Previously on this topic, see "Could the Crash Have Been Avoided?", "Pilots on Germanwings," and "More from Pilots and Doctors." Here is the next crop:
1) Bruce Holmes, a lifelong flyer and former career official at NASA, is one of the genuine pioneers of modern aviation. I described him and his work at length in my book Free Flight, and in the years since then, he's become a friend. He sends a message about the technological implications of this disaster:
Could this be avoided?: In certain statistical ways, the Germanwings suicide crash is a "Black Swan" event—imagined, but not anticipated (with a nod to Nassim Taleb). It is a little like forecasting earthquakes—not done on any practical timescales but done reasonably well on geologic time scales (as described in "The Signal and The Noise" by Nate Silver).
The fact of the matter is that we need to react on practical, not geologic time scales. To that end, my remarks have to do with the emerging capabilities to turn the airplane into an equivalent of a node on the Internet. While this is scary and promising in the same breath, I will speak to the promising facet.
The industry is within not too many months of having the first commercial deployment of true broadband air-to-ground WiFi capability (far beyond the performance of current purveyors of email in the sky), making the "connected airplane" a reality. In addition, the industry is not too many months from having the computational means for assessing flight path optimality and conformance (in both safety and economic terms) very rapidly (in "fast-time" as we like to say in the tech business).
We imagined such a capability back in the days of the AGATE program at NASA [JF note: a program to bring modern technology to civilian aviation, described in my book]; now these capabilities are becoming reality.
So here is the scenario: Pilot goes rogue, for whatever reason. Airplane is irrevocably connected on the Internet (actually Intranet, with necessary security features). Airplane begins to perform in ways not aligned with logic (plenty of modern engineering tools available, such as Bayesian Belief Networks, to compute this assessment). Ground control systems recognize the deviation and under parameters that project "something bad appears to be happening," control is taken over by those ground systems.
Sidebar: After 9/11, NASA Langley conducted a successful flight demonstration at the behest of the White House of such a capability using our B-757 flight test aircraft. Such a system was not previously implemented because the air-to-ground networked bandwidth systems and flight path management computational capabilities did not exist. Soon these systems will exist.
The benefits of solving this problem range from the obvious (eliminating the rogue-in-the-cockpit scenario), to the not so obvious—that is, obviating the need for secure doors on the cockpit and making it possible for every 10-year old girl or boy who might be inspired by an opportunity to see what happens in the front of the airplane a chance to do so.
2) Don't blame the $99 fare. A number of readers write in to protest an earlier comment, from Adam Shaw in this post, that the relentless drive to cut costs in air travel was shaving the margin in pilot qualifications. A sample dissent:
I think there's some good points your correspondent makes here, but I think it exaggerates the importance of lower pilot salaries to low-cost airlines.
Low-cost carriers base their business on using smaller regional airports, packing more passengers on board, avoiding money-losing routes, having simple low-maintenance fleets, and flying their planes for more hours of the day. A cheaper wage bill helps but those are the key factors.
Staff costs clearly are a big part of airline costs—mostly around 20 percent-25 percent—but while pilots are the best-paid employees outside of head office they're a small part of the workforce. Qantas, the airline I know best, has nearly 30,000 employees but only a bit more than 2,000 pilots—and a number of those aren't even working for the airline, but on "leave" that allows them to work for rival carriers.
I think as journalists we tend to think pilot pay is a really big deal for low-cost airlines because it's the battleground for a big political fight that plays out in the media. Airlines want to weaken regulation in the area because it's part of their cost base (unlike, say, oil prices) where their actions can make a difference. And unlike, say, route planning and reduced turn times, changes to the current status quo yield clear losers who will make their case to journalists. Losers who, in this case, are heroic figures in peaked caps.
For what it's worth I think your correspondent is right to deplore any decline in standards for commercial pilots—the huge respect in which the industry is held derives in large part from a safety record that is born of its heavy regulation.
But I think if he's seeking someone to blame, it shouldn't be the punters seeking $99 airfares—if those prices can survive $150 oil, they can certainly survive higher pilot pay—but the companies, lobbyists and legislators allowing that regulation to be watered down.
Similarly:
I really need to take issue with your implicit (via highlighting) endorsement of the statement that "when people start looking for whom to blame, the answer is simple: Joe-six-pack who wanted a $99 flight from New York to L.A."...
The simplistic idea that the market determines everything we get (and so, deserve) is easily challenged. While we might, in fact, "get what we pay for" (though even that trope is suspect), we frequently don't get the price-driven result we want. Otherwise, we would likely be reading equally simplistic statements about how the reason we have cars with no seat belts, no passive restraints, no pollution controls, etc. is because we insist on having a $5000 car...
In fact there is infinitely more desire for that $5k car than there is for the $99 flight, and eliminating or watering down all the government specs for safety and environmental impact would go a long way toward meeting that desire, but the reality is clearly going in the opposite direction ...
Joe-six-pack was never given the choice of getting that $99 ticket by accepting the risk of a minimally qualified pilot. Quite the contrary, the automotive equivalent of what Joe-six-pack ends up with is a car that appears to have all the safety and pollution control features, but doesn't really, because the Department of Transportation colluded with the auto industry to quietly lower the standards, without drawing any undue attention from the public that was theoretically "demanding" (via price seeking) the change.
3) The puzzling economics of pilot training. I won't bother to spell out all the aviation lingo in this one. The reader's point is that regulatory requirements and economic/practical realities are pushing in contradictory directions:
I don't see how you're going to simultaneously eliminate P2F, up the hour requirement to 1500, and keep the pilots paid. You then have a much larger contingent of pilots (since each pilot has to complete another 1000 hours of non-ATP time) competing for the same number of instructor, jump school, and banner towing jobs, which will do nothing but further depress wages in those areas, and make being a pilot an even more non-viable career track for most people. To build time, the pilots will either be working for almost free, or back to P2F. Sure, it might make the actual ATPs have an easier time of it, but it would be a lot harder for pilots to get there in the first place.
Also, unless they get a job flying a corporate jet, most of the hour building time will be in pistons or turboprops, not jets, so at some point they're better off just jumping into jets and learning what they're actually going to do, rather than shooting VFR approaches in a 172 for hundreds of hours.
4) Co-pilot by choice. I've received a number of notes from professional pilots on why they prefer to remain in the right seat of the cockpit, in first office/"co-pilot" role, rather than becoming a left-seat captain. For instance:
I’ve been a professional pilot for over 30 years, the first 14 or so as a flight instructor and the last 16 as an airline pilot. As a flight instructor, I spent almost five years at Airline Training Center Arizona, the Lufthansa owned school where I’m assuming Andreas Lubitz trained.
1. All of the hand wringing over how to prevent another incident like this is wasted time, a solution looking for a problem. As you’ve pointed out, being on a commercial airliner is statistically one of the safest places you can be, safer even than your own home. The chances of a renegade pilot intentionally crashing an airplane are vanishingly small.
That said, if a pilot wanted to do it, it would be next to impossible to stop them, regardless how many people are the in the cockpit. The primary reason the FAA requires a flight attendant or other crew member to be in the cockpit when one of the pilots is using the lavatory (which is the ONLY reason we are allowed to leave the cockpit) is to be there to check through the peephole in the cockpit door to ensure that when the other pilot notifies the flight deck via the interphone that he/she is ready to return THAT IT REALLY IS THE PILOT AND NOT A BAD GUY.
The "two people in the flight deck at all time" procedures were developed over some months after 9/11. One of the problems the security experts had with simply allowing the pilot in the cockpit to unlock the door (via an electronic switch located on the control pedestal) was that the cockpit interphone audio quality is not great, and pretty much anyone could use it to say, “Hey, it’s _____, I’m ready to come back in”. Passwords were tried and quickly dismissed after some uncomfortable situations ensued from miscommunication and forgetting the password of the day. Why the rest of the airline world didn’t follow the U.S. procedure is beyond me.
As a normal security precaution, this system works really well, but if the pilot in the cockpit IS the bad guy, the element of surprise/shock at being attacked would pretty much preclude the other person in the cockpit, be it a flight attendant or the other pilot, from putting up a effective defense ...
2. Lubitz's experience or perceived lack thereof has NOTHING to do with his actions. Adam Shaw has some valid concerns regarding some of the regulations regarding pilot training and pay, but I don’t see any connection in Lubitz’s suicide to any of Shaw’s arguments. Back handedly attacking ab initio programs (“250-hr button twiddling geeks”) like the one Lufthansa has been operating for decades is ridiculous—Lufthansa’s safety record speaks for itself. [JF: Ab initio programs are ones in which candidates start out with no flying experience whatsoever and are trained from the start to fly big jet airliners.]
Properly conducted ab initio training is MUCH SAFER for the pilot, the airlines and the traveling public than forcing aspiring airline pilots to fly “gritty, shitty and temperamental” equipment. Yep, that’s how I did it for the first couple of years, and no, I don’t look back on it fondly.
Airline ab inito programs like Lufthansa’s are every bit as rigorous as the flight training conducted by the military (and let’s not forget that the U.S. military basically invented ab initio training. Active duty fighter pilots are put into combat, and C17 pilots fly missions around the world, with less flight time than Lubitz had). Quality of experience is every bit as important, if not more so, than quantity of experience.
I'll argue all day that a pilot who has trained from day one to operate in an airline environment is much more suited for an ATPL at 250 hours than a flight instructor who spent 1500 hours doing airwork and traffic patterns in a C172 (not that there is anything wrong with this—I've given 6000 hours of dual instruction and though it made me a pretty good GA pilot it did not really prepare me for airline flying).
The multi-crew license (MCL) is the future of commercial aviation, and if properly regulated and conducted will be the answer to the looming pilot shortage (it's for real this time) ... Pilot pay, especially at the commuter airline level, is a different issue and, unless it turns out Lubitz did this because he didn't feel his pay was adequate, way off subject ...
3. Last, beating a dead horse, the whole co-pilot issue. In today’s world, there are always two pilots operating an airliner and they are referred to as the captain and first officer. They are ONE ANOTHER’S CO-PILOT. But yes, the first officer is often referred to as "the" co-pilot.
I think it's important in a story like this, however, for the public to understand how the airline hierarchy operates i.e. the seniority system.
As a pilot, your date of hire (seniority) determines EVERYTHING about your airline career: Where you’re based, what equipment you fly, your schedule, vacations and yes, whether or not you’re a first officer or captain. Hoot Gibson, a test pilot and shuttle astronaut and in anyone’s estimation an outstanding aviator, started at Southwest Airlines as a “bottom of the list” first officer just like every other pilot at the airline.
I’m almost 50 years old, have 20,000 flight hours and 14 years with my company and could upgrade to captain whenever I chose (yes, I'd have to go through upgrade training, but VERY few people fail to become captains solely because they can't pass).
Upgrading now, however, would mean moving from being a relatively senior first officer, with my choice of base and schedules, to being a relatively junior captain, where I’d be on reserve (flying wherever the company sent me, often at the last minute) and a good chance of being moved to another base. Although it means giving up a bit of money (captains typically earn about 30% more than first officers) I’m not willing to make the lifestyle changes that upgrading to captain would ensue ...
Waiting to upgrade means that, on a regular basis, I fly with captains who are ten or twenty years younger than me, have half my experience and who have only been with the company for eight or nine years, but who were willing to take an early upgrade in order to earn more money ...T he misconception that the captain is always older and more experienced (much less a better pilot, which is yet another topic) than the first officer needs to be put to rest.
Similarly:
I was chatting with a friend of a friend a few years back who, after graduating from a small, liberal arts school, began flying 19-seater Saabs for [a regional airline] ....
He was several years in and I asked him which seat he sat in—he was a co-pilot. I asked why and he said that while the pay was lower, he was near the top of the seniority lists, so he basically got whatever trips he wanted, and worked the schedules he wanted to. As a guy in his 20s, the pay was less important than the lifestyle. He had a good record and could probably move up to a captain's seat had he wanted, but he'd be bumped down to the bottom of the seniority list, working oddball trips at the airline's discretion ...
When you're flying internationally, it's generally assumed that you get a better product on a foreign carrier (especially Asian carriers and Gulf-based ones, and especially in higher classes of service). But it certainly sounds like US regulations are far more stringent for crews, and no matter how much free booze you get, I first want a competent flight crew. Of course, it's hard to put a price on safety, but easier, apparently, to put a price on legroom.
Thanks to all. Will do another reader update when there are some significant new facts or views to offer.
This article was originally published at http://www.theatlantic.com/international/archive/2015/03/can-we-learn-from-germanwings/389057/
