Airline pilots who’ve forgotten how to fly

There’s been a lot of interest in my previous blog entry “AF 447 and the loss-of-control epidemic”. People pointed out that there had been nine fatal LOC airline accidents since 2000, not just the seven I’d mentioned. I’ve since added them to the list.

One of my regular correspondents, David Connolly – who flies 747s – says I am talking rubbish when I ascribe the primary blame for the loss of piloting skills to the national aviation authorities who set the requirements for the training exercises that airlines must provide for their pilots.

He says the airlines are to blame, and he’s right.

But so am I as well.

The exercises airlines are required to make their pilots practice in recurrent training are heavily loaded with engine and systems failure exercises, based on the problems airliners of the 1950s could be expected to face. These have very little relevance to modern aeroplanes where mechanical failure is rare, and much more easily managed when it does occur.

Modern aeroplanes are full of computers which, as everyone knows, are pretty reliable most of the time, but when they fail they do not necessarily do so transparently, and can leave you confused. In the Air France 447 case, the computers disengaged because they knew they were being fed incorrect airspeed data, and computers can only work when they are being fed data that’s worthy of processing.

But the regulators do not require pilots to practice managing the results of subtle, computer-related failures.

And when failures do happen in the air, the airlines want the pilots, if possible, to leave the autopilot, autothrottle and flight director engaged.

Come on, guys, recurrent training is done in a simulator, it’s safe to try things out there! Chop all the automatics, kill the flight director, and let the pilots practice directing the aeroplane by using their brains, hands and feet. That’s the practice they are never going to get unless you give it to them!

Of course Connolly’s right, the airlines could do all this as well as providing the out-of-date statutory box-ticking exercises they are compelled by law to practice, but doing more would put up the costs of those who did compared with those who didn’t bother.

I don’t like regulating everything, but regulation has one undeniable benefit: it levels the playing field. Airlines will always base their training around the statutory minima because they know their competitors are doing that. So if the statutory pilot training minima are wrong for today’s highly automated airlines, the airlines argue they have no choice but to provide this irrelevant training at the expense of what’s really needed.

The AF447 crew was not prepared for what happened. Whose fault was that? The pilots, the airline, or the guys who make the rules about how the airlines must train their pilots?

Let’s hear from you.  


12 Responses to Airline pilots who’ve forgotten how to fly

  1. Oluf Husted 5 August, 2011 at 6:34 pm #

    Dear David,

    Thanks for blaming both the Autorities and the Airlines that, in Scandinavia at least, are way too friendly.

    Just overpayed lobbyist! (for the industry they should govern)

    You can add the two pilots, on the Taiwanese ATR72 cargoplane in dec. 2002, to your list of casualties.

    Oluf Husted
    Slagelse, Denmark

  2. Aimee 5 August, 2011 at 9:25 pm #

    David, perhaps it’s a case of Airbus versus Boeing pilot skill philosophy?

  3. Chitragupta 5 August, 2011 at 10:23 pm #

    The biggest problem isn’t the quality of training but it is the quantity of training. Fellow pilots work their way into middle management and pad their wallets by reducing training cost at the line pilots expense. By reducing the training budget each year upper management rewards them bonuses for a job well done. The line pilots and their Type A personalities suck it up and work harder to pass the up coming check ride. The School House instructors and Check Airman suck it up by trying to stuff ten pounds of potatoes into a five pound bag.

    Then add in the extra training that has been mandated by the FAA in the last twenty five years, wind shear, flight into terrain, cockpit defense, Crew Resource Management, how to be Politically Correct and not using the three G’s on ATC nor in the hotel van, etc, etc, etc.

    Then the training schedule is a rolling daily start time so you come in late on day one and leave early on the last day so they don’t have to put you into a hotel one extra day because you missed the last flight home. By moving up your daily classroom/simulator times you only have enough time to grab a meal then hit the sack and up again for the next day, four hours earlier. Those four hours could be used to study for the next day’s training and/or relearn information from the previous day.

    Engines still fail, electrical equipment still shorts out, hydraulics still lose fluid but now you have a much smarter aircraft and the training keeps getting less and less.

    Just adding stuff to the cockpit doesn’t mean the crew understands it. They may be able to regurgitate the correct procedures in the simulator and the Check Airman passes them but when out on the line can they now remember what page that obscure software command is six months later after doing nothing but accepting the visual approach or the straight in ILS since the last time they were in the simulator?

    I won’t even go into the changes in procedures that still get the same thing done but every few years someone reinvents the wheel just to prove to management they were the ones for the job. Now the line pilots must comply with the all new method and get whacked on the knuckles for fall into their previous methods that the last time they were in the simulator were perfectly acceptable. This goes on and on and on with established equipment that has been flying longer then the pilot is old.

    Just looking into the cockpit and seeing six massive LCD displays and then nothing but blank real estate doesn’t mean it is any easer to fly then the mind boggling rows and rows of gauges on the B707s. At least the B707 was straight up about what you saw was what you got. The new planes it is how fast you can type and how good your cheat sheet is on stuff your suppose to know but never use in the real world except once a year every time you’re back for recurrent training. Or that dark and stormy night when ATC gives you a command that you were taught six months ago (you get it, yea, yea yea, I got it) and now in that rolling, tossing cockpit you are expected to quickly comply.

  4. rik.drixen 6 August, 2011 at 10:38 am #

    David Learmount’s article hits the middle of the rose.

    One more point needs attention. Read carefully the BEA report. It expplains, while in alternate law, the pilot has no authority on the autotrim, from the moment the autopilot disconnected until the impact, the trim staid pitch up until the end. Centered as the aircraft was, all issues possible were fatal. No pilot could save that situation on that type of aircraft.

  5. David Connolly 6 August, 2011 at 11:44 am #

    2011-07-06 CA-932 / B-2443 EDDF ZBAA 1+1
    Remarks :AP/AT TOC@FL330 F/O 9.6 BLK HRS

    David, I would never ascribe nor consign your scribal works to rubbish. If I thought so, I would not be arsed in commentating upon them. I am of the view that one’s written word should always be the longer shadow of one’s spoken word. I also believe in practice approaching perfect, without ever quite making it. I paste a plog entry of July 6 last, only using AP and AT @ TOC. Hard work that, particularly throttle tweaking, with #1 on the left pinkie. Good practice particularly, but not A good practice generally-if it is not in a sweat. I have not repeated since, but good to know I am still capable, confidence is capital. On the B-744 MEL all three autopilot systems may be inoperative, with the proviso that -“Flight crewmembers are limited to 5 flight hours per scheduled flight.” A limiting principle and not a regulated target.
    On regulation, I agree with your contention in general of lawful regulation having an undeniable benefit of making a level playing field. Rule of law is not rule by law, it enables equality of opportunity to offend or not, if you will. It is opportune to quote one of your great historical countrymen, namely John Dryden 9 August 1631 – 1 May 1700 and his poetic political satire: Absalom And Achitophel of 1681 … “For lawful power is still superior found; When long driven back, at length it stands the ground”. That is our principle today, if in practice found frequently wanting, it shall always be thus. We aim for perfection upon this mortal coil, for if we achieved it, we would stagnate and regress by letting the unattainable perfect being the enemy of the attainable good.
    As professionals, good enough, should never be enough. And regulatory law is optimal when seen as a limit, not a target.
    Regulators can be witting or unwitting parties to regulatory capture. That is to say, being less than objective of the industries they are regulating by being a former employee, shareholder or both of what they are regulating in which the company interest takes priority over the public interest. So they can agree to set a minimum standard equating with minimum cost which can prove costly in the long term. This conflict of interest is nothing new of course. Ignorance of the law is no defence. In the same way, despite the company culture and regulatory short comings and indeed aircraft short comings-in my Boeing view, being mitigating factors, Captain Marc Dubois of AF 447’s A-332, failed to take command and his ignorance of the unmitigated stalled condition of the aircraft, is sadly, no mitigating defence of de-facto if not de jure dereliction of duty. If the two guys in the front command seats don’t understand what is happening and admit it,as they did, it is past time to jump in-literally. Professionalism is always bringing more than minimum to an endeavor. A good contrasting parallel example is Alaska Airlines Flight 261’s MD-80 of Jan 31 2000, with the catastrophic nose-down failure of the horizontal stabiliser, unwittingly made worse by the crew’s trouble shooting. They shot themselves in both feet with a single bullet in the end. The relatively calm and stoic efforts of Captain Ted Thompson and FO Bill Pansky are a CRM gold standard of never giving up and trying to retrieve the impossible, while also talking to ATC and the terrified pax and staying over water to avoid population zones. Ted’s last words, 2 seconds from a 13300 fpm VS were, facing the Pacific Ocean, “Here we go !” and there they did.
    As you say, it is best to devote more recurrent sim training to the unexpected unrequited abdication of that seductive AP/AT,FMS, ND & PFD, suddenly and without warning. Yikes !, that’s a bit pessimistic.
    Or is that being an informed and prepared optimist in keeping the blue side up, a bit of both, I guess ?
    I recommend watching the pasted 1975 B-747 video link to appreciate airlife without an AT in general and without an AT in a quad in particular. Not to mention the pre-CRM era of PF-PNF dilution of tasks with an F/E goalkeeper, I hold an F/E licence since 1998. While my academic understanding under knowledge is still deep my skill set from that seat now is like your long division currency.

  6. David Connolly 6 August, 2011 at 4:03 pm #

    Upon reflection from my previous post, I should have added that the letter and spirit of the regulated law are mutually exclusive with monotonous regularity. I learned that valuable lesson from my experience of working as a legal clerk for Cohen & Naicker solicitors on London’s Edgeware Road near Paddington Green police station in 1986-1987.

  7. Chris Brown 7 August, 2011 at 10:02 pm #

    Dear David,

    The treasure in the tragedy of AF 447 is what it can teach us.

    Thank you, David, for calling for dialogue. Thank you Wood’s Hole Oceanographic Institution and Bureau d’Enquêtes et d’Analyses for recovering data we can all work from.

    Thank you, David, for offering to share the blame. All of us share the blame, all the way down to the passenger who wants a cheaper flight. But in the end, blame is irrelevant. Blame only serves economic interests which demand simple answers. These answers will not be simple.

    Let us all be brave and try to speak truth as we see it. Let us all bring our training and experience to bear on the recovered data and mine it for lessons and solutions.

    Chris Brown
    Air Canada (retired)

  8. alan giles 9 August, 2011 at 1:49 am #

    As a non pilot simply a hobbyist something doesn’t make sense with the AF447 disaster.
    from the report the aircraft was stalling, the crew focused on lack of speed and tried to address what they thought was the cause of stall warning by applying power.
    The flight data recorder showed the angle of attack was real problem , but with 5 or more large display screens available, why was this information not flashed large screen to the pilots

  9. David Learmount 9 August, 2011 at 9:57 am #

    That question is being raised widely. But if pilots will not take any notice of their attitude indicator (artificial horizon) – and this crew never referred to their pitch attitude in their discussions – would they consult an aoa indicator even if they had one?

    The logic is that the aircraft provides stall warning (based on aoa), there is stall protection in Normal Law, so why do you need another piece of data?

    If your attitude is 13deg nose up, your airspeed a moderate 160kt and fluctuating, but you are dropping at more than 10,000ft/min, it says something about your aoa. If you are pointing up at an attitude that approximates to the attitude adopted at rotation for take-off but are actually going down fast, the picture should be clear. But stress does strange things to perception.

    Airbus, and others, will be reviewing the aoa indicator option when all the wisdom is assembled in the BEA’s final report. But that will not be for a while yet.

  10. ehvee8r 15 August, 2011 at 9:27 am #

    Chris Brown has hit the nail on the head. Blame serves nothing relevant in any disaster, the causes and effects for such a disaster are too numerous to demand a simple reason or even a simple solution for that matter.

    Training, Management and Culture will all have major lessons to learn from AF447 – and with proper transparency future accidents may just be avoided.

    In my opinion (for what its worth) I think its just too easy to blame flight crews in order to satisfy the families, media, regulators, the airlines and the manufacturers. There are far too many variables and underlying factors for concluding why the crew “did as they did”

  11. Gordon Tritten 31 August, 2011 at 1:52 pm #

    AF447 has done a lot to focus on the man-machine interface, training, regulation and raw flying skills. Rarely mentioned is the motivational aptitudes of those at the sharp end. To satisfy the double digit growth of air travel we have had to lower our standards during selection and the (global) majority of current professional pilots are not motivated by the love of operating and “piloting” a flying machine which in the past was nurtured and honed in a military aviation background or via a self starter route. To many nowerdays it is just a job. Pitot icing and UAS is well documented officially (eg: in Airbus training manuals and notices and the AFR Notice To Crew) and there is a wealth of information on-line, IF your professional motivation and interest is sufficient to drive you there outside of formal and organized training in the sim or classroom. IF – as AF maintains – the crew acted professionally why did they find themselves bemused by the tell-tale signs of the specific multiple ECAM warnings, the loss of AP engagement and the appearance of Thrust Lock ? Granted their training may have been perfunctory but any mainline pilot worth his salt should understand the dynamics of high altitude flight in heavy metal at the limits of the certified aerodynamic envelope. IF they had they would not have allowed the aeroplane to climb. Granted that the psycological pressure of loosing automation and ASI while immersed in the ITCZ at night is significant but they seem to have handled this well except this was of no use if they had lost situational awareness through ignorance of the meaning of the blatently obvious clues and allowed the a/c to fully stall. Now that is scary in a heavy jet in broad daylight let alone nightime IMC when surfing the ITCZ. IF this sounds a little harsh then it is meant to be – professionals have a personal responsibility (now there’s an antiquated concept!) to their passengers to know and understand the environment in which they operate (whatever their training or background).

  12. Karl Montens 18 May, 2012 at 3:49 am #

    There are more important issues that are safety related than this but Mr Learmount is afraid to put it in writing. No one will argue that explosive growth has serious implications.