AVweb

« Back to Full Story

Should Airline Pilots Fly More -- Or Less?

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

In an AP story this week, the issue of pilot skills gets another look -- with increasing automation, pilots just don't get to fly enough to stay as sharp as they should be. The answer proposed most often is that procedures need to evolve to be sure that pilots spend more time as actual manipulators of flight controls and less as systems managers. But maybe that's a losing battle. Should pilots instead be rethinking what their proper role should be in the cockpit? In one ABC report we heard, the average total time on the controls for the crew for a typical flight was around three minutes; the rest is flown by the automation.

Modern avionics supply pilots with huge amounts of information, and they are getting better and better at conveying that information in ways that are intuitive and useful to the humans at the controls. Or maybe humans are just adapting to the machines' foible and quirks. But forget about that for a minute. What about the basic scan? Every instrument pilot who struggles to stay current understands that the scan goes first and there's no evidence to suggest that modern glass displays change that in ways that matter. You still have to look at the instrument, interpret, evaluate and decide. And any flight instructor or experienced instrument pilot will tell you there's a difference between scanning with the autopilot engaged and while hand flying.

There may be no more vivid example of this than the recent reports about what went wrong in the Air France 447 cockpit. For several minutes, the crew apparently could not interpret the airplane's departing flight attitude. Could that have been the result of just not enough hand flying? Conversely, would that flight have had a better outcome if the pilots were taken entirely out of the loop? That's a question that may be argued for years to come, but if we revisit that notion five or ten years from now, it seems to me more likely the answer will be that the airplane of the future would do better on its own.

That doesn't mean there won't still need to be humans in the cockpit. But maybe while we're ensuring those pilots maintain their flight-control skills, we should also be focusing on how to train airline pilots for the day when what they'll really need are better ways to manage complex systems and maximize cockpit resources. I suspect that day will come well before the pilots who are starting out today hang up their wings.

Comments (52)

Yes, they should fly more. But at the controls of aircraft they used to fly when they were training for their CPL/ATPL. Aircraft like the C-150 or 172 will keep they flying skills on an acceptable level. 10 hours a year or so should be a good start.

Posted by: John Piepers | September 1, 2011 12:10 AM    Report this comment

Spot on, John. A few hours flying light aircraft or even gliders with no automation should be a mandatory part of maintaining a commercial pilot's license. I occasionally teach current jet pilots wanting to get back to "grass roots" aviation, and without exception they have all said that the miss actually flying the aircraft, and even the good ones forget basic things like speed control with no auto-throttle or maintaining balanced flight without a yaw damper.

Posted by: Mike Ellis | September 1, 2011 5:21 AM    Report this comment

This goes beyond Flight-control skills, it has a lot to do with automation designs. Everything is automatic, even the standby instrumentation. If the main instrumentation fails and the standby system fails, with what do you ascertain the standby can be trusted after it recovers if it does? Mechanical instruments may have been more expensive to maintain, but their reliance and trust developed a better cognition in recovery of failures. In addition, aerodynamic training is gone, the upset recoveries are non-existent, it cost too much to train. FAA does not require spin recovery training anymore, why?

In the classrooms, theory can be discussed, but why not show it on an airplane like the good old days, money? Are the bean counters riding in the back, they are probably thinking in terms of their probabilities?

Posted by: Rene Blanco | September 1, 2011 6:29 AM    Report this comment

Automation = green. You save a lot of fuel by letting automation fly the plane. Don't expect airlines to go back to hand flying around the sky with hands on the throttles.

As for pilotage, that's no different from driving a car. Look at how many people know how to drive UNTIL the completely unexpected happens. Some people "get" the big picture physics, others don't. That goes beyond skills.

Posted by: Mark Fraser | September 1, 2011 7:49 AM    Report this comment

One of the biggest difficulties is with failures. As system complexity increases pilots (and engineers) are not able to anticipate the complex interactions. It is akin to the infamous "blue screen of death" with your computer operating system. What is happening? Who knows. As a program manager for one of the major 142 training organizations, I see a lot of automation confusion when things don't go according to plan in the sim. The tendency is to try and program the aircraft out of an undesired state rather than manually get the situation under control.

Posted by: Shannon Forrest | September 1, 2011 8:24 AM    Report this comment

This reminds me of the old "joke" about the "foot rests" that airline pilots use aka rudder pedals. Like most humor, there is an element of truth there.

I have heard commentary about pilots relying too much on "George" and every other technological aid to flying my whole career, 43 years, so far. The truth is all pilots need to know how to use those things as well as how to "hand fly" the aircraft. Both take practice.

The bottom line is knowing your aircraft and its' systems + practice.

Posted by: William Zollinger | September 1, 2011 8:44 AM    Report this comment

"Needle, ball & airspeed." That basic skill should be mandated as a core skill in IFR training, and then practiced and made part of IFR currency requirements. Any IFR pilot should be able to do it upon demand. In the airline cockpit that has another pilot to figure out why "George" is doing something unexpected, the odds of a successful outcome would be greatly enhanced.

Posted by: David MacRae | September 1, 2011 9:15 AM    Report this comment

I think there is too much emphasis on coumpters to fly. It's the rule not the exception now that airlines train to let automation do the flying. The problem is the pilot is not totally in the loop and when a failure occurs there is a lag for the pilot to assess the situation then take control. In some cases that time is crutial. During my many hours of automated flying, I always hand flew the plane to 18,000 or 20,000 feet before turning on the autopilot. Then when not IFR I hand flew the last 10,000 feet to landing. It helped me keep my flying skills. I was questioned by some of my co-pilots and explained my position on automation. I think total dependency on automation is overrated.

Posted by: Clif Walker | September 1, 2011 9:31 AM    Report this comment

The problem is greater than loss of hand-flying skills. As a Part 142 simulator instructor I find that reliance upon SOPs, checklists, and company procedures gets in the way of cognitive assessment and decision-making. I see crews fighting fires ask for "delaying vectors" so they can complete checklists, fearful of what the "company" or the FAA might say if they act independently and/or decisively based upon airmanship skills. I've watched departures in basic VFR conditions, continue into IFR conditions when a simple declaration, about-face and landing could have been made VFR! Automation-Dependency is not the only contributing factor to the loss of airmanship decisionmaking skills.

Posted by: George Horn | September 1, 2011 9:54 AM    Report this comment

to be really good at something you need to practice it frequently. doing emergencies once or twice a year when you know the instructor is going to do every dirty trick to you does not do it. I agree with cliff and that all pilots need to take full advantage of the automation but that they also must frequently hand fly the airplane so their scan and feel for the airplane becomes automatic. most airlines I hear, have rules against hand flying the airplane and insist that the autopilot be used at all times. I feel that any pilot should be required to log a certain percentage of hand flying the airplane.

Posted by: william Lawson | September 1, 2011 10:00 AM    Report this comment

Nobody can predict when that "Oh, Sh*t" moment is going to happen. The lights go out, the shiny side isn't up, surprise grabs you. Even triple redundant systems fail. You must know how to recover the aircraft. Upset recovery training such as taught by APS in Phoenix should be mandatory for airline pilots. Four days and 6k is dirt cheap compared to the cost of one accident. The skills learned apply to all aircraft. Besides, it's loads of fun. It was the best money that I've ever spent.

Posted by: Sid Love | September 1, 2011 11:00 AM    Report this comment

As a long time GLIDER pilot and instructor, I would highly recommend that airline pilots take a little glider flying which is the most basic of all piloting. At one point in their training all SWISSAIR pilots were required to fly gliders.

It would also suppress the fear of "dead stick" landings. Remember the Hudson River landing, by a glider pilot. I myself gave a glider check ride to Bob Pearson, the Air Canada captain who glided a B767 90 miles to Gimli for a successful landing (see Gimli glider). The Boeing manual listed this all engine out condition as "not survivable" And it can be really fun!

Posted by: HILLAR KURLENTS | September 1, 2011 11:04 AM    Report this comment

Piots have to be able to fly the aircraft. Counting on the "magic" to be able to do it all is foolish at best. Requiring some amount of hand flying in all phases of flight would be a good start.

Jay

Posted by: Jay Hulbert | September 1, 2011 11:07 AM    Report this comment

Hand flying has become the backup procedure to the simpler and less tedious method of letting the autopilot take care of the details so the pilot can focus on the overall picture. If pilots take this hand flying to heart there will altitude busts and nav errors when they get distracted. We hand flew the 727 all the time but that was technology from the 60's. I would love to fly the 747 like I do my Piper cub but noise abatement, airspace constraints, controller expectations, company policies aren't conducive to that. I've heard, use the automation for the last 20yrs and although I'm a slow learner eventually became a believer.

Posted by: William Rucker | September 1, 2011 11:16 AM    Report this comment

I believe that every certification check, every proficiency check and every training in lieu of a proficiency check should all include an engine out ILS to a missed approach and a non-precision approach to a landing, conducted without the use of the autopilot/automation. Many 142 and 121 programs are so locked into "training the automation" that they forget that the pilots should also have adequate skills to handle the aircraft when the automation quits or circumstances require the pilot to actually fly the aircraft.

Posted by: THOMAS OLSEN | September 1, 2011 11:23 AM    Report this comment

If, in the three minutes of confusion in the Air France crash there had been a "straight and level" button, and the pilots had been trained to use it, the outcome could of been different. The machine lost valid speed readings for 20 odd seconds, handed over to the fully trained and current pilots and the result was terrible. It is notable that the fully automatic drones now used by Air Forces across the world have had almost no crashes, and those they have had have been when some cowboy in the caravan tried to land it by hand. It will only be a matter of time before passenger aircraft get similar technology -- it does not mean that private aircraft need to follow.

Posted by: Brian McCulloch | September 1, 2011 11:26 AM    Report this comment

I don't think – in most cases anyway - the problem rests with a lack of flying skills at all. The problem more likely rests with the way humans perceive, process and act on large blocks of information.

In all the “automation accident” cases I have knowledge of it appears that the pilot(s) abruptly were confronted with an overload of information that was confusingly contradictory but obviously demanded some immediate action. They then failed to properly diagnose what was actually happening prior to taking the absolutely normal human action of fixating on and following one information pathway…which subsequently turned out to be the wrong one. By way of example, fighting a spurious increasing airspeed indication right into a stall has been a recurring theme in these accidents so it would seem some training emphasis on noting & analyzing the sometimes subtle clues that the speed might not really be increasing or abnormally high might be productive.

Posted by: John Wilson | September 1, 2011 12:49 PM    Report this comment

Pilots need to know when to hand-fly the airplane, and how to do it. I like to hand fly several minutes of each leg, just to stay in the loop.

Posted by: Unknown | September 1, 2011 1:39 PM    Report this comment

John is correct. Information overload, can lead to disaster. Add to this audible warnings, and confusion takes on a new level. Too often the complaint becomes; "Today we're analysts."

Even though being an analyst seems like a new problem...over the years, I have spoken with many WWII bomber pilots, and they lived in constant emergency, and analyzed everything in order to get back home, given battle damage, weather, breakdowns, etc.

In many ways this is a new old problem, which gets back to the ability to act under pressure, which, in my opinion, most flight crews are quite good at.

I think training will need to focus on these traits, and we'll achieve the balance needed in today's flightdeck.

Posted by: Greg Andrews | September 1, 2011 1:39 PM    Report this comment

What may be overlooked in the question of automation is the fact that it is an assitant to crew/pilot workload relief, hence should be treated as the secondary (backup)flight crew and not the primary. Lots of pilots are lulled into a false sense of security due to the reliability of the systems and numerous successful flights completed using these systems.

The vigilance that should be maintained by the human element lapses most times to a point where some pilots are guilty of dozing, or simply no longer keeping a basic scan going, or practicing observation and interpretation while in flight. As such the situational awareness suffers and when "George" does a back flip, the crew/pilot is caught in 'I don't have a clue of what, where, how, when" mode, and that is a recipe for disaster.

When both the crew and the automation are kept in their proper perspective, and limitations of both are acknowledged, then together they work wonderfully in the real world of aviation.

Posted by: David Christmas | September 1, 2011 2:08 PM    Report this comment

Hand flying in the AF447 case is irrelevant. To fly by hand you need to know what's happening. Try turning off all the screens in the dark at 40000 feet and then flying the airplane "by hand".

I think there is still some logic to having a mechanical airspeed and AOA indicator on board, or at least something that's independent of all integrated systems to inform the crew of what's happening during system failures.

Posted by: Paul Hekman | September 1, 2011 2:09 PM    Report this comment

The Air France crash, was it a failure of flying skills or a failure of decision making? I believe that the pilots became so fixated on the loss of airspeed information that they ignored everything else. This is very similar to Northwest Flight 6231 on December 1, 1974. Here is a link to that accident: http://en.wikipedia.org/wiki/Northwest_Airlines_Flight_6231. In their case they forgot to turn on the pitot heat. When the pitot tubes iced up. They eventually got both an overspeed warning and a stall warning. The pilots response was the same as the Air France pilots. In their case they spun in from FL 240. This was in 1974. Automation was not an issue. Here is another link: http://aviation-safety.net/database/record.php?id=19920730-0 This is TWA Flight 843. It took of out of KJFK bound for KSFO. Shortly after lift off they got the stick shaker. Did we learn anything in the ensuing years? Since we did not learn much from this accident maybe what we need to do is come up with a "safe mode" for operating the aircraft when anything goes wrong (as long as it is not engine related). Pitch and Power for whatever phase of flight you are in. Once you have determined you are in the safe mode, then deal with the problem.

Posted by: David Heberling | September 1, 2011 2:38 PM    Report this comment

"...had been a "straight and level" button."

Brian,

There is a "straight and level" button on every airplane -- it's the pilot. Roll in the shortest direction to the horizon until wings level and the lift vector is pointed up, then apply back pressure until the rate of descent stops. (Or if you find yourself nose high with decaying airspeed, unload, roll to 90 degrees of bank, let the nose fall to the horizon, and then roll wings level.)

I must be old-school, but I'm appalled to hear that airline pilots can now complete training and be hired without ever having practiced actual unusual attitude recoveries.

I still remember in pilot training being told to close my eyes as the instructor took the airplane, and 30 seconds later being told to open my eyes and find us at 135 degrees of bank and 30 degrees nose low as the instructor said, "You have the airplane."

Posted by: Gary Dikkers | September 1, 2011 2:41 PM    Report this comment

As a previous airline check airman and sim instructor, I noticed the degredation in flying skills during the 80's when we flew the then 'highly-automated' DC-10's. One brief example is not including the airspeed indicator in your scan while on final approach, since the autothrottles are maintaining the selected speed...unless, of course, the autothrottle clicks off, which I have seen as a jumpseater.

I don't have a problem with automation, but I do have a problem with over-reliance on automation. Sometimes, a pilot needs to be a pilot due to automation failure or having to do what has become a challenging visual approach where automation can't be used (like the old IGS13 @ Hong Kong's Kai Tak). There are plenty of YouTube videos that illustrate that, BTW. In order to maintain my personal skill standard, I often fly the airplane to top of climb, and certainly when below FL180, without a FD or autothrottle engaged. The profession demands that one has to be proficient with both the automation and hand-flying the aircraft. Unfortunately, many airlines, the FAA and many "drivers" don't understand the concept of being a competent pilot during all flight regimes regardless of the level of automation available.

Posted by: Manny Puerta | September 1, 2011 2:45 PM    Report this comment

"Since we did not learn much from this accident maybe what we need to do is come up with a "safe mode" for operating the aircraft when anything goes wrong..."

David,

Good idea, but hardly new. Seventy years ago the Ju-87 Stuka had an automatic dive recovery system for the fairly common event when the pilot blacked out during the high-g pull after releasing his bomb.

Posted by: Gary Dikkers | September 1, 2011 2:50 PM    Report this comment

"Good idea, but hardly new."

Gary,

That may be true, but we still keep making the same mistakes. Screw up our airspeed indications and all bets are off. Of course we never hear about the crews that successfully recover from this situation. Fixation is a killer.

Posted by: David Heberling | September 1, 2011 2:56 PM    Report this comment

I may be stepping out of line here, since the closest I get to any airline cockpit is when the door is open as I'm leaving the airplane. But I've come to distrust GA autopilots, from a couple of incidents in which they inexplicably failed. In one, I was on top of it and immediately hand flew when it went off line. In the other, I was complacently looking out the window, not monitoring, when it suddenly commanded a hard right turn and nearly flipped the airplane over causing it to lose several hundred feet before I hit the disconnect button on the yoke. I wonder of complacency--i.e., it's been working while the pilot's been doing something else, is the root of the problem. Human tendency is to get distracted when there's no conscious need to do otherwise. Just wondering.

Posted by: Cary Alburn | September 1, 2011 3:28 PM    Report this comment

Hand fly the departure (via FMS directed command bars)to cruise altitude and the descent/approach to landing and you've covered all the "training" you'll need. cruise via autopilot.

Posted by: Stephan Gnecco | September 1, 2011 4:02 PM    Report this comment

Instructing on several generations of Jet Transports, I found that too many pilots were fearful of manual flying out of distrust of their own skills, or a fear that "They" don't want us to do that; which was often an excuse. If you have never flown the aircraft with the Autopilot/Autothrust/Flight Director off, you will not be able to do so when one or more of these systems fail. A good pilot can fly the aircraft as well as the automation; sometimes better. The Air France Pilots were utterly unable to recognise the situation they were in and had no idea what to do. These pilots were not "Properly Trained and Fully Qualified" as has been asserted. At my airline, upsets are a regular part of the training curriculum.

Posted by: Brian Hope | September 1, 2011 4:11 PM    Report this comment

Brian sez: "At my airline, upsets are a regular part of the training curriculum."

As an airline passenger, I would hope that they would be part of every airline's training curriculum! I don't fly airliners, but I believe every pilot, regardless of the size of the airplanes they fly, should have upset training. In my case, I chose to take a basic aerobatics course more than 30 years ago, and there have been several occasions since when the ability to easily get out of an unusual attitude was important. Just because it's big iron doesn't make it immune to being upset, obviously.

Posted by: Cary Alburn | September 1, 2011 6:39 PM    Report this comment

Do jets and turboprops still have a dedicated AoA indicator on the panel? I wonder why AF447 and the Colgan air pilots did not make use of that instrument to confirm a stalled condition.

Posted by: scott dickey | September 1, 2011 6:51 PM    Report this comment

Scott, the answer is "NO". On the Airbus, our FMS has a page where I can look up the AOA, but you have to know where to look. It is also not practical to keep that page permanently up.

I hand fly the Airbus up to 10,000 feet then accelerate to climb speed. "George" can have it after that. On descent, I go manual once on vectors. In the Airbus, it is important to also kill the autothrust so I can get some thrust lever time too. That keeps the "scan" going (what we call the Airbus "stare"). None of this is in our policy and procedures manual. I do upset training in my own airplane.

Posted by: David Heberling | September 1, 2011 10:33 PM    Report this comment

"what we call the Airbus "stare"

Hah. Love that! Sounds like you have a good plan to stay in the loop scan wise. I'm sure some pilots don't do as much.

Posted by: Paul Bertorelli | September 2, 2011 9:37 AM    Report this comment

David, thanks for the answer. That's interesting that the AoA is not depicted in some little corner of the PFD. It would be nothing to have that very important raw data available. Airspeed is nice but AoA is where the rubber meets the road. It's hindsight for sure but I think had AF447 or Colgan seen the AoA clearly in the stalled condition maybe they would have stopped pulling back on the yoke...

Posted by: scott dickey | September 2, 2011 11:00 AM    Report this comment

"It would be nothing to have that very important raw data available. Airspeed is nice but AoA is where the rubber meets the road."

Agree Scott. It's hard to believe that every airliner (let alone every airplane) doesn't have an AOA indicator mounted at the top of the instrument panel right in front of the plot's eyes.

Not AOA displayed through the FMS on an electronic screen, but an actual AOA indictor showing -- as you said -- raw flight data. The one performance flight instrument that always shows exactly the flying condition of the jet and how close it is to a stall.

Posted by: Gary Dikkers | September 2, 2011 4:25 PM    Report this comment

"I hand fly the Airbus up to 10,000 feet then accelerate to climb speed. "George" can have it after that. On descent, I go manual once on vectors."

Excellent David. I like your attitude. I'd think every self-respecting airline pilot who takes pride in being called a pilot would do the same. Other than collecting a paycheck, what's the sense of sitting there and letting the autopilot fly the airplane?

I've never flown a airliner, but did fly fighters in the Air Force. We had an informal squadron policy that we didn't log flying time for the time on autopilot. I never used the autopilot, but did sometimes use the "altitude hold" function -- for example, when flying in weather and I needed to look down and behind me to find and pull a pub from the map case.

Posted by: Gary Dikkers | September 2, 2011 4:38 PM    Report this comment

"I hand fly the Airbus up to 10,000 feet then accelerate to climb speed. "George" can have it after that. On descent, I go manual once on vectors."

Excellent David. I like your attitude. I'd think every self-respecting airline pilot who takes pride in being called a pilot would do the same. Other than collecting a paycheck, what's the sense of sitting there and letting the autopilot fly the airplane?

I've never flown a airliner, but did fly fighters in the Air Force. We had an informal squadron policy that we didn't log flying time for the time on autopilot. I never used the autopilot, but did sometimes use the "altitude hold" function -- for example, when flying in weather and I needed to look down and behind me to find and pull a pub from the map case.

Posted by: Gary Dikkers | September 2, 2011 4:39 PM    Report this comment

"It would be nothing to have that very important raw data available. Airspeed is nice but AoA is where the rubber meets the road."

Agree Scott. It's hard to believe that every airliner (let alone every airplane) doesn't have an AOA indicator mounted at the top of the instrument panel right in front of the plot's eyes.

Not AOA displayed through the FMS on an electronic screen, but an actual AOA indictor showing -- as you said -- raw flight data. The one performance flight instrument that always shows exactly the flying condition of the jet and how close it is to a stall.

Posted by: Gary Dikkers | September 2, 2011 4:39 PM    Report this comment

I have no idea how those double entries got there. Monitor please remove.

Posted by: Gary Dikkers | September 2, 2011 5:01 PM    Report this comment

Scott, They had a whole PFD full of information. It meant nothing to them because they fixated on the erroneous airspeed indication. There are numerous examples of flight crews all fixating on one issue to the exclusion of flying the airplane from which they did not recover. Fixation is a killer.

Posted by: David Heberling | September 2, 2011 5:40 PM    Report this comment

Don't get me wrong, I believe automation is great at reducing work load and at performing monotonous boring tasks like the cruise portion of a flight. However, automation has to be monitored to make sure it is in the mode you selected, or is performing the task you programed it for. Pilots have to stay in the loop with the automation. It is in moments of inattention that mistakes are made. It is also too easy to make an entry error. That is why two sets of eyes are essential to verifying that the entry made was the correct one. Most GA pilots do not have the luxury of a two man crew. In that case, the lone pilot has to double check everything they do with their autopilot. Just because an autopilot has gone off on a tangent on you is no reason to not use them. Like everything else on an airplane, the autopilot has to be maintained to a high standard if it is to operate properly. Used properly, an autopilot can reduce the single pilot workload to a manageable level. However, the use of automation should not be used to the exclusion of piloting skills. Pilot are not machines. Because of that skills that are not used eventually atrophy. If pilots are to save the day when the automation takes a dump, you would think airlines would emphasize hand flying skills, upset training, and primary flight instrument failures. The FAA requires us to endlessly train for engine failures that we may never see in our careers. Automation failures are no where in the curriculum.

Posted by: David Heberling | September 2, 2011 6:26 PM    Report this comment

Your glaring mis-statement "We're forgetting how to fly" is only party correct. I think "Never learned properly in the first place is true also. The airlines WILL NEVER TEACH YOU 'HOW TO FLY'. They DO teach the NEW automation and landing approach systems. Some students should have a tattoo on their body sayin. 1. Never to fly instruments! 2. Too excitable to fly as PIC! 3. Unable to make solo landings! 4. Never to fly near thunderstorms! 5. Unable to follow simple rules! And I have not started on personality problems that make a pilot a minus in a crew. It was a great career, but I would not do it again with todays pay and hours required. I am just as good a mechanic could fix things quicker and more cheaply than most and still charge the going rate IN MY OWN BUSINESS! Today GET A BA IN A FIELD OF INTEREST BEFORE FLYING PROFESSIONALLY! 13,000 hrs J3 to L1011 EAL Arnie Allison

Posted by: Arnold Allison | September 3, 2011 12:56 PM    Report this comment

Like certain anatomical parts, everyone has an opinion. This problem has been under construction for a long time. The planes themselves are being designed to be flown by automation. I believe the intent was to make them so easy to operate that anyone could do it, thereby allowing downward pressure on pilot's pay. Certainly the plan for the third world was to put pilots in aircraft that would be operated in much the same way as any computer controlled machine. After a 28 year career flying for a prominent US carrier, I did a two year stint for an overseas airline. Most of my expat colleagues were also old guys, schooled in the basics of our craft. The local pilots, save a thin veneer of older military-trained aviators, were incredibly thin in experience with few basic flying skills. Most were very smart individuals, capable of rote memorization of an extensive list of technical documents. But when the scenario deviated from "The Plan", the analytical skills and basic instrument flying and aircraft handling ability was just not there. That there aren't far more accidents than there are is a miracle to me. While I may be picking on the overseas situation, from what I hear, and see in their recent incidents, my old airline is becoming much the same. All the above causal factors, coupled with an incredible lack of pride in the profession these days. Managements wanted to take pilots down, and they have succeeded. Most I know still working really don't give much of a damn.

Posted by: Bill Mcclure | September 5, 2011 7:56 AM    Report this comment

Brian McCulloch’s posting came closest to getting to the nub of this topic. “It is notable that the fully automatic drones now used by Air Forces across the world have had almost no crashes, and those they have had have been when some cowboy in the caravan tried to land it by hand. It will only be a matter of time before passenger aircraft get similar technology”. And said drones are frequently controlled by “operators” on a different continent. In summary, the day of the commercial pilot is over; most of them just refuse to recognise it yet. In response to the emergence of “pilot error” as the predominant cause of commercial aircraft accidents, manufacturers have been designing the pilot out of the system for years. And of course this also plays directly into the commercial interests of airline owners; much cheaper to have an “operator” sitting in a comfortable, air conditioned control room monitoring the automatic operation of a number of aircraft and able to interact on the very rare occasions when the multi-layered automation fails than pay multi-pilot crews in each aircraft. Only major problem left to solve is the psychological one of getting 500 plus passengers to get in the back if no one gets in the front. Maybe a job for the multitude of unemployed (and therefore cheap) actors who’d love to wear the uniform, shades etc and even make the announcements! But seriously, this is the way ahead and we do need to start thinking it through.

Posted by: Trevor J Murray | September 5, 2011 9:54 AM    Report this comment

Stategic Air Command Policy all thru the late 1960's and early seventies was No autopilot useage until you had logged 200 Hrs. of hand flying in the previous 12 months. That policy was derived from the blood and subsequent death of several USAF Generals attempting take off in a heavily laden KC-135 on Autopilot and without "recent' hands on flight experience, from Mc Connell AFB, Witchita Kansas in the late 60'S. They Crashed into the Cessna plant killing everyone on board. When I went thru KC-135 transition training we were given unusual attitudes at night above FL350 with the attitude indicator 'covered'. Real world experience is the Best Teacher!!

Posted by: Buz Allen | September 5, 2011 1:42 PM    Report this comment

When I started flying, there were actually some aircraft still equipped with the old Superhomer radios and NDB approaches were common. When we installed our Garmin 430 back in the mid-90's, I found that there was a huge learning curve. I felt that there was just too much going on in that one box! Anyone who thought that they could jump into approaches to minimums, etc without some practice was very much mistaken. I have no doubt that much confusion and distraction was caused by this. Similarly, when the new glass panels became available, there was even more of a learning curve. It took awhile, but the training community eventually addressed this. Now, it seems, that GA pilots, like our airline brethren, are being encouraged to let the autopilot do the hard stuff. I'm sorry, but I will never stop hand flying approaches. I don't get to do enough of them as it is, so I want to be ready when the bells and whistles crap out! Last year my mechanic told me of a customer of his who, while flying the same aircraft and autopilot as me, doing a coupled ILS, had his airplane suddenly enter an uncommanded, almost 180 degree roll inside of the marker in the clouds. I'll bet that he was happy that he was not just a video game player, and had the wits and skills to recover! My instructor still bombards me with partial panel and upset recovery while under the hood, as well as single radio (not GPS)orientation drills. It makes me sweat, but I wouldn't want it any other way.

Posted by: Steve Tobias | September 7, 2011 3:53 PM    Report this comment

Air France 447 meteorological conditions www.weathergraphics.com/tim/af447/

Take a look at this site scroll down the page, read and see for yourselves the conditions of the weather where AF 447 flew through. I've flown in conditions like that, I can tell you that the turbulence and the combination of the loss of the Standby Instruments for 20 seconds and not having any other attitude instrumentation to rely on is very disconcerting. They did not have a prayer in Heaven to succeed in recovering this aircraft.

140+ Computers along with multiple failure's chimes and bells trying to talk and re-synch with each other, hard to identify which has priority. Fly the aircraft first "Maintain Aircraft Control" is not there, where could they have gone? How could they confirm that their Standby Attitude Indicator was showing the correct indication when the aircraft is already in an unusual attitude to start with. Engines at Full Power at high altitude adding more nose up input to the wings in thin air due to the under-wing thrust mounted engines. How could they have at night inside these CB's in violent flight minimize disorientation?

Posted by: Rene Blanco | September 7, 2011 4:46 PM    Report this comment

I've followed this thread and read an awful lot of excuses. The First Rule of flight is "Fly The Airplane". It always applies. If you can't do that, you have no business in the front seat. Apparently, a lot of pilots haven't been in an unusual attitude since their first instruction, if then. Go get some upset recovery training.

Posted by: Sid Love | September 7, 2011 7:26 PM    Report this comment

Sid Love... I just checked your two post's here and I don't think you speak from experience. You mention basic principles that apply ALL of the time i.e. "Fly the Aircraft" Maintain Aircraft Control, Unusual Attitude Recovery for a cheap price of 6K's, etc.

I am not trying to be contentious or presumptuous, but this accident goes beyond their capability to maintain aircraft control. I am an Airbus Captain with 24,000+ hours and over 12000 in type. I flew A-7D's for 11+ years with around 1000+ in type. Lots of unusual attitudes, lots of loss of controls in the old Dog-Fight world. We do unusual attitude recoveries every year where I work. These guys did not have a chance or with what to recover their loss of control.

These are no excuses, these are well experienced opinions, slow down you are jumping to conclusions. Read and think, you may learn something that will help you some day.

Posted by: Rene Blanco | September 8, 2011 7:07 AM    Report this comment

I fear that the new generation of pilots may never get the opportunity to master their piloting skills. It takes many hours of hands on, pilotage, single pilot IFR and getting lost and finding your self without turning on the GPS to gain the confidence necessary to become a good pilot.

Posted by: Bryan Carricaburu | September 12, 2011 8:43 PM    Report this comment

Bryan, Very well said!

Posted by: Rene Blanco | September 12, 2011 10:18 PM    Report this comment

Well said Brian. I think in a perfect world, one could make a strong case why all pilots should first learn to fly a glider or sailplane before moving to powered-flight. A sailplane is the ultimate test of a pilot's skills -- there are no second chances.

Posted by: Gary Dikkers | September 13, 2011 9:26 AM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?

Register

Enter your information below to begin your FREE registration

« Back to Full Story