Two Pilots, One Pilot, No Pilots
January 17, 2023
FOURTEEN YEARS AGO I appeared on a TV show with an aeronautics professor who predicted that pilotless commercial planes would be up and running within ten years. I described her prediction as preposterous, laughable, possibly even deliberately misleading.
Enough said?
And no, I’m not gloating. You can’t really gloat over a contention that’s ridiculous from the start.
Those of you who’ve been following my posts and columns for any length of time are familiar with my frustration whenever the conversation turns to autopilots and the prospect of pilotless planes. “People have a vastly exaggerated understanding of how cockpit automation works, and how pilots interact with that automation,” is a statement I often repeat in interviews. It’s one of several overused mantras of mine, but it’s true, and it gets to the gist of why the oft-cited professors and researchers are usually so wrong.
At one point the New York Times ran an op-ed of mine on the topic. I hoped it would shed some light, maybe change a few minds. But I’m arguing into a pretty fierce wind. We’re living in a world where people seem to think a $75 drone from Amazon and a Boeing jetliner are the same thing, and our tech-fixated age has no appetite for old-school luddites like me, even if we’re right.
Sure enough, this irritating whack-a-mole is back in the news. The latest version is slightly modified: The experts seem to be admitting, finally, that a truly pilotless jetliner is eons from fruition. Instead, the new version goes, we’ll be zipping around soon in planes with only one pilot instead of two. I could link to any of the recent articles discussing this, but I’m lazy and annoyed and I don’t feel like it. You can find them if you need to.
This time the proposition isn’t quite as silly. Still, things aren’t as simple, or as likely, as they’re making it sound. Could a commercial plane be safely operated by a lone pilot? Sure, most of the time. But would it be the wisest choice? Notice my wording, and although the risks are impossible to quantify, suffice it to say that within the realms of airline safety, “most of the time” won’t cut it.
There are certain applications here I’m open to discuss. For instance, the possibility of permitting one pilot in the cockpit during the cruise portion of long-haul flights, instead of two, thus allowing longer flights to be staffed with two pilots instead of three, or three instead of four. (Currently, long-haul flights carry augmented crews that work in teams.) That sort of thing.
Perhaps certain cargo flights could, at some point, be flown single-pilot. But again I’m skeptical. There’s a lot of work to do, both technologically and human factors-wise. As it stands today, even the most modern two-pilot cockpits become extremely busy at times, and many of the highest-workload scenarios have little to do with automation. Automation on or off, both pilots can find themselves task-saturated.
And I’ll stop there. There’s not much I can say that I haven’t already said. Please refer to my greater manifesto on the subject, HERE. Please read the whole thing before leaving your snarky comments below.
Related Stories:
AUTOMATION MYTHS
PILOTLESS PLANES? NOT SO FAST.
Upper photo courtesy of Unsplash.
Lower photo by the author.
Leave a Comment
Maximum 1500 characters. Watch your spelling and grammar. Poorly written posts will be deleted!
33 Responses to “Two Pilots, One Pilot, No Pilots”
You are viewing newest comments first. Click to reverse order
This post reminds me of a related JOKE that I can’t remember where I read it, but will re-post here:
In the future, commercial jets will be flown by a pilot and a dog. The pilot, to reassure the passangers that the aircraft is flown by someone capable. And the dog, once that they both are inside the cockipt, to bark and bite at the pilot if he even attempts to touch any of the controls.
Am a big fan of your column and, of course, “Cockpit Confidential”. My Q: do you do any recreational flying? Is there a Gee Bee in your garage? Does mastering aerobatics make you a better commercial pilot?
Elon Musk has taken no end of flak over giving the name of his driver-assist feature “Autopilot.”
As a rated pilot Musk has a pilot’s expectation of what an aviation autopilot does, but constant complaint in media and Internet message boards is that most reasonable people expect the function to completely automate the driver function. So they supposedly kill themselves and other people with that assumption and the word is that Elon lied and misled them just to get richer.
It doesn’t help that the “FSD” software package actually is an attempt, so far unrealized, at providing so-called Level 5 driver automation. Opinions on this run a full spectrum, from “it will save lives and in fact already has” to “unrealistic wishful thinking that will inevitably lead to wide scale carnage.” And of course Musk is a liar who pathologically lies so he can get richer.
So it should be no surprise that there is a general public array of misconceptions regarding aviation automation. After all we have drones and cruise missiles that do all sorts of things without a pilot on board so why not airliners?
These pilots are obscenely overpaid. But even if the people get rid of half of the little rascals, there won’t be enough cost savings to offset the costs of infrastructure pork, bribery of politicians and regulators, union-busting and insurance company demand$.
There will be no Pilot 1. No pilot Zero. Ever.
Toy problem. Solved.
Now go fly a balloon. To Montana or something.
Published this morning:
https://www.axios.com/2023/02/03/fully-autonomous-flight-planes
Railroad executive here. Automation of operations in the railroad industry — freight, long-haul passenger, commuter passenger — is not within my lifetime, unless people are willing to accept concurrent massive losses in capacity and safety, and massively increased costs. I don’t see how it would be any different in aviation. We’ve studied automation to death — even implemented it at colossal expense on some isolated, boutique, operations that aren’t remotely applicable to everyday operations — and the only thing I can say about it is Waste Of Money.
At best it substitutes the judgment of programmers and engineers, who almost by definition do not understand and cannot understand what they’re trying to automate because they’ve never done it, for the judgment of professionals who quite literally bet their lives on their knowledge, expertise, and a little bit of paranoia.
At worst, the automation requires the operation to be modified to adapt to the limitations of the automation. Loss of capacity, fluidity, reactivity, and increase in costs results. The labor savings is ephemeral; it simply replaces the on-board crew with a retinue of office staff, which have every possible financial and legal incentive to bloat their ranks and study everything to death.
Mr. Smith I sometimes disagree with, but on this subject he’s 100 percent on target.
I do have one prediction I feel is fairly safe: in commercial aviation the first place we’ll see a start on full “pilot-free” automation is in ground handling and taxi. At large, busy airports it’s complex enough to create a lot of pilot workload and be a fruitful source of error, and it’s also a capacity bottleneck. And most importantly, it’s vastly easier to make it fail to a safe situation since the aircraft can’t fall out of the sky.
But even this, once you look at it closely, seems desperately far away. It’s another one of those things where once you get into the details it’s a lot harder than it first looks even to get to where it could be done with absolutely _no_ pilot interaction with the aircraft or ground controllers between pushback and runway threshold even 95% of the time. And getting from there to 100%, where it can recover the system from something like a flat tire, is _much_ more work.
(Sorry for the separate post; I wanted to add this as a reply to my earlier post but the “Reply” button here hasn’t worked for me for quite a while.)
From the argument I had in the comments on the recent Mentour Pilot YouTube video on single-pilot aircraft, I think the trigger for the latest round of this has been ChatGPT and similar “AI” advances.
This has been yet another round of, “Wow, that looks impressive” improvements, and in certain ways it _is_ impressive. But that invariably triggers a fresh round of “We’re obviously just about to have AGI (Artificial General Intelligence)” claims and similar from people who haven’t been following the field and haven’t seen the same claims over and over (and over) again since the 1970s and even earlier, which continue not to pan out. And these people not only don’t know the answer to, “Well, what’s different _this_ time?”; they don’t even know to ask that question.
While ChatGPT has some serious implications for improving Siri’s “understanding” of simple commands, but it’s not only done nothing at all for having computers reason, but made things worse on that front. It produces answers that sound more authoritative but are actually just as often wrong as a random Internet search and can’t provide references that let you check whether it’s wrong or right; you can only decide if its answers are correct by already knowing the answer or doing the same research you’d do without ChatGPT. As Ian Bogost said, “Once that first blush fades, it becomes clear that ChatGPT doesn’t actually know anything—instead, it outputs compositions that simulate knowledge through persuasive structure.”
And all it takes is one “Colgan air” accident, that you’ll have the public fuming over it (for valid reasons or not…), and that’s the end of that.
I’m not a pilot. I have just enough acrophobia that I can’t even imagine handling a landing. But I have several friends who are recreational pilots. Some of their stories are amusing but also frightening.
Like the one who flew up a mountain pass without any guarantee that his single-engine plane could reach the altitude at the top of the pass, and allowed himself to get into a situation where he wouldn’t have been able to turn around. (He managed to get through the pass, obviously, but it could have been another “pilot error” fatality.)
Or one who took off from SMO in “scattered” weather, and flew through a cloud into clear sky above the clouds. He called the tower with a “weather report” that “scattered” (less than half the sky covered with clouds) was now “broken” (more than half clouds). He was hoping to bamboozle them into helping him out, but the controller insisted he answer the question, “Where are you”. He had no idea.
I suppose an automated plane would handle that better — it would *always* be flying on instruments and not subject to the disorientation that occurs when you can’t see anything but white around you. But how good would it be at avoiding mountains? DOes it carry a complete contour map of the world?
I doubt that an automated pilot could do any better. And avoiding other aircraft in a sky full of even more aircraft than we have now (no pilots = cheaper flights = more craft)? Or dealing with the situation that Captain Sullenberger handled? I doubt it.
The more complicated something is; the more likely it is to fail due to a single part failing.
This applies to aircraft, every kind of machine, software, cities, all kinds of systems, everything – large and small.
When – it happens in aviation again, as it has in the past, I want pilots on board.
And want live air traffic controllers doing that work. The recent upleasantness at Kennedy proves their value, yet again.
I have no interest in those people who are citing the costs, until they include in their comments, what the cost of their spouse, child or mother is worth.
I think I have heard this discussed, but am interested in your opinion.
Instead of going from 2 to 1 pilots, how about 2 to 1.5. Specifically – create a new classification, where a flight attendant gets a lot of special training, and is qualified for “standard operation” (simple take-off,landing,cruising), and “emergency operation” as a second in the cockpit to help the pilot.
Clearly such a person would be expected to know and be comfortable with all the switches/gauges and making “simple” decisions.
Pay would be between flight attendant and pilot.
Alternatively, maybe it is a pilot who has completed a lot of their flight training, but are not FAA certified (yet).
Gotta put this out there: the motivation of companies to reduce reliance on human employees is manyfold… cost is part of it, but there are others. See Patrick’s other article about human workers. The SW meltdown last week seems to have been because SW couldn’t get the right humans into the right places. SW pilots last week started threatening a strike if they don’t get their way… how many companies, from fast food french fry workers to airlines, would rather have robots that can’t strike, call in sick, fight for work rules that benefit the employees, etc.
Maybe not in our lifetimes, but SOMEDAY there will be driverless cars and pilotless planes… all we’re debating is the when.
I think we’ll get pilotless airplanes right after we get lawyer-less courtrooms. After all, the law is written in black and white and AI is getting better all the time – soon we’ll let the machines determine guilt. We might as well have doctor-free hospitals too! There are already robots assisting in operations. It’s just a matter of programming. If you think I’m being sarcastic . . . I am.
Statistics has got us down to 2 engines on large pass aircraft and I think 1 engine on small passenger aircraft.My guess statistics I pilot
Hope the details of getting a plane to fly itself goes goes better than self-driving automobiles.
Flying and driving are very different tasks. Flying is much more regulated and therefor predictable. That doesn’t mean I think autonomous aircraft are in the near term, but I don’t think the lack of self-driving cars is a meaningful argument for or against automatic flight.
Certain tasks could be more automated in aircraft than they are today. Tasks that require hours and hours of unblinking observation, followed by split-second reflexes. The TCAS computer is but one good example. Today, the TCAS provides indications to the crew, and the crew must then act (or react) quickly and correctly based on the indications. Would it not be simpler to allow the autopilot to react to the alerts directly?
Airlines today get unscheduled maintenance when airspeed limits are exceeded for the flaps and landing gear. There’s no reason those systems couldn’t do their thing automatically, at exactly the right time, every time. But still with the ability of the crew to command a manual/abnormal extension if needed to meet an unexpected or emergency situation.
I work in safety at my company when I’m not flying. I’ve read over 10000 pilot written safety reports over the last 5 years. Never once have I read a report on a major incident and thought “This would have gone smoother if there was only one pilot in the cockpit”.
As previously mentioned: Wouldn’t saving on one (or two) pilots be a very small percentage of total operating costs? So saving on the pilots is going to make only a very small difference.
Patrick, can you give a ballpark figure of pilot costs? And Fuel costs? Writeoff of the airplane (even if it flies for 30+ years) is also significant.
While I don’t have exact numbers in front of me, the operational hull-loss rate of military drones is appallingly high, even for military aviation (which has historically tolerated a much higher hull-loss rate, even outside times of armed conflict, than commercial aviation ever has or would). And the US military, so far as is publicly known, isn’t asking troops to get on their remotely piloted aircraft, either.
Those two problems will have to be solved before anyone in the general public is remotely comfortable with a remotely piloted aircraft, to say nothing of an autonomous aircraft, which is *significantly* more complicated.
If I were betting money on the future of the industry, I would bet that the only major change along these lines we’ll see in the next 30 years would be the introduction of a remote-pilot capability (first in addition to, then maybe in place of, the relief pilot) on long-haul cargo flights. If that goes well, it’s not a great leap to imagine two-pilot long-haul passenger flying (and maybe single-pilot short-haul flying) where one pilot is aboard the aircraft and a second is located at a ground facility.
But there are significant technology and infrastructure cost hurdles to overcome and, as someone else noted, the savings in direct operating cost isn’t particularly large. As a rough number, about 1/10th of the price of an airline ticket pays for the entire crew, including flight attendants, whose presence would presumably still be required.
Capitalism on steroids is kind of comical when you think about it. The motivation for a one pilot cockpit is cost savings and the willingness to sacrifice some safety. Yet Delta’s CEO makes $12.4M a year. If the CEO can’t live on less than that, then by all means look for cost-cutting in other areas.
In 2019, Wired Magazine reported on a Garmin autopilot called The Safe Return Emergency Autoland System that ” … lets passengers hit a big red button to bring the plane to safety if the pilot’s incapacitated.”
https://www.wired.com/story/cirrus-garmin-vision-jet-autoland-safe-return/
So, the technology exists that makes single-pilot commercial flights possible and is currently in use in small private aircraft and jets.
As a passenger, I think there’s a certain amount of security in knowing that the folks up front are just as interested in their own safety as they are in yours. As one pilot said (and I may be paraphrasing), “If we crash, I’ll be the first one at the scene.”
Never mind the technological and infrastructure issues that would have to be overcome to safely operate thousands of planes via remote control. Just managing the uplink/downlink frequencies and bandwidth would be a nightmare.
We’ve spent the last 40 years refining the Crew Resource Management concept – with remarkable success. It wasn’t just changing our procedures, we changed the CULTURE of aviation. That was hard but very necessary (*cough* police, *cough* medicine) . And while CRM still applies in a single pilot situation, taking one pilot out of the equation negates many of those gains. It would be a big step backward.
I think of all the times I have caught my flying partner in a lapse or error, and when I have had my own mistakes pointed out to me. It’s another big misunderstanding that professional cockpits are error-free. We make mistakes all the time, but mostly catch them before they propagate into problems. I’m actually single-pilot qualified in one jet I have flown, but never wanted to exercise that privilege. I value the presence of a second pilot too much.
For discussion …
FAA Certifies Cirrus Vision Jet’s Safe Return Becoming the First Jet Aircraft to be Certified with Garmin Emergency Autoland
Duluth, Minn. and Knoxville, Tenn. (30 August 2020) – Cirrus Aircraft and Garmin International, Inc. today announced Federal Aviation Administration (FAA) certification of the Cirrus Vision JetTM Safe ReturnTM system – a revolutionary system enabled by Garmin’s emergency autoland technology that enables passengers to land the Vision Jet with just the touch of a button.
https://cirrusaircraft.com/aircraft/vision-jet/
The first serious question to be asked and answered is, “How much money in direct operating costs would be saved across the entire airline industry by allowing single pilot operation? And what percent of total operating costs is that?”
The second serious question to be asked and answered is, “How much more money will insurers charge the airlines for the additional risk?”
The third serious question to be asked and answered is, “How do you feel about having an autopilot navigate to an airport and land your aircraft after the only pilot on board has a stroke somewhere over Chicago or the Pacific Ocean?”
I’m not a pilot, but I am a software engineer, which I think gives me close to equal authority to say the idea of pilotless aircraft is preposterous and laughable. In the unlikely event it does happen, no way in hell will I ever set foot on one.
This is consistent with all the automation hype elsewhere.
We were supposed to have completely self-driving cars in five years. That was in 2016. We’re not even close. Tesla’s so-called “autopilot” is nothing close to auto, in fact so far from it, it can’t even detect and avoid crashing into a bright white 18-wheel semi trailer on a clear sunny day in the middle of a straight highway.
A friend of mine who writes software for car automation in Silicon Valley tells me that because navigating a left turn with oncoming traffic is so difficult to automate, “self-driving car” prototypes simply take three right turns and cross straight instead of trying to turn left. 😀
As long as the hype is so strong and the deliverables so paltry (“fake it till you make it” – Liz Holmes will soon be going to the federal pen for 9 years for that), we’ll keep being subjected to a barrage of lofty “visions”. The only appropriate reaction is to wave off and point out it’s all hype. Show me some actual results first, then perhaps we can talk. So far, 99% of it is vaporware. The results today are it takes at least two certified pilots. It will remain that way for the foreseeable future, no hype can distract from that.
I notice you don’t even bring up (at least here) the first thing that came to my mind, with regard to a one-pilot airliner: What if the pilot becomes incapacitated, for any reason? Not a likely contingency, on any given flight, but if there’s just one person capable of flying the plane, the critical importance of that sole individual’s uninterrupted well-being becomes staggeringly high.
I’d like your opinion on this: drone commercial airplanes. We already know they’re being used in military applications, with pilots flying drones half the world away and then going home for dinner at night. No more having to get pilots to their origin airports, everyone can fly the airplanes from fixed locations (which would have helped with the Southwest mess a few weeks ago, although crews were an issue there too).
Any reason this wouldn’t be feasible in the next couple of decades? Is there any push for this at all in the commercial aviation industry?
(I feel kind of bad asking about this, because it seems pretty clear that one of your favorite parts of being a pilot is the actual flying part. But I’m sure the airlines would move to this in a heartbeat if they thought it could save a buck and be implemented safely).
Automation is currently great for tasks and processes that are repeatable and predictable such as assembly lines, package processing etc but for roles that have a level of uncertainty at random times the intelligence is still not there.
To be honest I think flight automation as an augmentation role has a better chance of succeeding in the future compared to driving automation due to the fact that the airways have defined rules that tend to be followed compared to the appalling level of human incompetence on the roads.
As long as cars, which don’t need to take off and land and stuff, aren’t even close to being operated driverless (and they are not, because going back and forth the same route in Disneyland at 15 mph is not “driving”), I wouldn’t even think about entrusting a multi-million-dollar jet and a few hundred people to a computer. (Which could probably be hacked.)
Also, I don’t quite see the WHY.
Many of these ideas admit that they will of course need pilots on the ground to step in and to take remote control of the aircraft at times, and those people will have had to fly real airplanes, or they wouldn’t know how it reacts under certain conditions and in certain maneuvers.
So, the airlines wouldn’t even save any money. Or not much.
And once the first pilotless plane crashes, they would probably scrap the project anyway. (Or regulations would.)
Not to mention, the entire aviation infrastructure would need to be redesigned. Imagine how long a project, and how expensive, THAT would be.