pilot was "behind the helm" that night and as the system claimed the speed was too high he pulled back.

          The result was a slow stall towards the ocean. He did ask for help and the captain who had been sleeping understood what was happening but it was too late then.

          You find (not in their voices) what was said in the cockpit until the end, on YouTube.

        3. David Hicklin Bronze badge

          >. Air France Airbus that went down in the Atlantic.

          I think they also made reference to the sudden event of going from a smooth flight (OK thunderstorms were nearby) to the next instant the alarms going off and throwing control back to the pilots and expecting them to "just take over" without being fully situational aware. They then became focussed on one thing with one pilot pulling up whilst the other pushed down until it was to late.

          All they really had to do was to fly straight and level.

          Self driving cars have the same problem where the driver is expected to take over in an instant if the computer craps out.

    5. mutt13y

      two or zero are the only options

      Many accidents where pilot error is the cause were precipitated by errors in monitoring or automation that were then mishandled by the pilots.

      So if there is no pilot you do remove one of the last lines of defence.

      There are probably many incidents where the automation went crazy and the pilots just handled it as they were trained and so we never hear about it.

      That said there is a rule against having one person in the cockpit because of the spate of murdercides. So who is going to be there to make sure the one pilot does not intentionally crash the plane?

      1. Anonymous Coward
        Anonymous Coward

        Re: two or zero are the only options

        Yes there are many times pilots will just override autopilot. The most common being that ATC ask us to do something that we know none of the Autopilots can handle.

        Also, note that a plane doesn't have one autopilot that you turn on and it does all parts of flying the plane. An aircraft has multiple different autopilots that each do different tasks, part of the role of the pilot is selecting the right auto pilot for each phase of flight, and that often has to change to cope with differences between expectations on the ground and reality in the air (weather, other traffic, idiots with drones etc)

        1. Scene it all

          Re: two or zero are the only options

          And even the most advanced automated landing system, the "Cat III" system installed at major airports, that can bring a plane right onto the runway in zero visibility, will not work at all if the crosswind exceeds a certain limit.

          1. werdsmith Silver badge

            Re: two or zero are the only options

            And even the most advanced automated landing system, the "Cat III" system installed at major airports, that can bring a plane right onto the runway in zero visibility, will not work at all if the crosswind exceeds a certain limit.

            This applies to human pilots too. If conditions are too bad to land at your destination airport, then divert to another airport that has an into wind runway and better conditions.

            1. Anonymous Coward
              Anonymous Coward

              Re: two or zero are the only options

              It does, but we can land planes in much rougher conditions than an autoland system can.

              1. Ivan Headache

                Re: two or zero are the only options

                I’m guessing that there’s no auto-land system at Madeira Funchal airport..

                1. werdsmith Silver badge

                  Re: two or zero are the only options

                  Aircraft that can't land at Madeira - which is up to road from Funchal at Santa Cruz and named after a footballer CR7, can go to Porto Santo, The Canaries or Marrakesh. They need to have sufficient contingency if the runaway is closed.

                  Auta landing systems are for visibility minima, not wind conditions. Pilots going to Madeira need special training so I would say that this route would not be one that is up for automation.

      2. bazza Silver badge

        Re: two or zero are the only options

        >There are probably many incidents where the automation went crazy and the pilots just handled it as they were trained and so we never hear about it.

        Not really. Each incident is reportable, they all get reported and the reports are publicly available.

        But I agree, such events do indeed occur, and the reports makes for sobering reading! Qantas had one such incident with an Airbus, which I don't think was ever satisfactorily resolved or fully understood.

    6. Mike 16

      Pilot Error

      I suspect that a major reason for so many crashes being put down to Pilot Error is that the pilot is dead and anyway cannot afford the legal team to go against the army deployed by airlines or manufacturers to scapegoat them.

      1. stiine Silver badge

        Re: Pilot Error

        Mike, you would be wrong. The point of the investigations is to determine the cause of the accident. When it turns out that the pilot did X when training indicated action Y, then the pilot gets the blame. If the training didn't cover this well, or at all, the the training is at fault.

    7. Lars Silver badge
      Coat

      @Furious Reg reader John

      My problem and the reason I wonder about it, is perhaps that I have had the habit of looking at all those "air crash" programs and some of the human errors are indeed stunning.

      Often, of course, it's a combination of both, but still, take the plane where one engine caught fire but the pilot shut down the working engine,

      Or the three pilots trying to change a tiny light bulb in the cockpit and forgot they had not switched on the autopilot and ended in the Everglades, and so forth, and what about the suicides.

      I would claim pilots are not keen on deleting automation, or as Juan Browne on the Blancolirio channel said about the difference in approach to automation between Boeing and Airbus (having experience flying both brands), Airbus has the better approach.

      1. ChoHag Silver badge

        Said with the calm assurance of somebody who's never had to deal with any pressure, ever.

    8. Ken G Silver badge

      That's the psychological barrier referred to in the article. Yes, it probably is as safe or safer but it's a hard sell to get passengers on an unmanned aircraft knowing there will be one death per 5 million boardings even though that's about the same as piloted aircraft over the past 30 years.

    9. tip pc Silver badge

      737 max issues where due to Boeing not wanting training and badly implementing critical software feature while not telling anyone it was there.

      The failure was in design not the meatballs driving the plane.

  3. Anonymous Coward
    Anonymous Coward

    Well Muskie's fixed Tesla, fixed SpaceX, fixed Twitter, now what else is there to do?...

    "This is Captain Clarence Oveur. Hope you're enjoying Musk Air flight 767. I'm a little busy at the moment... please leave your message after the tone and I'll get back to you"

    1. Anonymous Coward
      Anonymous Coward

      terrible straw man.

      you'll be blaming conservatives next.

      1. Anonymous Coward
        Anonymous Coward

        As a conservative, I thought it was fucking hilarious. Who cares if its true, or even likely, if it causes liberals to crap their drawers.

  4. Anonymous Coward
    Anonymous Coward

    If you work in a shop or an office it is required by law for there to be two people in case the other one has an accident or emergency. Now I try to imagine the same shop or work place carrying 100's of people and hurtling through the sky at great speed that could potentially crash land anywhere because the one person that can land it safely is no longer available. What could possibly go wrong?

    You could argue well if they can do automatic landings then it's not a problem. I won't trust a computer to drive a car and I'm certainly not trusting a computer to land a plane I'm on that could hit any number of variables. Next you could say drone pilot or remote control. Makes sense except drone pilots fly these tiny little planes so I'm also certainly not trusting someone miles away using whatever communication system that can't feel the plane and adjust.

    You can do it trains sure. It's on a track it'll just stop and so will all the trains behind. (Obviously doesn't always happen but the % is very very low)

    It's madness I tells ya and all to try and save money... I hope someone has some common sense and tells to jog on.

    1. KarMann Silver badge
      Headmaster

      When you say things like 'required by law', you kind of have to specify which law, where. I've never experienced such a law anywhere I've worked, whether due to lack of such a law or lack of enforcement, I know not. But don't just assume that that goes for everyone.

      ETA: The use of 'jog on' suggests you're British. I've only worked remotely in (from) the UK, which would explain why I haven't noticed that law/regulation. In the US, I've been alone in shop/office plenty of times, and I'm pretty sure there's no law/regulation/rule against it there.

      1. Anonymous Coward
        Anonymous Coward

        It sits under,

        Health and Safety at Work Act 1974.

        Management of Health and Safety at Work Regulations 1999.

        You can work alone however unless an assessment says it is safe to do so you can't. Sure they can make you but if something happens then it's on them. So technically working in an office would not be safe. What if you fall over and badly injure yourself? Fall down the stairs? If you are in a shop what if you are manual handling stock and you injure yourself? Both these examples mean that yes I was wrong that it isn't illegal however employers don't do it unless they want to take that risk.

    2. VoiceOfTruth Silver badge

      -> If you work in a shop or an office it is required by law for there to be two people in case the other one has an accident or emergency

      It is not. You have repeated a widely-held but wrong belief.

      Employers ARE required to assess the risk to lone employees (as they are with other employees). If the risk to the lone employee is deemed unsafe just by them being alone then the usual remedy is to have another employee present.

      There may be circumstances or types of jobs which always require more than one person present purely for safety reasons.

      1. werdsmith Silver badge

        There are many lone worker solutions available found by a simple search engine.

      2. DS999 Silver badge

        There are a lot of convenience stores in the US where it is almost always one person there. They may add a second during morning & evening rush if they are somewhere frequented by commuters, but the typical is one person behind the register. I assume if they need to visit the restroom they lock the door with a "be right back" sign and hurry!

        Many small shops are similar, though they may be open for limited hours and have only one employee (often the owner)

      3. David Hicklin Bronze badge

        >> If you work in a shop or an office it is required by law

        It may not be the law but in general it is undesirable and should be avoided if possible (assuming your employer cares about you of course).

        I have been in situations like this, usually late at night trying to fix some broken IT kit but my manager did phone me every 30 minutes or so to check I was still OK (and was it fixed yet!)

        1. swm

          When I was at college (50+ years ago) I was working in the computer center. The only other person was the computer operator. I heard a funny noise and I investigated. The computer operator was on the floor outside of the computer room barely moving. Since it was break time I wondered who to call.

          I called the campus police who drove up in their station wagon which had a stretcher in the back. The campus police stayed with the computer operator while I was busy trying to get the stretcher out of the campus police station wagon. A dean saw me and thought I was trying to rip off the cruiser and came running over (just as I planned). When the dean got to me, he recognized me. So I said, "Good, you can help me with the stretcher."

          Back inside we loaded the computer operator onto the stretcher, put him in the cruiser and drove to the hospital (about a block away. The doctor on-call diagnosed it as a stroke and decided to do nothing (correct decision) but admitted the operator to the hospital. This resulted in all of the operator's possessions being locked up including all the keys to the tape library etc. Next morning there was some confusion as the computer personnel tried to retrieve the various keys. The computer operator did recover fully.

          If I hadn't been there the operator would have laid on the floor until the next morning.

    3. tip pc Silver badge

      You could argue well if they can do automatic landings then it's not a problem. I won't trust a computer to drive a car and I'm certainly not trusting a computer to land a plane I'm on that could hit any number of variables.

      they've been researching auto land since the 1940's with the first demonstration on an aircraft on 3 July 1950 right here in Blighty.

      https://en.wikipedia.org/wiki/Blind_Landing_Experimental_Unit

      https://en.wikipedia.org/wiki/Autoland

      I'd suggest that auto land is far more developed than auto drive and auto land likely has less variables than auto drive which has uncontrolled space all round to deal with.

      if all vehicles where auto driven and nothing else allowed on the roads (no people, pets, animals etc) then auto drive would be easier to implement. 99.9% of the time airliners are flown in controlled spaces and it'll be known that nothing is In front of it to be crashed into.

    4. David Hicklin Bronze badge

      >> You could argue well if they can do automatic landings then it's not a problem

      You will then also hit the other extreme that already exists with the current automation: the pilots do not spend enough time actually flying the plane to stay proficient

  5. tip pc Silver badge

    2 Pilots for a reason

    There are 2 Pilots for similar reasons like why there are 2 engines, design diversity for FADEC, Autopilot and other critical systems etc.

    They say its cheaper to have several computing devices than a human but those finances will turn on their head once the public stop flying those airlines.

    if £150k for a meat sack is too much for an airline to afford then they shouldn't be in the airline business.

    1. Yet Another Anonymous coward Silver badge

      Re: 2 Pilots for a reason

      But that only applies to planes above a certain number of passengers.

      The single engined float planes that take off from the water in a busy city center, surrounded by high buildings, mountains and suspension bridges, in an uncontrolled airspace full of private pilots, police/news/medic helicopters. These only need one pilot because they have < 16 seats.

      I suppose this is much easier flying job than taking a 737 from one Cat5 airport to another Cat5 airport.

      1. steelpillow Silver badge
        Boffin

        Re: 2 Pilots for a reason

        It's more about sustaining pilot presence and alertness on a long flight. Small planes take short hops and a pilot will cry off the flight if he feels unwell. Big planes fly for hours, pilots need a break periodically and are more likely to become unwell during the flight.

    2. CrackedNoggin Bronze badge

      Re: 2 Pilots for a reason

      Before the public stop flying there will be loads of lawsuits.

      Training the pilots might be cheaper.

  6. Rameses Niblick the Third Kerplunk Kerplunk Whoops Where's My Thribble?

    The requirements for a full flying license are also incredibly onerous, which creates a bottleneck in the supply for qualified pilots. For most European airlines, you need 1,500 hours flight time before you get a full license. Until then, you're on provisional terms and need a fully qualified pilot operating alongside you.

    So coming up with a solution which eliminates the primary training opportunity (ie co-piloting) is the way to train more pilots?

    1. JohnTill123

      I don't think they are interested in training more pilots. They are interested in cutting costs at all costs. Training pilots is someone else's problem.

      1. Neil Barnes Silver badge

        And of course far better from the company's point of view for liability: if they trained the pilot and he screws up, it's their fault; if someone else trained him then 'Well, he was qualified, guv!'.

        1. tip pc Silver badge

          'Well, he was qualified, guv!'.

          nope, its up to the airline to ensure the pilots are fully trained and certified in that type of aircraft. That's one of the things with the 737 Max crashes, Lion Air (I thought it was Ethiopian airlines) wanted training for their pilots and Boeing moaned about it and tried to put them off.

          https://fortune.com/2020/01/14/boeing-lion-air-extra-737-max-pilot-training-simulator-crash/

    2. Neil Barnes Silver badge

      Rameses, the US has already come up with that solution; see my post above.

  7. Anonymous Coward Silver badge
    Black Helicopters

    Simple solution

    Train the co-pilot to also be able to push a drinks trolley and advertise scratch cards. That way they can be in the cockpit for the critical phases of flight, but in the cabin for the cruise phase. If an emergency were to happen in cruise, (s)he's just got to get through the door and can get back to being a (co-)pilot.

    Simples.

    1. ChoHag Silver badge
      Joke

      Re: Simple solution

      This is the best sign yet that the vultures are circling their own corpse.

      Did 5 people take this comment seriously?

      Belated icon for the Americans on AC's behalf.

      1. Apollo-Soyuz 1975
        Windows

        Re: Simple solution

        In fairness, we Seppoes have to hear people make comments like that earnestly… in real life.

        It’s all so tiresome.

  8. nautica Silver badge
    Boffin

    Next up: get rid of that pesky extra engine.

    The requirement for two pilots in the cockpit has always had an over-riding psychological component which no amount of automation can provide. The pilot simply must have someone, another expert---physically present---to talk to; to discuss---among the more trivial topics---life-and-death situations.

    Of course, since there's a way to insert a computer/computers into this argument/equation, the 'Single-Pilot' scenario will eventually win out, much as did Boeing's ability to completely design a brand-new aircraft, using ONLY software, and convince (would "...pull the wool over their eyes..." be not strong enough?) the powers-that-be that this approach is perfectly fine; that nothing could possibly go wrong.

    --------------------------------------------------------------------------------------

    Why do you think that single-pilot General Aviation crashes are so common?

    1. Yet Another Anonymous coward Silver badge

      Re: Next up: get rid of that pesky extra engine.

      For the same reason that motorbike accidents by young men on the road are so common?

      1. Anonymous Coward
        Anonymous Coward

        Re: Next up: get rid of that pesky extra engine.

        The importance of multiple pilots in preventing accidents is covered in the human performance exams for PPL. Essentially it's easy for a single pilot to misunderstand a failure due to confirmation bias etc. Two pilots using CRM are less likely to pick the first root cause they think of, then subconsciously select evidence to support that.

        1. I ain't Spartacus Gold badge

          Re: Next up: get rid of that pesky extra engine.

          Who's going to read out the checklist, while the flying pilot pushes the buttons? Will it be an automated voice on an iPad? Who runs the radios while the pilot is flying the thing?

          How much money is going to be wasted by planes screwing up in the airspace around airports, because the single pilots have to manage talking to ATC and going through the landing checklists while also working out what holding pattern they're in.

          Not to mention splitting up the tasks during an actual emergency.

          If we're going to do this, we should at least have every airline ticket entered into a free prize draw. The prize being that passenger gets to be copilot for the journey. They can at least keep the poor pilot company, if nothing else.

  9. trevorde Silver badge

    Press release from Tesla Air

    Using the experience from our car division, we are just *months* away from a full, self-flying aircraft. Just as soon as I've sorted out Twitter.

    Elon Musk

  10. Anonymous Coward
    Anonymous Coward

    QF32

    Not sure about this. Look at QF32. 4 pilots on board. 2 normal crew, a check pilot and a check check pilot. Possibly one of the most experienced crews in aviation history.

    Even then, the workload was intense. With only one pilot this could have ended very differently.

    1. druck Silver badge

      Re: QF32

      Great example. The computer couldn't even work out the landing distance by itself, without those skilled pilots there would have been 469 fatalities.

  11. Adam Foxton

    Boeing

    A poster child for why we need very tough Corporate Manslaughter laws with the punishments at the top levels of the corporation.

    1. Antron Argaiv Silver badge

      Re: Boeing

      Not only the people at the top. The Corporation, itself, needs to be punished.

      I'll believe corporations are people when one of them goes to jail.

  12. Richard 12 Silver badge

    They're literally insane

    This is possibly the most stupidly self-destructive and dangerous idea I've ever read.

    Sure, a single pilot can fly the sector when everything is going perfectly.

    But when something goes wrong, if there's only one pilot, everybody dies a lot of the time.

    "Pilot error" almost always means "lost situational awareness". Piling on a greater workload makes that far more likely.

    Just read the AAIB reports for "recreational" pilots.

    Then there will be even fewer pilots, although I suppose that won't be a major problem because nobody will want to fly at all.

    1. Horst U Rodeinon
      Flame

      Re: They're literally insane

      Do not forget the imperative that the Machine, i.e., aircraft, must be seen as infallible or many people would not even consider boarding one. That is the sole reason so many accidents are written off to "Pilot Error" even when the evidence suggests otherwise.

  13. VoiceOfTruth Silver badge

    It's the cost, nothing else

    -> There are two obvious drivers for the proposal – cost cutting and crew shortages

    No. It's just cost cutting. It is nothing else. Don't be fooled or misled.

    1. An_Old_Dog Silver badge

      Re: It's the cost, nothing else

      "Crew shortage" == "People are unwilling to do the jobs we have on offer, at the rates we're willing to pay, under the working conditions we have."

    2. I ain't Spartacus Gold badge
      Coat

      Re: It's the cost, nothing else

      There are two obvious drivers for the proposal

      I thought the whole point of the proposal was to only have one driver.

      ...I'll get my coat. The one with the parachute strapped to it please...

  14. Locomotion69

    Madness

    Three basic rules of flying an airplane (in this order):

    1. Aviate - keep the thing up in the air and under your control

    2. Navigate - determine where you are, and where you are going

    3. Communicate - tell traffic control/others what you are doing/going to do

    For 1 and 2, modern systems are great help, and in most cases better equipped to do the job then you are.

    But if one of these systems goes astray for whatever reason, it is all up to that single pilot in the front seat.

    Multiple pilots can divide 1 till 3 among them in emergencies, making them focused on the job, which is saving the lives of themselves and everybody behind them,

    And emergencies WILL happen.

    1. Yet Another Anonymous coward Silver badge

      Re: Madness

      That's why cold-war Soviet airliners with a separate navigator and radio operator in addition to the pilots were so safe.

      1. Jan 0 Silver badge
        Devil

        Re: Madness

        I think you'll find it was more related to how accessible the on board vodka was.

      2. MachDiamond Silver badge

        Re: Madness

        "That's why cold-war Soviet airliners with a separate navigator and radio operator in addition to the pilots were so safe."

        Sort of. The navigator was working at a plotting board with a map, stopwatch and circular slide rule. The radio operator was dealing with enough buttons and knobs to make a ham radio enthusiast salivate. The radio operator was likely feeding navigation beacon signal information to the navigator when available or trying to get a fix under direction of the navigator for an expected beacon.

        Modern radios just need the frequency punched in. Navigation is often by GPS with an inertial system (IMS) along with beacons feeding a computer that determines the best fix should one input drop out. Flight profiles are fed into the FMS (Flight Management System) so the plane does most of the navigation on its own. If ATC modifies the profile enroute, the pilot/co-pilot changes the program.

        There used to be a Flight Engineer that monitored the engines and fuselage until that become far more automated and flight crew was reduced to a bare minimum of two.

  15. Snowy Silver badge
    Coat

    Single point of failure

    With only one pilot what you have is a single point of failure, it is one of the reasons the pilot and the co-pilot generally do not eat the same food.

    <quote>Pilots and co-pilots are advised not to eat the same meals when they are working. If something is wrong with the meal (like food poisoning), the other pilot will not be affected and can take over. The rule is not mandated by the Federal Aviation Administration, but most airlines have their own rules about it,/quote>

    1. Throatwarbler Mangrove Silver badge
      Happy

      Re: Single point of failure

      As so aptly demonstrated in the seminal disaster documentary Airplane!.

      1. I ain't Spartacus Gold badge

        Re: Single point of failure

        Surely that documentary also showed the general unreliability of automatic systems. Which are liable to both deflation and self-combustion. As well as making unwanted advances on the cabin crew during flights.

  16. Electronics'R'Us
    Holmes

    Automation

    "As a computer, I find your trust in technology amusing".

    That is a very old line but it is just as true today. Technology fails. Regularly.

    For flight safety critical avionics, there are duplex (2) and triplex (3) architectures. The duplex approach requires the pilot to take over when the channels disagree (because with only 2, we don't really know, from an automation perspective, is incorrect.

    Even triplex systems can fail [1] if multiple sensors fail, as was the case with AF447. I remember looking at the telemetry and there was 'airspeed disagree' (so at least one airspeed sensing channel was not operating properly) followed by 'alternate laws' [2] which gives the system extra flexibility in operation in a difficult environment. When the system could not get agreement on at least two channels (and therefore had no reasonable way of knowing the real airspeed) there was nothing it could do - it required the pilots to actually take over with manual control

    A lot of pilot error has to do with extremely confusing errors and warnings, especially at low level / low speed. The checklists are designed for use where there is plenty of time (and height) to get to the bottom of a problem and even then there is cross checking by the crew and a strict division of duties (one pilot goes through the checklist to identify the issue and the solution while the other pilot keeps an eye on actually flying the aircraft is one scenario).

    Single pilot with a duplex system would have major overload as soon as something went wrong and a triplex system would be required for all safety critical systems (there are quite a lot of them in a modern aircraft).

    Failure rate is a numbers game and the usual threshold is that the probability of catastrophic failure is less than 10^^-9 per flight hour. Doesn't mean it cannot happen, just it is very unlikely. The corollary to Murphy's law is 'Murphy was an optimist'.

    All these systems have a lot of complexity (the channels have to be galvanically isolated, for example) so there is always a nagging suspicion that something has not been thoroughly analysed.

    [1] The failure of multiple pitot tubes on AF447 actually highlighted a known human factors problem with automation- over reliance on it. In that case, it was determined the pilots did not actually know how to fly the aircraft.

    [2] The control laws are what are used by the automation to determine safe changes to the aircraft flying profile so it does not exceed the safe operational envelope.

    1. Scene it all

      Re: Automation

      AF 447 is interesting. Even with the frozen pitot tubes, if the pilot flying (a very under-trained first office) had simply sat on his hands and done nothing, nobody would have died. The Airbus has multiple levels of degraded ability and would have continued on in level flight. Instead he panicked and did exactly the wrong thing because nobody had ever taught him how to fly an airplane instead of how to fly a computer. And that is the fault of Air France management. Maybe the managers are the ones who should be replaced.

    2. bazza Silver badge

      Re: Automation

      The F117 is interesting, has a quadchannel design (one up from triplex, I just don't know how to say it: quadrex?!). Something to do with taking battle damage I think.

      Mind you, that airframe needed all the help it could get to be able to fly at all!

      1. Neil Barnes Silver badge
        Headmaster

        Re: Automation

        Quadraplex, I think.

      2. SCP

        Re: Automation

        I just don't know how to say it

        Quadruplex. Typhoon FCS is also quadruplex.

        1. Electronics'R'Us
          Holmes

          Re: Automation

          Strictly speaking, Typhoon is a dual duplex system. There are 2 channels controlling one of the Canards and another dual channel that controls the other Canard and rudder.

          Disclaimer - I have worked pretty extensively on that system for the design authority.

          1. SCP

            Re: Automation

            Not as I recall it. Quadruplex on all primary flight control surfaces (foreplanes, inboard/outboard flaperons, rudder). Dual duplex on secondary surfaces (Leading Edge [I think] and engine cowls).

  17. SloppyJesse

    50% reduction in pilot fatalities

    "EASA's Safety Risk Assessment Framework for Extended Minimum Crew Operations (eMCO) and SPO aims to address the following points:

    [...]

    Pilot incapacitation"

    How does one address that? Fly in circles until they feel better? Or Hollywood style aircraft to aircraft transfers to get another pilot on board?

  18. Kev99 Silver badge

    Looks like the fools who've never had to do the job think they know what's best. It's one thing for a truck / lorrie to have a single driver but quite another for an airplane. If something goes cattawumpus on the ground one can at least pull over. If it happens at 30,000 feet it becomes put your head between your knees and kiss your sweet arse good-bye.

    As an aside, my brother once let me sit in the right seat in his Beechcraft Bonanza. Even with both of us watching the umpteen hundred miles of airspace we almost met another airplane coming at us. The other guy ignored our flashing the landing lights until he was less than a couple miles away. I'd rater not experience a computer failure under the same circumstances in a commercial airplane. It's probably more a matter of the airlines wanting to save money to keep their own back pockets and those of wall street and the various bourses fat.

  19. Anonymous Coward
    Anonymous Coward

    single pilot

    If your job duties ever involved shuffling backup tapes between drives and storage (or if you served in almost any of the western militaries) you're familiar with the phrase "Two is one and one is none!".

    We can talk about single pilot ops when zero pilot operations are routine, and having a meatsack pilot is needed for emergency standby only. Even then, it might not be a viable idea (given human reaction times to develop situational awareness after lengthy stretches of having no input into the process)

  20. AdamWill

    don't worry, Boeing's got it sorted

    "Boeing Southeast Asia president Alexander Feldman told a Bloomberg business summit in Bangkok last week. "The technology is there for single pilots, it's really about where the regulators and the general public feel comfortable.""

    A Boeing president telling us everything's fine? Well, I for one am reassured!

    1. bazza Silver badge

      Re: don't worry, Boeing's got it sorted

      I'm so reassured I'm going get on an Airbus instead, so I can get a grandstand view of the Boeing in operation. It must be wonderful, a sight to behold.

  21. Boris the Cockroach Silver badge
    Mushroom

    Where do they

    get their ideas? beancounter central?

    I'm responsible for making (among other bits) some aircraft landing gear parts, once the computer has swallowed the CAD model, the CAM software has generates a program(helped by yours truely), I've built the fixture and tested it all out to make sure it all works, I am not allowed by the company rule book to be anywhere near the inspection department while they are checking it.

    On the basis, if I've made an error , then hopefully they'll pick it up.

    If something as simple as I make has to have a 2 man rule, I'd damn well hope that something as complex as a commecial jet has a 2 man rule too.

    Icon .. for what happens if people involved with aircraft get it wrong...

  22. Spazturtle Silver badge

    I believe the push if for single pilot cockpits, but still having multiple pilots. Currently from transatlantic flights you need pilots in the cockpit and then another crew for the 2nd shift who spend the first half of the flight sleeping. Transatlantic flights also spend most of their time in un-congested airspace with usually little activity. With single pilot cockpits you would only need one pilot awake at a time but you would still two on board and in an emergency the second pilot could be woken up.

    Short hops are harder to switch to being single pilot as they spend most of their flight in congested airspace over populated areas and multiple pilots help with managing the workload.

    1. Richard 12 Silver badge

      In an emergency, by the time the second pilot has woken up, everyone is sleeping again.

      Permanently.

      There are multiple examples of the second crew walking in on an emergency with only enough time to utter "Oh crap" before the airframe interfaces with the ground (or water) at speed.

  23. steelpillow Silver badge
    Boffin

    Good luck with that

    Try and get this past the national trades unions. I can see BALPA, the British Airline Pilots' Association, on all-out strike if our airlines tried it on.

    Back in the day, when Boeing introduced the 747 Jumbo Jet, the airlines tried to pay the pilots the same as before. The pilots argued that great passenger numbers bring great responsibility and demanded equivalent pay increases. After a series of damaging strikes, the industry caved in. The joke arose, "Why does the 747 have a hump over the cockpit?" "I don't know, why does the 747 have a hump over the cockpit?" "So the pilots can sit on their wallets without banging their heads!"

    As others have pointed out here, there are many reasons why a long-haul jet needs two pilots until the day when it doesn't need any. The pilots' unions will be well aware of their members' views on this. One might as well put out a new standard for horse feathers.

  24. nautica Silver badge
    Boffin

    Learning to fly: no problem.

    “The Guide says there is an art to flying", said Ford, "or rather a knack. The knack lies in learning how to throw yourself at the ground and miss.”

    ― Douglas Adams, Life, the Universe and Everything

  25. Horst U Rodeinon
    Alert

    For the hours upon hours of boredom...

    when everything is going as you want, even zero crew in the cockpit may be adequate. However, if even one edge case occurs (a.k.a., moment of stark terror) and the machinery hasn't been programmed to recognize and deal with it, the outcome will be as welcome as a turd in a punchbowl.

    The automated systems also fail and they may not be able to recognize or accept they are the cause of the problem.

    As with all systems, there is an optimal amount of backup capability which should be required. Bean counters should have no part in determining what it is.

    1. Anonymous Coward
      Anonymous Coward

      Re: For the hours upon hours of boredom...

      Damn beancounters run the whole world. This is why the last two years have been like they were.

  26. Eclectic Man Silver badge

    Tedium

    In John Finnemore's excellent comedy radio series 'Cabin Pressure' the two pilots, Douglas (Roger Allan) and Martin (Benedict Cumberbatch before he got all magical) used to play word games* and puzzles on long flights to keep awake and alert. This was in part based on the experiences of Finnemore's father who was a commercial pilot. Any single pilot flying a long route would get terminally bored. You need someone to chat with, discuss things and generally make life acceptable.

    *My favourite game was take the title of well-known book, movie or play and remove the last letter to make another but hopefully amusing title. So, for example 'Far from the madding crowd' becomes 'Far from the madding crow', 'Three Men In a Boat' becomes 'Three Men in a Boa'. I have recently thought of 'The tale of Peter Rabbi (t)', 'On the Origin of Specie (s)', Animal Far (m)', 'The Road to Wigan Pie (r)', 'My Fair Lad (y)', and my current favourite "Who's afraid of Virginia Wool (f)' . I welcome contributions from others.

    1. Irony Deficient

      Re: Tedium

      For books, there’s Stephen King’s I (t).

      If you include poems in the range of works, there’s Poe’s The Rave (n).

      One film that’s not widely remembered now is Abbott & Costello’s Lost in a Hare (m).

    2. Irony Deficient

      Re: Tedium

      And if you include albums in the range of works, there are two from the Beatles: Rubber Sou (l), and Sergeant Pepper’s Lonely Hearts Club Ban (d).

  27. Michael Hoffmann Silver badge

    Where to start?

    Reading the article and then going through the thread, there didn't seem to be too many, if any, actual pilots here?

    Former one here.

    The trend towards single pilot operation has been going on for years and isn't new. In GA, single pilot IFR was considered one of the greatest killers, due to work-load (next to VFR pilots flying into IFR). Automation helped immensely. There are jets now rated for single-pilot operations. I expect the next stepping stone will be freight, with massive long-distance flights done by 2 people taking 8 hour shifts or such. For people-movers there's a lot of inertia and psychology at play.

    That said, when things go wrong is when the workload split between 2 people becomes a life-saver - one flies the damn thing, the other does the diagnostics. Which can involve fiddling with the automation that suddenly isn't so automatic.

    THAT said, people who think the 2 pilots in the cockpit at all times somehow prevents a lone rogue pilot flying into mountains: have fun googling "airliner pilot suicide". As far as we can tell there were both pilots in the cockpit in most if not all cases. Though whether one was then incapacitated by the suicide-murderer we can mostly only speculate. IIRC, in one case there the recorders tracked what may have been a struggle? The most recent Chinese airliner case certainly showed a flight pattern where one pilot tried to dive directly into the ground, then what looked like an attempt to pull up, followed by the final dive.

    Also, can we stop with those idiotic "can a non-pilot land an airliner?" videos? At least until we "normalise" on a generic pattern as to what's being tested? Sometimes they're talked down, at least one video they actually had the gall to talk the non-pilot through turning on auto-land, after which they just sat there, and then claim "see, they can do it!" I have friggin logged time in a commercial 747 sim, including take-off and landing, and I don't think I'd get the damn thing safely on the ground without a LOT of help.

    Finally, ab initio training has been a thing for decades outside the US (not in the US, where it's always been "through the military" or "pay your dues because we all had to as well"). You get all the training from ground up and become an indentured serf for an airliner until you "pay it off". Seems to have become less popular, but makes a comeback in times of major shortages.

    1. Eclectic Man Silver badge

      Re: Where to start?

      Michael Hoffman : "can we stop with those idiotic "can a non-pilot land an airliner?" videos? "

      The landing speed of an Airbus A320 is between 137 knots and 155 knots (https://www.quora.com/What-is-the-landing-speed-of-the-A320). They maximum landing weight is 64,500 kg (https://www.lufthansagroup.com/en/company/fleet/brussels-airlines/airbus-a320.html). An inexperienced person trying to land one of those finds that on touchdown they are driving a racing car with unfamiliar handling at over 150mph. I used to fly gliders (some time last century) so I reckon I could get one down, in absolutely perfect conditions (long straight flat runway, no side-winds, perfect visibility etc.) but I doubt it would still be in one piece afterwards, and certainly not anywhere close to being airworthy (and I was really good at landing gliders too, never even bounced once).

    2. Triggerfish

      Re: Where to start?

      Doesn't having a single pilot go against all the ideas of safety from crew management training? That is that the other people are also there as a sense check against one person making a mistake?

    3. Anonymous Coward
      Anonymous Coward

      Re: Where to start?

      I am training for my PPL in the UK.

      Occasionally I get people asking 'is your dream scenario being asked "who can land this plane"?

      Fuck no!

  28. Michael Hoffmann Silver badge
    Unhappy

    Bit of a horrifying aviation article

    https://www.aviationpros.com/aircraft/commercial-airline/news/21270942/murdersuicides-by-pilots-are-vexing-airlines-as-deaths-mount

    Quote:

    "The risk of dying on an airliner has declined significantly in recent decades as a result of innovations in safety equipment, aircraft reliability and pilot training. After 5,005 people died on Western-built jets from 2001 through 2010, the total fell to 1,858 the next decade, according to data compiled by Boeing Co., AviationSafetyNetwork and accident reports. The odds of being on a plane involved in a fatal accident was about one in 10 million, according to Boeing.

    But deaths attributed to pilot suicides bucked that trend, actually moving upward, according to data compiled by Bloomberg. If the China Eastern crash is confirmed as the latest such suicide, it will mean that deaths due to intentional acts have exceeded all other causes since the start of 2021. "

  29. Cyberian
    Trollface

    Safer option

    Replace the airlines CEOs with AI. Then there will be no need to pay massive bonuses, perks and all that jazz. Redirect the funds to the flying staff.

    And when a "blue screen" situation arises, or "there's an unexpected [item|event] in the [baggage?*] area" there won't be a new episode of Air Crash Investigation. Just reboot the CEO. (Not to mention thay if the CEO gets hacked, there won't be as many casualties.)

    (*-replace with: cockpit, engine, pitot tube(s), wheel well, baggage area, fuel, electrical systems, cabin or whatever is appropriate)

  30. Strahd Ivarius Silver badge

    First rule of safe travel

    If you see Tom Hanks boarding, go out!

  31. Daedalus

    Sledgehammers to walnuts

    Yep, let's spend billions on automation to save millions on salaries.

    The nice thing about a mega tech project is that the top people can start one and then bail (with golden parachutes, of course) before the bill comes due.

  32. dwrolfe
    WTF?

    So how will this single pilot get trained?

    Let's assume we can overcome the safety and cultural problems and create a single pilot airliner that's safe to fly and people will pay to travel on.

    Who trains the single pilot? How?

    Will we still insist of 1,500 hours before letting them be a pilot? Hours on what?

  33. Daedalus

    OMAAD

    Maybe it could be "one man and a dog" as in "the man is there to feed the dog. The dog is there to make sure the man doesn't touch the controls."

  34. jollyboyspecial Silver badge

    SPOF

    Sounds like a good idea if you don't consider the consequences of failure. Sure in theory you could even automate a plane so that it could take off, fly to it's destination and land without the pilot doing much at all, but is it a good idea? With a single pilot on board what happens if systems fail? Could you rely on a single pilot to sort it all out figure out what to disable in order to take manual control and safely fly and land the plane?

    Let's not even consider what happened with certain automation in the 737-MAX. Given the role played by a certain regulator in that debacle I'm not sure any of this sounds particularly reassuring.

  35. Dom 3

    And in 2122?

    Will we require flight crew in 2122? Probably not. In 1922, lifts ('elevators' if you prefer) came with operators. Now safely automated.

    So it becomes a question of "when?", not "if?".

    For now I want two trained meatbags up front.

    1. Irony Deficient

      Will we require flight crew in 2122?

      I won’t — I’ll be far too busy a-mouldering then to worry about flight crew requirements.

  36. Where not exists

    Ghosts in the code: the near crash of Qantas flight 72

    https://admiralcloudberg.medium.com/ghosts-in-the-code-the-near-crash-of-qantas-flight-72-b4faebc90e27

    Watch your outputs.

  37. Cuddles

    No such thing as a single pilot

    "Pilot incapacitation: Detect whether the single-pilot during the cruise phase of the flight is no longer fit to fly. Ensure that the level of safety remains acceptable in case of pilot Incapacitation."

    A system that is capable of taking over when a lone pilot becomes incapacitated is a system that is capable of flying without having a lone pilot involved in the first place. This isn't trying to reduce the number pilots required, it's trying to eliminate them entirely.

  38. druck Silver badge

    Envoy Air Flight 3556

    Envoy Air Flight 3556 has just suffered a pilot passing out during climb out, the co-pilot took over, turned the aircraft around and landed safely. Unfortunately the pilot later died.

    With only a single pilot, the outcome could have been very different.

  39. RobDog

    FAA VS. EASA

    https://go.reg.cx/twdml/3c18f/63a8e3ff/62d64363/46xX?utm_source=weekly&utm_medium=newsletter&utm_content=article

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like