Last week, Air Line Pilots Association (ALPA) president Jason Ambrosi took on an Airbus-driven initiative to promote single-pilot operations. He addressed the International Aviation Club of Washington as leader of the organization that represents 78,000 pilots flying for 41 airlines in the U.S. and Canada.
In his address, reported by Forbes, Ambrosi noted the near-disaster in Austin, Texas, in which a landing FedEx crew narrowly avoided a collision with a Southwest Boeing 737 in foggy conditions. The FedEx first officer made the quick decision to abort the landing, and the two pilots worked together to effect a safe response to the near-tragedy. Ambrosi told his audience, “Some manufacturers and foreign airlines are actually working to design flight decks that replace the very safety features that averted these disasters. They plan to replace pilots with automation. Of course, that’s insane.”
Ambrosi told the listeners that the European Union Aviation Safety Agency (EASA) is working with Airbus on hammering out criteria for the manufacturer’s extended Minimum Crew Operations (eMCO). He called the plans “a gamble with safety; and a gamble with people’s lives.”
“To prevent this risk to safety from reaching our country,” he continued, “we must work together with aviation regulators and stakeholders to discourage it across the globe. We cannot allow foreign regulators to grease the skids for their manufacturers, trying to force our hand to undermine safety in our country.”
Airbus’s new commercial aircraft CEO Christian Scherer, who assumed his position in January, told The Sunday Times of London that single-pilot airline operations are “technologically feasible.” “And bear in mind,” he continued, “if you go to a one-man cockpit, you might as well go to a zero-man cockpit. Because it all needs to cater for the eventuality that this one guy just ate a bad oyster and is incapacitated and the airplane has to take over. So, one pilot or zero pilot is effectively the same thing.”
Just ask Captain Kevin Sullivan what he thinks! A former US Navy pilot who had moved to Australia, he was the captain of QF72 from Singapore to Perth on 7th October 2008 when it went into two rapid, uncontrolled and uncommanded pitch-down manoeuvers which caused severe injuries to a number of passengers and crew.
These manoeuvres were found to have been caused by major software issues on VH-QPA, an Airbus A330-301 which was the aircraft performing the flight that day. Sullivan wanted, needed, and relied on, the assistance provided by his fellow flight-deck crew, First Officer Peter Lipsett and Second Officer Ross Hales to get the aircraft safely on the ground at Learmonth Airport, Western Australia; meanwhile, eight of the nine cabin crew had their hands full dealing with the injured (including one of the cabin crew who was thrown to the ceiling and back to the floor in the rear galley on the first pitch-down manoeuvre), all the while trying to reassure all other passengers and prevent the onset of widespread panic.
This cost Sullivan his career, his passion, and his previously quite sound mental health; similarly with Cabin Attendant Fuzzy Maiava (a Maori from New Zealand) who remains permanently crippled by the incident. Others have sustained similar but maybe less serious long-term effects. I have named these folk without their permission, because I an sure they will agree with me.
And how successful would Sully have been without Styles alongside him, reading checklists, performing a whole range of supporting tasks, but most importantly of all acting as a sounding board, even for what at the time seemed like an absolutely hair-brained scheme for landing US 1549 wheels up in the Hudson?
I believe taking away a second front-seater jeopardises the safety of every single soul on board, and quite frankly the price in human suffering and possible death is far, far too high. QF 72 and US 1549 were but two examples where the presence of another pilot on the flight deck was paramount; there have been many others – and I can absolutely guarantee there will be more incidents to come as technology continues to invade and pervade our lives. Just remember – Artificial Intelligence was and is designed by humans – and so it is far, far from infallible. Keep BOTH front seats occupied by trained pilots. Anything less is pure and utter negligence.
Amen!
The NTSB investigation library is full of software induced incidents which I would guess Airbus aircraft are in the majority. The Boeing Max incident was not an isolated event it was an example of automations shortcomings. One could fill a book with similar incidents. For whatever reason, Boeing took it on the chin while Airbus gets by with a few new lines of computer code and an “Ooops” for an explanation. They’ve been flaunting their “fly by wire” technology for over thirty years now and yet those little software gremlins keep popping up. And the FAA/EASA folks don’t have a clue. They have let those gremlins out of the box with no idea how to control them. Just how many near miss catastrophes are needed for them to figure this out. We can only thank our lucky stars there were well trained humans up front.
I have to wonder if Mr. Scherer were sitting in the front row of Qatas flight 32, one of those marvelous A380s, if he would be so anxious to remove flight crews. The root cause was not a computer issue but there is no way in hell that aircraft would have made a safe landing withour the crews on board. This needs to be nipped in the bud immediately!
You can cherry pick instances that support your argument but then you leave out the over three times more often that perfectly good aircraft come to grief due to their crews.
I haven’t fully made up my mind on the topic of automation on the flight deck, but I will say that I’m pretty sure the NTSB investigation library is far more “full of” incidents that source to pilot error than it is to software induced incidents.
People worry that self-driving cars will eventually kill somebody. That’s probably true. The goal is to kill fewer people than the humans have been doing.
One of my best friends, a check pilot and instructor on 777 and 787, said pretty soon there will be only one person and a dog in the cockpit. The dog’s job is to bite the man if he touches anything.
Is that dog going to be AI?
History is replete with “incidents” where cheaper made things more dangerous and ended in disaster. Is it OK to build something that risks human life over saving money? Is a handful of deadly crashes considered acceptable because the money is favorable? Look what happened with the 737 MAX, where the accountants drove the engineering. Look at what it took to fix that.
I had the pleasure of watching a 747 crew practice emergency procedures, and while I marveled at how the computer flies the plane (a 27-year-old system), I also marveled over the competency of the crew’s rapid and professional emergency responses while flying the computers. Nobody was the backup for the other; he pilot’s and copilot’s actions meshed into one response.
Until we can be 100% assured that the machine will be infallible and capable of acting and deciding ON ITS OWN, the human in the cockpit will be then best insurance of flight safety, and the best backup system will be the second human.
The human is far from your goal of 100% infallible. The machine has passed up humans as “infallible” long ago.
Infallibility is impossible for people or machine, and not a reasonable standard. Pilot error is listed as the cause of 53% of aircraft accidents. Of course the ALPA would oppose this, they also opposed the elimination of flight engineer.
Yet look at GA software where Garmin provides an emergency landing function in case of single pilot incapacitation. It does not try to complete the flight but declares an emergency, notifies ATC and heads for the nearest suitable airport. It is fully capable of landing and bringing the aircraft to a stop. Of course it isn’t proven to do this with a wing torn off or a failed engine or in a tornado or in the shadow of an erupting volcano, but that would be a double failure, not a pilot incapacitation.
It is only a matter of time until one of the quarter million dollar pilots is made unnecessary. The financial pressure is too great and the solution not hard to imagine.
Well written Warwick. The biggest example that comes to my mind on why this should never happen is the DC-10, United Airlines Flight 232 into Sioux City IA. Without any hydraulic fluid, multiple crew assisting, no one would have survived.
Like we buy insurance policies hoping they will never be needed, and I had a tornado safe room installed and as the installer said as he walked away, “I hope you never have to use it”, same thing with the second pilot. A second crew member is a money well spent insurance addition to the cockpit. And they are even handy to have around without emergency necessity….sometimes. No to one pilot.
Truth once again being stranger than fiction, as trite and overused as the phrase has become nevertheless characterizes this bizarre moment. This proposition is one on which the traveling public will ultimately vote and the human instinct for self preservation will ultimately decide.
Airbus is hoping to capitalize on the airlines zeal/dream of having a pilotless cockpit thus saving money And at the same time, passing it off to regulatory governments as being “safer”. The manufacturer love to point out all the accidents that were human error. What’s being missed are all the times that a competent crew saves the day at the last minute that never gets reported, Situations where AI or remote control would never be able to react in time.
AI will always react faster than a human. However, an AI doesn’t have a “gut instinct” like a human. An AI cannot think “out of the box” like a human. As far as remote piloting an airliner, yeah, that’s where I think it’s headed. However, I’ll never trust a pilot who doesn’t have skin in the game. The possibility of one’s own death is far more motivating than someone else’s.
AI? Get off that crap. It’s basic logic. “if this is true? do that? There is no way in hell the software geeks can ever anticipate all situations and write computer code to respond. These are simply hi tech programmable computers and as they say in that industry, garbage in, garbage out.
The industry is moving in that direction for one reason only. Saving money. A few errors every once in a while is perfectly acceptable. Final approach is a hell of a place to be when the screens go blank.
And what percentage of these competent crews who save the day have to do so due to their own incorrect actions?
Why don’t you do the research and get back to us?
People have the ultimate say in businesses, like plummeting sales of EV’s after all the fires, cold weather issues, and high insurance rates and long charge times. People will chose to have pilots driving the plane rather than AI and people will not chose to sit behind AI version 1.07332 in the left seat.
This concept is a response to a scarcity of pilots and rising pilot wages (which are justified). 51 years flying, 48 in the RCAF & another 24 at AC flying large multi crew aircraft, mostly long haul. Lots of minor incidents because there were two of us at the controls that could have easily been accidents – one was a duplicate of the 767 IAH crash.
Hard NO to this!
Instead the military and airlines need to initiate “cadet” programs in high school (circa grade 9-10).
The “copilots” aren’t there on standby in case the primary pilot fails, to read the checklist aloud, or to act as a wetware autopilot. They’re there as an essential, independent, thinking and observing member of a much more capable organism called the crew. A well-trained crew can be, and almost always is, much greater and safer than the sum of its parts. Sophisticated automation doesn’t observe, think and question, and so doesn’t replace a crewmember.
Engineers are gonna engineer, but until you can show me a system that works better even in 2 dimensions than the average (read: crappy) driver, don’t try to sell me on a system that purports to replace a well-trained, thoughtful, observant crewmember operating in 3 dimensions. With paying passengers onboard. Operating in airspace and above a surface also occupied by human beings. Maybe AI will be an eventual replacement for a trained human but we’re nowhere near that level of performance.
Single-pilot airline crew? Probably no. But I see a future where, on long-haul routes, a two-pilot crew is in the cockpit for takeoff and climb, the crew then goes into rotating shifts of duty and rest with active ground-based monitoring of a single pilot in the cockpit, and then the aircraft operates with a two-pilot crew for descent, approach and landing. This reduces the need for backup crews in the low-workload enroute phase but preserves two-pilot operation for the higher workload phases of flight.
I would be in the bunk second guessing every move the airplane was making not getting a second of rest
As you’re sitting “fat, dumb & happy” in the back of you’re noggin’ you’ll be wondering “what did I forget or overlook” or “what’s gonna try and kill me today?”
I don’t think AI can begin to contemplate this and Single pilot can’t discuss this with big dog sitting in the right seat.
I fail to see what Boeing and Airbus have to gain by promoting such insanity.
The USAF is trying it with tankers. No way that experiment will work out well
I’m curious how the insurance companies feel about this issue. They are the ones who will have to make a payout if an accident occurs with only a single pilot in the cockpit. I know they don’t like insuring single pilot business jets. When a company I used to work for ran that through the insurance company, they were going to charge as much additional as it cost to put an FO on the payroll. If the insurance companies won’t cover single pilot pt121 ops, then this issue is dead in the water now, as it should be. I really doubt anyone in Congress would sponsor a bill to rescind the 2 pilot crew requirement for pt121 ops.
My sense is that neither AirBus or Boeing have the hardware or software to enable 121 single pilot operations, yet…
Some of this is being tested in the military, but single pilot ops with civilian passengers is a whole other ball game, even if AI worked perfectly.
Expect this debate to hit the cargo carriers first, that’s where it’ll be tried first.
I’m just curious… what percentage of an airline’s total operating costs are pilot salaries? I didn’t take the time to google it, and I may be totally out of touch, but it seems to me that it would have to be a small fraction. If that is the case, and the savings realized by nuking the FO are somewhat inconsequential, what’s the motivation to do so? Just because, so we’re told, technology is in development that can do what they do, only better? Maybe I am ignorant as to the eventual capabilities of AI to surpass the human brain, but the reference to Sully in a previous comment is a great example of a human brain unleashing a gazillion electric impulses that ended up formulating an arguably crazy solution with a survivable outcome. If, at some point, AI could accomplish the same thing without even breaking a sweat, then perhaps I should pay more attention to Elon Musk’s doomsday outlook on what is to come. Meanwhile, until that day arrives, I will consider all the talk of single pilot operations of airliners carrying millions and millions of people a year complete nonsense. At the very least, an FO in the right seat is an extremely effective, affordable, and just plain sensible insurance policy.
One pilot equals zero pilots, by safety’s one in ten to tje minus ninth.
Single pilot type ratings on small jets are underutilized due to INSURANCE.
Labor savings? GMAB. Decent wages are at the tip of the ponzi scheme and somebody needs a decent job.
Who taxis single pilot, while running radios, checklists, etc- as smoothly as today? Single pilot means monitor, not pilot. Maybe the one that interfaces with ATC as they will not be flying the jets directly.
Can ATC handle manned and unmanned jets equally in the same intervals? Nope. So why go to one if one is essentially none?
Insurance and safety snicker at this.
IYKYK.
Given the complexities in aircraft systems, air traffic control (ATC) traffic density, and the advantages of having two pilots (i.e., “four eyes are better than two”) under Part 121 operations in the United States, the rating for single-pilot operations would be quite low.
ChatGPT Rating: 2/10
Justification:
Complex Aircraft Systems: Modern commercial aircraft have highly complex systems requiring continuous monitoring and management, especially in abnormal situations. A single pilot might struggle to manage these effectively.
ATC Traffic Density: The airspace in the United States is among the busiest in the world. Handling dense ATC communications and complying with complex instructions would be challenging for a single pilot, increasing the risk of errors.
Safety and Redundancy: The safety benefits of having two pilots in the cockpit are significant. Two pilots provide redundancy, cross-checking, and shared workload, which are critical in maintaining high safety standards.
Regulatory and Training Requirements: Current regulations and training programs are designed around two-pilot operations. Adapting these to safely support single-pilot operations would require major changes and comprehensive validation.
Emergency Situations: In emergencies, having two pilots can be crucial for managing the situation effectively. A single pilot might be overwhelmed, leading to poorer outcomes.
Conclusion:
While technology and procedural advancements might someday make single-pilot operations more feasible, the current state of technology, operational complexity, and safety standards make single-pilot operations in Part 121 air transport operations in the United States highly impractical and unsafe, warranting a low rating of 2 out of 10.
Staffing costs are by no means a trivial amount, in the airlines or any other business.
Example: FedEx has about 5,000 pilots, 1/2 are FOs.
Average FO pay (just pay, not benefits) is $200,000.
That $500,000,000/ year. This is real money to the bottom line.
Example: DAL has about 15,000 pilots, 7,500 FOs
$200,000 X 7,500 = $1,500,000,000.
Now you know why the airlines want single pilot operations.
While sacrificing safety for cost savings is not justifiable, it does serve as an incentive for profit. Financial guys may prioritize short-term gains over long-term safety and reliability.
As a Capt. on the 727 30 years ago I predicted single pilot operations on 121 acft. and my fellow crew members would laugh at me. Now I am predicting O pilot cockpits. You can speculate with all the reasons in the world, but the real reason is “MANAGEMENT HATES the PILOTS.” They always have and will to the end.
MANAGEMENT HATES (PAYING) THE PILOTS.
This weeks poll shows overwhelming evidence that something stinks in Denmark. I have to admit, with the current number of issues with Boeing (and likely every single manufacturer of commercial aircraft) and all the other things going on in our industry, I always wonder which topics and issues never see the light of day.
Not only would I refuse to fly on anything large driven by a single pilot – I guess I would gladly refuse to board any commercial aircraft at this time, even if 2 or 3 pilots occupy the front end.
Its going to take a while until public trust folds, but eventually even the cheapest of travelers will figure out that more water enters the ship than we can scoop out. Amazing times we live in.
I don’t give a rodent’s posterior what the opinion of this barnacle is. I also don’t care the opinion of the software engineers designing the systems if their opinion is that their products are better than humans. This is all dog bite man stuff.