Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Transportation

Cathay Working On Single-Pilot System for Long-Haul (reuters.com) 94

schwit1 writes: Cathay Pacific is working with Airbus to introduce "reduced crew" long-haul flights with a sole pilot in the cockpit much of the time, industry sources told Reuters. The programme, known within Airbus as Project Connect, aims to certify its A350 jet for single-pilot operations during high-altitude cruise, starting in 2025 on Cathay passenger flights, the sources said. High hurdles remain on the path to international acceptance. Once cleared, longer flights would become possible with a pair of pilots alternating rest breaks, instead of the three or four currently needed to maintain at least two in the cockpit. That promises savings for airlines, amid uncertainty over the post-pandemic economics of intercontinental flying. But it is likely to encounter resistance from pilots already hit by mass layoffs, and safety concerns about aircraft automation.
This discussion has been archived. No new comments can be posted.

Cathay Working On Single-Pilot System for Long-Haul

Comments Filter:
  • Only one in the cockpit over the Pacific did not work out so well for Amelia Earhart.

    • by Ksevio ( 865461 )

      THAT's the first thing that comes to mind? I hope Airbus has improved their planes since the days Earhart was flying

    • Not solo (Score:5, Informative)

      by virtig01 ( 414328 ) on Tuesday July 06, 2021 @11:30AM (#61555779)

      Earhart was the only pilot of her plane. Airbus' Project Connect is proposing one pilot in the cockpit under certain circumstances. There would still be two pilots on the aircraft. This is different from today where long haul routes have three or more pilots aboard.

      • Earhart was the only pilot of her plane. Airbus' Project Connect is proposing one pilot in the cockpit under certain circumstances. There would still be two pilots on the aircraft. This is different from today where long haul routes have three or more pilots aboard.

        Everyone can see what the airlines eventually want here, though: no pilots in the plane.

        No crew = cheaper operation.

        So, if they get their way, it'll go from one active and one reserve pilot, to one human and software backups when he's sleeping/eating/taking a dump, etc, to eventually Welcome to Skynet Airways.

        The corporate mindset is always, ALWAYS shed costs... especially workers... and automate uber-cheaply. Because these people would whore out their own grandmothers for a nickels' worth of stock price in

      • There would still be two pilots on the aircraft.

        Exactly. Long-haul flights requiring work of pilots exceeding 8 hours from when they get to work before the flight to when they leave work after the flight have two cockpit crews. One sleeps off-duty in the cabin while the other operates in the cockpit. This meets regulations which don't allow pilots to work for more than 8 hours.

        Once the plane is at altitude and the autopilot is set, there's not actually a lot to do in the cockpit. Project Connect would reorganize the few parts of the operation which curre

    • Re:Haha (Score:5, Interesting)

      by CubicleZombie ( 2590497 ) on Tuesday July 06, 2021 @12:15PM (#61555979)

      Only one in the cockpit over the Pacific did not work out so well for Amelia Earhart.

      Or 150 people on Germanwings Flight 9525 [wikipedia.org]. There should never be only one person in the cockpit. In the US, if one of the pilots has to get up to go take a whiz, the head flight attendant has to be present in the cockpit.

      • One pilot could easily whack the other pilot on the head anyway, so... shrug.

        • Always the same binary logic that treats each and every likelihood as absolutely identical and all likelihoods above zero are "likely".

          Preventive measures are not useless because there is a tiny risk remaining. Preventive measures are typically useful when their cost is lower than the total risk they reduce, by reducing likelihood and / or impact of adverse events.

          The reverse is also true: preventive measures that cost ten times the risk they reduce are typically a bad idea.

          Denying this is Reddit-level of r

    • A more recent example is Andreas Lubitz.
  • by bhcompy ( 1877290 ) on Tuesday July 06, 2021 @11:17AM (#61555733)
    There's a requirement for multiple pilots in the cockpit across much of the world because of the risk from a single pilot, as a bunch of Germans found out when a suicidal pilot flew the plane into a mountain
    • Airbus' Project Connect would only use a single pilot at cruising altitude.

    • by GFS666 ( 6452674 ) on Tuesday July 06, 2021 @11:53AM (#61555877)

      There's a requirement for multiple pilots in the cockpit across much of the world because of the risk from a single pilot, as a bunch of Germans found out when a suicidal pilot flew the plane into a mountain

      Also, EgyptAir Flight 990, where the relief pilot flew the aircraft into the sea. And Malaysia Airlines Flight 370, which is increasingly looking like the pilot in command deliberately hijacked the aircraft and flew it in a pattern to hide it's changed flightpath and crash it with no one knowing where it had gone. Without the satellite pings, we'd never have known where it went.

      • by tlhIngan ( 30335 )

        There's a requirement for multiple pilots in the cockpit across much of the world because of the risk from a single pilot, as a bunch of Germans found out when a suicidal pilot flew the plane into a mountain

        Also, EgyptAir Flight 990, where the relief pilot flew the aircraft into the sea. And Malaysia Airlines Flight 370, which is increasingly looking like the pilot in command deliberately hijacked the aircraft and flew it in a pattern to hide it's changed flightpath and crash it with no one knowing where it

        • But two on duty.

          So when one of them gets up to have a pee the other can lock them out. Which has happened on several occasions.

          Having two pilots doubles the likelihood of a pilot suicide.

          • The more people, the more risk one of them may be an adversary? Reality and experience shows the opposite effect: the more people are in one spot, the less risk that one of them tries (and succeed!) committing an attack.

            All critical industries do it like this, the military has done it since prehistoric times. Typically, it even is three or more people from different backgrounds and for the cockpit crew, it was like this until a few decades ago.

            The main reason we have more events like this is the world going

    • Co-pilots have done that too, by locking the pilot out when he went for a pee, or by directly wrestling over the controls until they crash.

      A single-pilot system my halve your chances of getting killed that way.

    • by Solandri ( 704621 ) on Tuesday July 06, 2021 @12:56PM (#61556103)
      It's more general than that. Say the rate at which a single pilot makes errors is 1% (the actual human error rate [wiley.com] varies between about 0.1% to 5% depending on task). Adding a second pilot then reduces the chance of both making the same error to 0.01% - a 100x improvement. This is the fundamental principle behind adding redundancy to a system [wikipedia.org]. The only way moving from 2 pilots to 1 makes sense is if you're counting on safeguards in the automatic flight systems to make up for the lost error-checking of a second pilot. i.e. This is a prelude to eliminating pilots entirely.

      It's also worth pointing out that in an emergency situation, one pilot is supposed to fly the plane, while the other pilot troubleshoots the problem (pulls out the flight manual and goes through the manufacturer's checklist). Captain Sully wouldn't have been able to safely land his plane on the Hudson if he'd had to fly the plane and make snap judgements about where he could land, while simultaneously having his head buried in the flight manual so he could go through the steps for an emergency in-flight restart of a failed engine. When you drop below two pilots, you're essentially gambling (with the passengers' lives) that the failure rate of automated flight systems (i.e. accidents caused by equipment failure) will be lower than the failure rate of two pilots (i.e. accidents caused by "pilot error").

      And even if that's true and the automated systems are safer, it'll still be a tough sell to a public which is mostly ignorant of the principle of opportunity cost and the fallacy of hindsight. They won't think of it as "it's sad this accident occurred, but rest assured that the overall accident rate is lower than if we were still using pilots." They'll think of it as "this plane could've been saved if a human pilot had been in control."
      • by jythie ( 914043 )
        I've gathered there is already a lot of questioning if going from 3 down to 2 was a good idea after various crashes that involved 2 overloaded pilots.
    • Professional pilot here. Thatâ(TM)s not why there are multiple pilots on transport aircraft. Itâ(TM)s because the high-workload phases are too much for one person. The en route phase is the lowest workload and is absolutely manageable by one person.

      Also, lthough I donâ(TM)t know about Cathay specifically, not all airlines outside the US have flight deck doors.

      • Also it's 2021 and Slashdot still doesnâ(TM)t know how text works.
      • AF447 is a much better counter example for reduced cockpit presence than the ones give here too. Two people in the cockpit, high workload whilst the most experienced was resting resulted in deteriorating communication etc. The idea that a single person might end up in that scenario is frankly terrifying.

      • Professional pilot here. Thatâ(TM)s not why there are multiple pilots on transport aircraft. Itâ(TM)s because the high-workload phases are too much for one person. The en route phase is the lowest workload and is absolutely manageable by one person.

        While that is true 99% of the time, IMHO, it's the 1% that needs to be carefully thought out. When something goes wrong, who troubleshoots and who flies the plane? Granted, the relief pilot can get up and help, but that still takes time and adds one more complexity for a pilot trying to decide what has happened. While I am not a professional pilot, I have worked in nuke plant control rooms where most operations are routine and 1 person can easily handle the workload. it's when things go wrong that you can q

    • There's a requirement for multiple pilots in the cockpit across much of the world because of the risk from a single pilot, as a bunch of Germans found out when a suicidal pilot flew the plane into a mountain

      It's a fantasy to think that having 2 pilots in the cockpit solves this problem. It is trivial to overpower someone who trusts you. And no the requirement long predated suicidal pilots.

      The single pilot danger is out of fear from mistakes, fatigue etc. All of this is quite irrelevant when cruising on autopilot at high altitude. Hell we've plenty of examples of planes that happily kept going and overshot their destination by hours because of distracted pilots not paying attention. The risk of a single pilot a

    • I don't see how the presence of two pilots would reduce the risk:
      "Hey, would you like to pilot alone while I take a few steps along the cockpit?" says suicidal pilot who then takes and uses the emergency hatchet/safety axe/crash axe.

  • by Luke has no name ( 1423139 ) <fox.cyberfoxfire@com> on Tuesday July 06, 2021 @11:22AM (#61555753)

    >"It makes sense to say OK, instead of having two in the cockpit, we can have one in the cockpit, the other one taking a rest, provided we're implementing technical solutions which make sure that if the single one falls asleep or has any problem, there won't be any unsafe conditions."

    Then hire someone cheaper to sit in the cockpit and banter, look at the skies, etc.

    The amount of money spent on researching and implementing a single-pilot solution probably costs more than keeping the extra pilots employed, but some project manager / AVP is getting a big payout from this "research".

    • Have to agree there, this reads like pure MBA profit-increasing nonsense.

      I feel like once the the tech to meet all the requirements to accomplish this is mature will already have a fully automated flying system from taxi to landing and back at the gate and the one "pilot" would only be there to supervise the computer.

      • The tech has been here for decades. https://www.lockheedmartin.com... [lockheedmartin.com] It is public acceptance that has not. If I remember the story right, the maiden flight had the pilots seated with the passengers. But that may have just been lore.
        • Sure and many airports have CATIII infrastructure for auto-land procedures but that is still a far cry of it passing safety and regulatory muster enough to enable pilotless or single pilot long haul flights. Even if it was 80% there in 1972 those 20% of outside edge cases and variables will still take a couple decades at least to overcome the hurdles to be trusted with a transoceanic flight by itself.

    • by RobinH ( 124750 )
      Airline pilots, especially junior ones, are notoriously poorly paid jobs. The joke in the industry is apparently: "when you line up for food stamps, please don't wear your uniform." It wouldn't cost much to have a junior pilot in the other seat. It's just still cheaper to have nobody there.
      • by ghoul ( 157158 )
        Charge the Junior pilots to sit in the cockpit. They need the hours to qualify for higher pay.Unpaid interns who pay to be interns is already a thing in many fields.
    • keeping the pilot awaye ? easy .. cattle prod in the seat's bottom .. that'll wake the boyo

    • Boeing 737 Max 8 Automation kill a lot of people and now we want to add even more of it?

  • by rsilvergun ( 571051 ) on Tuesday July 06, 2021 @11:29AM (#61555775)
    for pilots to go to pilot school. The military isn't training enough pilots for free anymore. Airlines got billions in training subsidies from the US Military but changes in our Airforce mean we just don't need as many pilots.
    • by fermion ( 181285 ) on Tuesday July 06, 2021 @11:59AM (#61555911) Homepage Journal
      We have pilots, it is just that price pressures limit pay. Jet Blue reportedly pay as low as 45K to under 200K. United pays 100K more. Low priced airlines depend on low priced wages.

      The original long haul flights, 19 hours from Hawaii to California, had two crews. Personnel on the plane has been decreasing my entire life. The cockpit crew is going to depend on liability.

      • We have pilots, it is just that price pressures limit pay. Jet Blue reportedly pay as low as 45K to under 200K. United pays 100K more. Low priced airlines depend on low priced wages.

        The original long haul flights, 19 hours from Hawaii to California, had two crews. Personnel on the plane has been decreasing my entire life. The cockpit crew is going to depend on liability.

        Airlines are a feast-or-famine industry. Regulation kept costs artificially high (to the point that average people couldn't afford to fly). De-regulation freed up innovation and cost efficiencies, allowing Average Schmoe to get his $99 tickets, but at the price of closing unproductive routes and being at the mercy of the business cycle (and disasters). So pick your poison on which you think is the better end: only the rich and upper middle class fly, while everyone else goes Greyhound (or simply doesn't tra

  • by MDMurphy ( 208495 ) on Tuesday July 06, 2021 @11:32AM (#61555787)
    It's only been a few decades since there was always a navigator and engineer on the flight deck. For longer flights, you'd have two each. The reduction of flight crew has been gradually happening as technology has made them redundant.
  • So...it's an automated flight, and the live pilot is really just there for backup, or maybe to do the landing.... Or maybe that's just the next step?

    • by EvilSS ( 557649 )
      I mean technically most commercial jetliners can already land themselves. It wouldn't be hard to go from what we have now to 100% automated flight from takeoff to landing. But even so, you will need pilots in the cockpit in case something goes wrong, especially if that thing going wrong is feeding the flight computers bad data.
      • Back in the '70s, TWA's L1011 aircraft had some pretty sophisticated autopilots. So much so, that often they would announce after landing if it was the auto pilot, or the human pilot who made the landing.

        It was said to be advanced enough that a layman could land the plane on auto pilot, should the pilots become incapacitated. The layman simply had to stand on the breaks when they touched down. At least that was the rumor.

        • I read somewhere that aircraft are required to regularly perform automated landings (under pilot supervision) in order to maintain their CAT landing certifications.

          Like, at least once per month or something like that.

          • I don't know about the *aircraft* but I believe the *pilots* need a certification for using their certification. In the CAT landing, instead of flying, the pilot is *monitoring* and it's a skill that has to be maintained. They don't get to just fall asleep. The look at out the window and look at the instruments and, if something goes wrong, they have to recognize it and take over. Part of the reason that CAT landings were historically not popular is that you still have to have the pilot and you have to
            • by _merlin ( 160982 )

              Pilots often don't like doing hands-off landings, because trying to ensure you're not missing something when you're not controlling the plane yourself is very stressful. It's all well and good when it works, but if it doesn't work, it's the pilot's responsibility to notice, and the pilot's fault if they fail to do so.

        • by EvilSS ( 557649 )
          Not surprised, it's probably one of the first ones with that capability. Here is a video on how it's done on a 737: https://www.youtube.com/watch?... [youtube.com]
          • Actually, I think the L1011 was the very first to be rated for a CAT III autolanding...when I go back and double-check myself.

            But, as others have commented, Pilots weren't/aren't always keen on the autoland feature. Which, I can understand. Automation, while it makes their job easier, also makes them worth less money to the company, and threatens their livelihood. Collectively, we have some of the same fears with the emerging technology of autonomous vehicles on the roadways. We want a driver in the seat

  • by altp ( 108775 ) on Tuesday July 06, 2021 @11:35AM (#61555799)

    So much of commercial air travel is already automated, and the automation is only going to get better. Yes, we hit a stumbling block with the Boeing sensors and computers. Hopefully we've learned from that and can move forward and iterate quickly. Granted "quickly" with the FAA means 5 to 10 years.

    But, as we improve automation and auto-pilots, the human pilots are mostly just to make sure things don't go wrong and fix them when they do.

    I've talked with commercial pilots where the only flying they do is crossing the country. 90% of it is all on auto-pilot, they hardly do any "real" flying as it is. I think it is likely that by improving the ergonomics / workload and automation in a cockpit so that it can be handled by 1 pilot, we will also improve the overall safety.

    • funny you say "we hit a stumbling block" when it's only Boeing making shoddy work. Do you own Boeing stock? Maybe Boeing should just go out of business instead, they aren't the only commercial jet maker.

      • Yea, because an Airbus [youtu.be] has never had disasterous outcomes from pilots fighting automation. Or... maybe you own Airbus stock?

        • you mean the automation saving a lot of lives after the pilot managed to put the aircraft into an unrecoverable low energy situation?
          without envelope protection the plqne would have immefiately stalled and crashed killing everyone instead of slowly gliding onto the tree tops.

          • not really relevant, they were doing stupid stunt at airshow and went off plan to were crowd were, and went lower than the trees there, and sucked in tops of trees to engines. Human dumb-assery 100% the cause.

            • That was actually my point. Human dumb-assery where the automation actually saved many lives. Without it there would have been no survivors.

        • oh you mean first one of three produced doing demonstration at air show with some "chartered passengers" where dumb-ass pilot doing fly-by was 40 feet up and hit tops of trees, sucking in branches and leaves and so had a crash?

          Not relevant. Stupid stunts at airshows with passenger jet are dangerous, yes

    • "We won't repeat our recent errors" is a form of hubris that's only so many notches below "we know what we're doing" that I don't think it makes much of a difference in this particular context.

    • funny how we forget how "Scarebus" got its name.

    • So much of commercial air travel is already automated, and the automation is only going to get better. Yes, we hit a stumbling block with the Boeing sensors and computers.

      It was actually the pilots who crashed those Boeings by reacting "wrongly" to the aircraft systems.

      (ie. they hadn't been trained to fly the aircraft after Boeing had modified them)

  • One pilot = flying bombs.

    Realise there's no way to insure that the pilot won't switch off the auto pilot ( a hammer will do fine ) and fly
    the jet straight to a target. There's no way in hell that a single pilot will ever be safe. Guy blows a fuse and goodbye building x or
    nuclear plant x or whatever.

    Can't convince me otherwise since 9 11

    • How do you reconcile your concerns with the fact that pilots occasionally need to pee during flights, and leave their co-pilot alone in the cockpit while they do so? If there were a significant number of pilots who wanted to commit mass murder we'd have this problem already.

      • by ghoul ( 157158 )
        They always ask the Chief air hostess to come in and sit in the cockpit while they pee toh make sure the copilot doesnt go suicidal. Same can be done for this new method
        • This is only due to "security" measures implemented after 9/11. Before that, the cockpit doors didn't lock and weren't re-enforced so crazies could break in and hijack the plane. But the pilot couldn't go suicidal because the rest of the crew could overpower them. Once the doors *locked* now you have an issue where a rogue pilot can do a suicide flight. So now you have to have the flight attendant in there. Not to fly the plane, of course, but to ensure that the other pilot can get back in!
        • No such thing as a Chief air hostess (or any air hostess) on cargo flights. And unless the cargo flight is ultra long haul, there really are only two pilots onboard and nobody else.

          • by ghoul ( 157158 )
            I dont really care if the pilot wants to suicide on a cargo flight as long as they dont take it down over population. BTW pilots on cargo flights wear parachutes so the copilot can jump if the suicidal pilot locks him/her out.
      • If there were a significant number of pilots who wanted to commit mass murder we'd have this problem already.

        Yep. The reason it's not very frequent is because .... it's not very frequent.

        It's not because there's two guys in the cockpit.

        • We have had infrequent murder-suicide events in the cockpit and most of them happened while there was only one pilot was in the cockpit, no matter the reason (by accident, by chance or by force).

          The rest of the time, there's two people in the cockpit (we do have women, not just guys, please update your world model) and with two of them in the cockpit, we curiously didn't have suicide events. Those only happened after one person succeeded in subduing the other or locking them out. Having no other person in t

      • A mitigated risk is equal to an umitigated risk?

        Mitigation measures to risks are worthless if they don't mitigate it to zero?

        Please.

        Humans have unlearned statistics in the last ten years.

    • There's no way in hell that a single pilot will ever be safe. Guy blows a fuse and goodbye building x or
      nuclear plant x or whatever.

      Can't convince me otherwise since 9 11

      One pilot can easily whack another pilot on the head. Think of that the next time you go outdoors.

    • You clearly haven't a fucking clue about avionics, but a hammer also works on unsuspecting co-pilots.

  • The technology for this is decades old:

    https://www.youtube.com/watch?... [youtube.com]
  • Airframe by Michael Crichton will want you to keep the Senior pilots IN THE COCKPIT.

  • No pilots. Period. Those overpaid primadonnas don't do much anyway, and electronics and computer power are already up to the job - flying an airplane is far easier for a computer than driving a car, and enormous progress has been made on that front in the last decade. Of course, dinosaurs will never go for this, but most of them will not last much longer anyway. For the younger generations, flying in a plane without a human pilot will be as natural as interacting with computers already is for (most) of the
  • I'm sorry, this isn't how you instill confidence in air transportation especially for a long haul carrier.

    • I'm sorry, this isn't how you instill confidence in air transportation especially for a long haul carrier.

      Oh? Long haul air travel has a confidence problem? Really? I mean we're talking about the safest form of transport imaginable, and one which has been made increasingly safer through many years of automation.

      • most of the time, unless a pilot is alone on the flight deck and has issues at home or with people in general. There's also that pesky issue of pilots dying inflight. [nbcnews.com] While statistically it's very safe the human factor can be risky and as long as we have that I want multiple, skilled and experienced pilots onboard.

  • I think the FAA will say no to 1 man killing this!

  • Uhmmm, who's gonna fly the plane if the pilot needs to go to the bathroom? And what if the pilot gets incapacitated for some reason?

    This sounds like a stupid idea which passengers will not appreciate.
  • A couple of things computers do poorly that pilots do generally well are:

    Communications. Interpreting what ATC is telling you can be easy, or very difficult depending on where you are and the experience of ATC. Training software to interpret what is spoken, often with poor radio quality, odd dialects, or with stepped on transmissions is near impossible.

    Observation. ATC calling to ask for visual assistance in regards to another plane. Another tough nut to crack for software to interpret and respond intellige

  • As in planning to have planes crash, as opposed to risk minimizing.
  • by VAXcat ( 674775 ) on Tuesday July 06, 2021 @04:36PM (#61556875)
    Pretty soon, the cockpit crew will be a pilot and a dog. The dog's job is to bite the pilot if he tries to mess with the automatic controls.
    • by bn-7bc ( 909819 )
      Nah that safety system is ti easy to bypass (ir pilot gives og a dogy treat) , good joke tho
      • by twosat ( 1414337 )

        The version that I've heard is that in the future planes will have just a pilot and a dog in the cockpit. The pilot is there to feed the dog; the dog’s job is to bite the pilot if he tries to touch the controls.

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...