Autopilots may dull skills of pilots, committee says

[cdnpilot77]

http://www.cnn.com/2011/TRAVEL/09/01/ai ... ?hpt=hp_t2

Washington (CNN) -- Long hours flying under computer control may have dulled the skills of airline flight crews, according to a U.S. advisory board that recommends more manual flight time for pilots.
"They're becoming very dependent upon using the autopilot, the auto-throttles, the auto flight system, the computers, to actually operate the entire flight," said Kevin Hiatt, a former airline pilot who sat on that board.
"What happens is, you don't actually hand-fly or manipulate the controls, whether it's a control yoke or a sidestick controller," Hiatt said. "Therefore, your computer skills get greatly enhanced, but your flying skills start to get rusty."
Those concerns were highlighted by the 2009 crash of Air France Flight 447 off the coast of Brazil.
When the crew received warnings that the aircraft was stalling high above the Atlantic Ocean and the autopilot shut down, the co-pilot started pulling the nose up -- exactly the opposite of what he was supposed to do. When the pilot returned to the flight deck to correct him, it was too late.
French investigators found that the pilots had failed to discuss earlier stall warnings and had received no high-altitude training to correct the problem.
All 228 people aboard the Airbus A330 from Rio de Janeiro to Paris were killed in the crash. Investigators say that more than 70 bodies will never be recovered.
The same January, a Colgan Air commuter plane crashed in icy conditions on its approach to Buffalo, New York, killing 50 people. Investigators said the pilot had countermanded the aircraft's system recommendations, something Hiatt and his panel say suggests that he had forgotten some key procedures.
The U.S. advisory committee, commissioned by Congress and working under the Federal Aviation Administration, found that jetliner crews are relying too much on autopilot. In some cases, pilots have the stick for less than three minutes during takeoffs and landings, and landings also can be done on autopilot, Hiatt said.
The committee found that the problem is not the fault of the industry or regulators but the result of evolving technologies and standards over the years.
The FAA would not comment on the recommendations, which were issued in late July. The Airline Pilots Association, the union that represents flight officers on 39 North American airlines, said airline safety "is a testament to the high levels of skill brought to the cockpit by the professional airline pilot."
The panel recommended that airlines provide guidance for manual flights in their operating manuals to encourage more actual flying by pilots. But experts say the problem may get worse because of the way younger pilots are trained.
"When you bring on a new pilot who has not been through some of the things that some of the older guys have, they've never flown an airplane that had anything but some computer activity on it," retired commercial pilot Jim Tilmon said. "They don't understand what to do necessarily when something goes wrong with their computer."

[glorifiedtaxidriver]

No shit. They didn't need a study to figure that out.

[FL020]

I have to laugh and shake my head..... It's as bad as the cdn government!!
We needed a multiple thousand dollar survey for CNN to broadcast and say that pilots are losing skills because a/p's are driving the airplane? Come on..

[DanWEC]

Pilot error is a component of accidents. Automation has decreased the overall accident rate- BUT. I think we may see very, very dangerous total fatality rate increase when the pilots in question (us) don't adequately deal with an emergency situation- due to simple lack of practice.
In other words, we are better equipped to deal with 95% of problems, but the last 5% might result in a total fatality when otherwise it could be mitigtated.

Despite the procedural complication, is it crazy to impose regulations like, say, 10% of every flight must be hand flown? IE one ascent/decent, or one heading change of all flights must be done manually? It just gets the blood flowing in the cerebrum. I don't suppose any pilot would really be averse to actually flying his bird for a few minutes after a few hours of monitoring it. Unless, it adds more risk than necessary... in which case we have become woefully obselete.

[Blakey]

I don't doubt the results of the study but in my experience as a check pilot, I've seen a lot more problems with pilots who did not know what the fancy Auto-Pilot was doing because they didn't spend enough time using it to operate the plane. We lost a Cormorant due to this.

Catch 22 I guess. More flying in general is the way to better skills but we certainly don't need a study to tell us that either!

[glorifiedtaxidriver]

I don't doubt the results of the study but in my experience as a check pilot, I've seen a lot more problems with pilots who did not know what the fancy Auto-Pilot was doing because they didn't spend enough time using it to operate the plane. We lost a Cormorant due to this.

Catch 22 I guess. More flying in general is the way to better skills but we certainly don't need a study to tell us that either!

I can see that, but what I see happening now is a lot of guys starting out with 200 hrs say right seat in a PC 12, then maybe right seat in a -8 or some other twin with an autopilot. From there they go to air canda or porter or jazz or something like that with no real hand flying time. Autopilots can be learned, but there's no substitute for good old fashioned hands and feet. You can't learn that in a month or two.

[Doc]

Totally disagree with the "study". Auto pilots give pilots a complete "overview" of the situation. An auto-pilot is an indispensable aid to pilots. Companies who special order modern A/C without auto-pilots because "we pay our pilots to fly..." are just kidding themselves. Sure, when everything goes for a crap, one should actually know how to fly.....are you listening AIR FRANCE????

[Bede]

Doc, you have a point, but you're presenting a false dichotomy. No one is suggesting no autopilots, just competence with both hand flying and automation. I would be willing to bet you already possess the hand flying skills because of your background, but like another poster said, if you start in a PC12 and work your way up, there is a chance your skills won't be as developed.

Bad hands and feet is what cause approach and landing accidents.

[Doc]

Doc, you have a point, but you're presenting a false dichotomy. No one is suggesting no autopilots, just competence with both hand flying and automation. I would be willing to bet you already possess the hand flying skills because of your background, but like another poster said, if you start in a PC12 and work your way up, there is a chance your skills won't be as developed.

Bad hands and feet is what cause approach and landing accidents.

Good point. I'd be willing to bet the average airline pilot who came up through the burgeoning "cadet" programme used in some European airlines couldn't hand fly a 172! This will become a real problem...I guess it already has?
When I read the "reports" on the Air France fiasco, I was physically ill! At least us old guys can turn off the "magic" and fly the puppy.

[Tiny Tyke]

Part of the problem is, if you try to hand-fly with some Captains, they get very nervous. Could be because they aren't comfortable hand-flying themselves.

Also, some company SOPs are overly restrictive as to what you can/can't do when you are hand flying. These procedures, designed for safety, discourage the development/maintenance of hand-flying skill. Additionally, SOPs should be designed to keep the flight deck flow the same, with or without the AP engaged.

Ex. - Most companies say that the PF selects the ALT Selector when the AP is engaged, but the PM does it when the AP not engaged. Couldn't the PM just do it all the time?? (leave the pilots some discretion in high workload scenarios)

[jjj]

Part of the problem is, if you try to hand-fly with some Captains, they get very nervous. Could be because they aren't comfortable hand-flying themselves.

Also, some company SOPs are overly restrictive as to what you can/can't do when you are hand flying. These procedures, designed for safety, discourage the development/maintenance of hand-flying skill. Additionally, SOPs should be designed to keep the flight deck flow the same, with or without the AP engaged.

Bingo Tyke.

You can spin the dials on the MCP, talk on the radios, make entries into the FMS and so on while your co-jo is in the lav.

Take away the auto-pilot and airline guys aren't even allowed to turn 10 degrees left on their own.



JJJ

[Doc]

Also, some company SOPs are overly restrictive as to what you can/can't do when you are hand flying. These procedures, designed for safety, discourage the development/maintenance of hand-flying skill. Additionally, SOPs should be designed to keep the flight deck flow the same, with or without the AP engaged.
)

When your stall recovery technique has you descending for THIRTY SOME ODD THOUSAND FEET, it's time to toss the SOP in the trash and try something else????
Are we perhaps adhering a little too closely to SOP's? Are SOP's actually killing pilots? You HAVE to know when to deviate from an SOP! Hate to oversimplify Air France's problem, but it seems a simple thing to lower the nose for some flying speed, then maintain a cruise deck angle with cruise N1. Guess that wasn't in their SOP?

[SAR_YQQ]

Hate to oversimplify Air France's problem, but it seems a simple thing to lower the nose for some flying speed, then maintain a cruise deck angle with cruise N1. Guess that wasn't in their SOP?

Stall recovery on a heavy is not as simple as that - sure the physics remain the same, but the overall momentum of a heavy jet is not something that can be instantly overcome. I was in the sim on friday, took a 270,000lbs jet to the buffet with AP engaged - perfect recovery had us losing 1200ft before we had a positive rate of climb. This was accomplished using properly functioning airspeed and AOA indications.
I'm not saying that the AF crew didn't mess up - but this isn't a prop job where "Pressure, Power, Rudder, Level, Climb" works with less that 20' loss.

Ever try flying an aircraft 14 hours without any sort of automation? I imagine that would be exhausting. FD/AP is the best hands and feet pilot that exists - you provide the brain.

[pika]

Bad hands and feet is what cause approach and landing accidents.

How so?

Here's some general guidance on what appears to be a major contributing factor.

http://www.mtc.gob.pe/portal/transporte ... sf_rat.pdf

[Beefitarian]

SAR_YQQ I just would like to point out Doc advocated for the use of auto pilot a few posts earlier. Let's all be friendly.

I think everyone who has flown even small planes has been in a situation where they get busy. We all understand the value of auto pilots and FMS as a tool.

There is a problem when people are only flying for a few minutes per flight and getting to a point where they are nervous flying the plane they are in charge of.

1200' was a perfect recovery. A sloppy one should only take 5000 leaving 25000 of the thirty. I am left wondering how much of their altitude was taken by downdrafts? The start of that incident was a giant cloud was it not?

[SAR_YQQ]

SAR_YQQ I just would like to point out Doc advocated for the use of auto pilot a few posts earlier. Let's all be friendly.

No malice was intended in my post.

[Beefitarian]

Just hoping to keep a light mood going.

I'm still kind of frightened about this..

Part of the problem is, if you try to hand-fly with some Captains, they get very nervous. Could be because they aren't comfortable hand-flying themselves.

I am supposed to ride for vacation in February, I hope our crew is current on type.

[SuperchargedRS]

The first person that blew a check ride during my time as a CFI was a lufthansa capt who was trying to get his single engine ATP over here before he had to fly back. He busted the MDA on an RNAV by 150' This was in a Cessna with a G1000. Changed the way I feel about hours and ratings for sure!

[bcflyer]

There is a problem when people are only flying for a few minutes per flight and getting to a point where they are nervous flying the plane they are in charge of.

1200' was a perfect recovery. A sloppy one should only take 5000 leaving 25000 of the thirty. I am left wondering how much of their altitude was taken by downdrafts? The start of that incident was a giant cloud was it not?

While I am a HUGE advocate of being able to hand fly the airplane (and do so on a regular basis with no nervous looks from anyone I fly with) recovery from a full stall at high altitude can easily take 5000' or more before the plane starts to fly. We did an exercise in the sim that involved a FULL stall (not just approach to the stall and recover at the buffet) at 37000'. We knew it was coming and were fully prepared for the recovery. What we weren't prepared for was the fact that it took almost 6000' before we were flying again. While it certainly doesn't excuse the actions of the AF crew it was a very eye opening demonstration on the effects of high altitude on stalls. Its not just a simple lower the nose, add power and all is good. It takes a long long time......