Automation Cautionary Tale
Thread Starter
Join Date: Jun 2006
Location: Australia
Posts: 1,186
Likes: 0
Received 0 Likes
on
0 Posts
Automation Cautionary Tale
Seems to me those of us heretics who enjoy the pleasure of keeping our manual flying skills in top order are in a minority and it will ever be so.
Diverging views on automation dependency has been with the airline industry since the first glass cockpit 737 came into service in the 1980's.
In a recent issue of Aviation Week & Space technology, correspondent Richard Hipper writes to the Editor and says: Quote:
"The deterioration of hand-flying skills dates to the airline industry's "pushing" automation on pilots as early as 1990 (AW&ST Feb 1-14 p.51). In 1991, I retired after 30 years of military and airline flying. I used to hand-fly airliners to altitude before engaging the autopilot for cruise, then hand-fly to landing, opting for the autopilot when required by company policy.
One IMC night in 1990 - during a flight into the New York area - I was getting a line check from a company FAA designee. I hand-flew most of the flight, including entering a holding pattern and ending with an ILS approach to minimums.
Even though I held altitude within 20 feet and airspeed within five knots, the check pilots critique was :"You have to use the autopilot more." When I asked if there was anything wrong with my flying, he said "no", but I should use the autopilot more. Sadly, up-and-coming pilots might never get that great feeling of being "one" with their aircraft." Unquote
His letter took me back to the time in 1989 when I underwent my first British instrument rating which was conducted by a senior CAA Examiner. Having already flown over 5000 hours in command on the Boeing 737-200, I found the test straight forward and the Examiner seemed happy with the result. The 737 simulator had a Collins FD108 FD which was quite new in those days but old fashioned by comparison with some of today's flight directors.
In fact, most of the pilots in my previous airline preferred to fly without the use of the FD 108 because it was not that accurate and you had to continually fiddle with its settings, especially in the pitch mode. For that reason I opted not to use the FD108 for my instrument rating renewal with the CAA.
At the debriefing, the CAA Examiner asked why I did not use the FD for the test. I told him I felt could fly more accurately without it as my instrument scan was faster than just concentrating on one flight instrument. He didn't agree, saying although my flying smooth and well within tolerances, I should use the FD in future as a matter of policy. We parted ways amicably with my brand new British ATPL firmly clutched in my hand. He had his opinion and I had mine.
Recently, I saw a copy of the simulator assessment sequences used for a major Japanese airline. One paragraph caught my attention. It said candidates must use the flight director until touch down on an ILS. Also, that the autopilot can be used in the test. Once again, the term automation addiction came to mind.
I would have thought that a company wishing to assess a prospective candidates pure flying ability, both on instruments and visual flight, would require the candidate to demonstrate he could fly within required tolerances without the aid of an autopilot and flight director.
Diverging views on automation dependency has been with the airline industry since the first glass cockpit 737 came into service in the 1980's.
In a recent issue of Aviation Week & Space technology, correspondent Richard Hipper writes to the Editor and says: Quote:
"The deterioration of hand-flying skills dates to the airline industry's "pushing" automation on pilots as early as 1990 (AW&ST Feb 1-14 p.51). In 1991, I retired after 30 years of military and airline flying. I used to hand-fly airliners to altitude before engaging the autopilot for cruise, then hand-fly to landing, opting for the autopilot when required by company policy.
One IMC night in 1990 - during a flight into the New York area - I was getting a line check from a company FAA designee. I hand-flew most of the flight, including entering a holding pattern and ending with an ILS approach to minimums.
Even though I held altitude within 20 feet and airspeed within five knots, the check pilots critique was :"You have to use the autopilot more." When I asked if there was anything wrong with my flying, he said "no", but I should use the autopilot more. Sadly, up-and-coming pilots might never get that great feeling of being "one" with their aircraft." Unquote
His letter took me back to the time in 1989 when I underwent my first British instrument rating which was conducted by a senior CAA Examiner. Having already flown over 5000 hours in command on the Boeing 737-200, I found the test straight forward and the Examiner seemed happy with the result. The 737 simulator had a Collins FD108 FD which was quite new in those days but old fashioned by comparison with some of today's flight directors.
In fact, most of the pilots in my previous airline preferred to fly without the use of the FD 108 because it was not that accurate and you had to continually fiddle with its settings, especially in the pitch mode. For that reason I opted not to use the FD108 for my instrument rating renewal with the CAA.
At the debriefing, the CAA Examiner asked why I did not use the FD for the test. I told him I felt could fly more accurately without it as my instrument scan was faster than just concentrating on one flight instrument. He didn't agree, saying although my flying smooth and well within tolerances, I should use the FD in future as a matter of policy. We parted ways amicably with my brand new British ATPL firmly clutched in my hand. He had his opinion and I had mine.
Recently, I saw a copy of the simulator assessment sequences used for a major Japanese airline. One paragraph caught my attention. It said candidates must use the flight director until touch down on an ILS. Also, that the autopilot can be used in the test. Once again, the term automation addiction came to mind.
I would have thought that a company wishing to assess a prospective candidates pure flying ability, both on instruments and visual flight, would require the candidate to demonstrate he could fly within required tolerances without the aid of an autopilot and flight director.
Last edited by Tee Emm; 23rd Feb 2016 at 10:38.
Join Date: Dec 2008
Location: Birmingham
Age: 39
Posts: 181
Likes: 0
Received 0 Likes
on
0 Posts
Sorry,
I will be honest. In my company we have really good pilots and they fly a lot of handflying here. So tell me why is it necessary to fly by hand to the toc and holding? Makes for me no sense. I agree that we should practice hand flying but to which degree? Would you fly a raw data ils to minimums with minimum weather just that you can say: wow I'm the best jockey on the bird? Wouldn't let sit my family into this bird. Just be professional and use the assistance for what it is made for. 6 days and 18 sectors and you fly the last leg raw data into a snow storm in Russia? Jesus....
So finally don't be selfish and act like a pro and I get my satisfaction from other things in life.
Sorry that must be said
I will be honest. In my company we have really good pilots and they fly a lot of handflying here. So tell me why is it necessary to fly by hand to the toc and holding? Makes for me no sense. I agree that we should practice hand flying but to which degree? Would you fly a raw data ils to minimums with minimum weather just that you can say: wow I'm the best jockey on the bird? Wouldn't let sit my family into this bird. Just be professional and use the assistance for what it is made for. 6 days and 18 sectors and you fly the last leg raw data into a snow storm in Russia? Jesus....
So finally don't be selfish and act like a pro and I get my satisfaction from other things in life.
Sorry that must be said
Only half a speed-brake
Devils advocate too:
- over here RVR 550 m is only allowed with FDs (CAT I approach lights - EASA);
- workload management is a fair and vital assesment point, more so for the captains;
- check rides in SIM are not scheduled for the reason of brushing up or showing off one's manual handling skills.
- over here RVR 550 m is only allowed with FDs (CAT I approach lights - EASA);
- workload management is a fair and vital assesment point, more so for the captains;
- check rides in SIM are not scheduled for the reason of brushing up or showing off one's manual handling skills.
Join Date: Mar 2008
Location: Above the Transition Level
Posts: 160
Likes: 0
Received 0 Likes
on
0 Posts
Sure there's a time and a place for hand flying. But i agree with Tee in that we as pilots should be able to hand fly to that level of proficientcy and never be in a situation that our scans have deteriorated to a point we feel uncomfortable hand flying.
Unfortunately not all can anymore with company policies dictating the level of automation that must be flown because "it's safer" or "reduces the amount of unstable approaches".
Unfortunately not all can anymore with company policies dictating the level of automation that must be flown because "it's safer" or "reduces the amount of unstable approaches".
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
This hot topic is being beaten to death on many occasions via various threads. It is becoming circular; and in the end our opinion will count for nought. The powers that be, including XAA's DFO's, CP's & HOT's will decide, and I doubt the XAA's will dare to get involved too directly.
However, there does seem a consensus that manual flying skills and especially scanning skills have deteriorated to be close to danger levels. I agree with those who say hand flying during a climb at constant power following an FD is not going to maintain skills. There are too few changes. Below FL100 to landing is quite different, and traffic density and terrain/weather difficulty must be taken into consideration before 'practicing'.
What are the opinions on these:
1. manual flying allowed (not necessarily encouraged) but no FD is forbidden.
2. visual approaches allowed (not encouraged) but no finals inside OM and LNAV/VNAV data must be inserted for OM waypoint.
3. visual approaches policy is preferred to be flown with automatics.
4. If flown manually with LNAV/VNAV guidance A/T is encouraged to prevent low speed excursions.
Do these impositions encourage the maintenance of basic flying skills? or has this attitude........
Unfortunately not all can do so anymore with company policies dictating the level of automation that must be flown because "it's safer" or "reduces the amount of unstable approaches".
now become the norm? We have read that FAA, NTSB and even Airbus is beginning to advocate more manual flying to combat auto-dependancy. If airlines don't follow this philosophy what chance is there? FBW will be the norm for all pax jets within 15 years. There is that amount of time to sort this out.
However, there does seem a consensus that manual flying skills and especially scanning skills have deteriorated to be close to danger levels. I agree with those who say hand flying during a climb at constant power following an FD is not going to maintain skills. There are too few changes. Below FL100 to landing is quite different, and traffic density and terrain/weather difficulty must be taken into consideration before 'practicing'.
What are the opinions on these:
1. manual flying allowed (not necessarily encouraged) but no FD is forbidden.
2. visual approaches allowed (not encouraged) but no finals inside OM and LNAV/VNAV data must be inserted for OM waypoint.
3. visual approaches policy is preferred to be flown with automatics.
4. If flown manually with LNAV/VNAV guidance A/T is encouraged to prevent low speed excursions.
Do these impositions encourage the maintenance of basic flying skills? or has this attitude........
Unfortunately not all can do so anymore with company policies dictating the level of automation that must be flown because "it's safer" or "reduces the amount of unstable approaches".
now become the norm? We have read that FAA, NTSB and even Airbus is beginning to advocate more manual flying to combat auto-dependancy. If airlines don't follow this philosophy what chance is there? FBW will be the norm for all pax jets within 15 years. There is that amount of time to sort this out.
Join Date: Sep 2011
Location: FL390
Posts: 240
Likes: 0
Received 0 Likes
on
0 Posts
One IMC night in 1990 - during a flight into the New York area - I was getting a line check from a company FAA designee. I hand-flew most of the flight, including entering a holding pattern and ending with an ILS approach to minimums.
By hand-flying from cruise, into a hold in busy airspace, and then an ILS to minima, what you actually did was massively increase the workload on your colleague who has to actively monitor every moment of your flying in addition to their other tasks. Have they flown with you before? Did you tell them you'd be hand-flying the whole thing during the brief?
So nice day, quiet airfield, relaxed flight deck - I'll hand fly from top of descent or vice-versa. When the external factors start to mount up, then the autopilot and flight directors are there as tools to be used.
Amen. By now that horse has been killed, subsequently severely belabored, butchered, turned into steak, devoured and excreted.
Agree, what's sad and quite alarming at the same time is the Automation fans disparagement of those who choose to keep up their manual flying skills.
Probably because theirs are lacking.
Probably because theirs are lacking.
Yes that's all very well, but I think Fursty has it right.
In terms of threat and error management, hand flying an aircraft for long periods in busy airspace increases the potential for making errors, and is therefore questionable in terms of safety. PM will have a doubled workload and PF will have reduced spare capacity for monitoring aircraft systems and situational awareness.
Above FL200, cabin crew will be coming into the cockpit, and the potential for these 'distractions' to lead to flying errors - especially in RVSM airspace - is also increased.
Now that we do not have navigators or engineers on the flight deck, the two remaining pilots need to do more than just fly, and this is probably one reason for the increase in the use of automatics - to free us up a little. This of course has the unfortunate side effect of reducing our practice in hand flying and we need to do something to mitigate against this.
Being required to perform and record manually flown approaches with no F/D and no A/THR on good weather days, (in a similar way that we used to do for our practice autolands) would be an easy way to start this process.
In terms of threat and error management, hand flying an aircraft for long periods in busy airspace increases the potential for making errors, and is therefore questionable in terms of safety. PM will have a doubled workload and PF will have reduced spare capacity for monitoring aircraft systems and situational awareness.
Above FL200, cabin crew will be coming into the cockpit, and the potential for these 'distractions' to lead to flying errors - especially in RVSM airspace - is also increased.
Now that we do not have navigators or engineers on the flight deck, the two remaining pilots need to do more than just fly, and this is probably one reason for the increase in the use of automatics - to free us up a little. This of course has the unfortunate side effect of reducing our practice in hand flying and we need to do something to mitigate against this.
Being required to perform and record manually flown approaches with no F/D and no A/THR on good weather days, (in a similar way that we used to do for our practice autolands) would be an easy way to start this process.
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
Being required to perform and record manually flown approaches with no F/D and no A/THR on good weather days, (in a similar way that we used to do for our practice autolands) would be an easy way to start this process.
Except there are some major airlines that strictly forbid such an act of heresy: sadly.
Except there are some major airlines that strictly forbid such an act of heresy: sadly.
It's perfectly understandable that airline policy would want the computer to fly the plane - it can do a much better job of it - smoother, more efficiently and it lowers the pilots' workload. All fair points, so why do we even need pilots? Because we like to think that when the computers go wrong, the pilots will take over and save the day. Except they can't, because either they've forgotten how to, or worse still they've never actually (hand) flown a jet because their airline won't let them. Does anyone else think that is crazy? Surely there's a happy medium here. If we need 3 takeoffs and 3 landings every 90 days, surely that can be extended to include 3 hand flown climbs and descents and 3 hours of cruise? (PICs discretion, obviously. If things start to get busy then AP on and scrub that one.)
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
The circle keeps turning; but here we go.
We the pilots like to think we can handle the problem when HAL goes AWOL, but some can and some can't depending on who you work for, their training & culture, and your own attitude as well. The pax sure as heck expect us to be able to handle it; they also expect us not to screw up in our interaction with HAL. Either way they expect us to be in charge & capable.
The airlines want a safe & efficient operation that takes care of the pax and makes a healthy profit without scaring anyone. So the techies design ever more all singing all dancing a/c that have so many back-ups and 10 times more reliability than ever thought necessary. They then sell these to the nervous & greedy airline managers who then decide that the best way is to use all the toys and keep the pilots dump and be simply airborne train drivers. If reliability holds up its part of the deal then case proven. The pax are none the wiser and they travelled A-B in total bliss; that is until the pilot is called upon to perform and does not fulfil their side of the bargain. And whose fault was that? Was it really pilot error or something more obtuse and dare I say it, "not to be spoken about even in whispers." Could it have been a corporate culture failing? OMG, how could that be? They passed their 6 monthly tick in the box checks from yester-year. They were legal. Can't be our fault; must have been the sharp end jet jockeys.
It's very rare the AAIB/NTSB points a finger directly at corporate company culture as a root cause of an accident. They often say the crew had not had a specific training that might have helped them avoid this particular accident. You can't think of every eventuality, but if the training is broad, in depth, comprehensive and the crews are well practiced in the science of a/c handling, systems knowledge & management and the black art of airmanship then they will stand a solid chance of coping with many surprises. (The beneficial results of that have been demonstrated in some cases, but that trait has not been heralded as an industry-wide requirement.)
Sadly that training philosophy is no longer the case and many operate to minimum standards and quotas. The techies are left to come up with the best & cheapest solutions. That's what sells a/c.
We the pilots like to think we can handle the problem when HAL goes AWOL, but some can and some can't depending on who you work for, their training & culture, and your own attitude as well. The pax sure as heck expect us to be able to handle it; they also expect us not to screw up in our interaction with HAL. Either way they expect us to be in charge & capable.
The airlines want a safe & efficient operation that takes care of the pax and makes a healthy profit without scaring anyone. So the techies design ever more all singing all dancing a/c that have so many back-ups and 10 times more reliability than ever thought necessary. They then sell these to the nervous & greedy airline managers who then decide that the best way is to use all the toys and keep the pilots dump and be simply airborne train drivers. If reliability holds up its part of the deal then case proven. The pax are none the wiser and they travelled A-B in total bliss; that is until the pilot is called upon to perform and does not fulfil their side of the bargain. And whose fault was that? Was it really pilot error or something more obtuse and dare I say it, "not to be spoken about even in whispers." Could it have been a corporate culture failing? OMG, how could that be? They passed their 6 monthly tick in the box checks from yester-year. They were legal. Can't be our fault; must have been the sharp end jet jockeys.
It's very rare the AAIB/NTSB points a finger directly at corporate company culture as a root cause of an accident. They often say the crew had not had a specific training that might have helped them avoid this particular accident. You can't think of every eventuality, but if the training is broad, in depth, comprehensive and the crews are well practiced in the science of a/c handling, systems knowledge & management and the black art of airmanship then they will stand a solid chance of coping with many surprises. (The beneficial results of that have been demonstrated in some cases, but that trait has not been heralded as an industry-wide requirement.)
Sadly that training philosophy is no longer the case and many operate to minimum standards and quotas. The techies are left to come up with the best & cheapest solutions. That's what sells a/c.
Only half a speed-brake
Maybe the effectiveness could be raised. To start, let's begin by not calling names such as "Automation fans", "opponents of keeping up manual flying skills", "Probably lacking manual skills" at people who state unwelcome facts. That's ad hominem, in my best French. stilton, care to join?
Join Date: Jun 2011
Location: Devonshire
Age: 96
Posts: 297
Likes: 0
Received 0 Likes
on
0 Posts
As a pilot I was checked periodically on my abilities to cope if ever something went wrong (engine(s) failed or flaps not work properly etc.) A/Ps are very much more reliable than they used to be on a York, for example.
But the same principles ought to apply, surely? Different speeds and weather minima may be applicable for landing, but this is, I assume, an exceptional circumstance...
(Stalling was done on the B170, DC7c and Britannia but at a safe height, lightly loaded or empty.. briefed by Training Captain who sat in RHS and done gradually. Others may notice differences. All three types had "straight rigid wings" and were well below F/L 15.0 )
LT
But the same principles ought to apply, surely? Different speeds and weather minima may be applicable for landing, but this is, I assume, an exceptional circumstance...
(Stalling was done on the B170, DC7c and Britannia but at a safe height, lightly loaded or empty.. briefed by Training Captain who sat in RHS and done gradually. Others may notice differences. All three types had "straight rigid wings" and were well below F/L 15.0 )
LT
Last edited by Linktrained; 25th Feb 2016 at 16:06. Reason: stalls
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
It is quite sensible that this discussion should not separate along the lines of the autonomies and the hands & legs brigade. Surely the industry should striving to create rounded pilots. Those who can handle both techniques with relaxed aplomb.
Some operators prefer the trained monkeys and have rigid narrow SOP's to ensure that: others encourage crews to be more a pilot than a robot. However, do they have a full understanding of how to use the automatics and when to use them to their best benefit?
What is needed is the middle ground; a pilot who has the ability to do both and the common sense to choose when to do which. That has to come from the training department day 1.
Future pilots may well have a different route to mine. I was brought up in GA and then in a piloting airline who trained and trusted its crews to make the choice I suggested. It was then a little frustrating to join an airline who lent towards the trained monkey method, but did allow those who could to do so. It was always a pleasure to demonstrate and help those who wanted to learn: equally disappointing to be with those who didn't. The real irritation came when the nomadic road took me to a straight-jacketed trained monkey outfit.
There is definitely not an industry wide standard and it will not be achieved unless the authorities stipulate it. What I find astonishing, and curious, is I suspect that many of the TM's fly with high productivity multi-sector per day short-haul operators. You'd expect the opposite, but...........
Some operators prefer the trained monkeys and have rigid narrow SOP's to ensure that: others encourage crews to be more a pilot than a robot. However, do they have a full understanding of how to use the automatics and when to use them to their best benefit?
What is needed is the middle ground; a pilot who has the ability to do both and the common sense to choose when to do which. That has to come from the training department day 1.
Future pilots may well have a different route to mine. I was brought up in GA and then in a piloting airline who trained and trusted its crews to make the choice I suggested. It was then a little frustrating to join an airline who lent towards the trained monkey method, but did allow those who could to do so. It was always a pleasure to demonstrate and help those who wanted to learn: equally disappointing to be with those who didn't. The real irritation came when the nomadic road took me to a straight-jacketed trained monkey outfit.
There is definitely not an industry wide standard and it will not be achieved unless the authorities stipulate it. What I find astonishing, and curious, is I suspect that many of the TM's fly with high productivity multi-sector per day short-haul operators. You'd expect the opposite, but...........
Join Date: Jun 2011
Location: Devonshire
Age: 96
Posts: 297
Likes: 0
Received 0 Likes
on
0 Posts
The Cross Channel Car Ferry might be considered a Multi Sector operation with up to a dozen sectors flown per day, by one crew.
Lydd to Le Touquet was flown at 1000ft. with the return at 1500ft. The minimum AP level was also 1500ft. This allowed less than 5 minutes use of AP per 20 minute flight, inbound, ONLY !
Some did not bother for so short a time !
LT
Lydd to Le Touquet was flown at 1000ft. with the return at 1500ft. The minimum AP level was also 1500ft. This allowed less than 5 minutes use of AP per 20 minute flight, inbound, ONLY !
Some did not bother for so short a time !
LT
Last edited by Linktrained; 25th Feb 2016 at 22:43. Reason: bits
Future Industry Embarrassment
Regardless of what one thinks, there have been some fatal crashes in recent years where aircraft that are mostly or completely airworthy have been flown into the sea or ground by pilots who have forgotten how to fly and land.
Put that in your actuarial pipe and smoke it.
There's a good chance that this type of crash will come to dominate the statistics. Insurance pay outs and fines will always be far larger for such avoidable crashes than ones which are caused by overwhelming weather, unavoidable equipment failure, acts of god, etc.
Put that in your actuarial pipe and smoke it.
There's a good chance that this type of crash will come to dominate the statistics. Insurance pay outs and fines will always be far larger for such avoidable crashes than ones which are caused by overwhelming weather, unavoidable equipment failure, acts of god, etc.
Yes. The industry "experiment" of minimal training and practise (sic) of pilots to reduce overheads in modern autojets has surely been shown to be literally fatal?
And time for the XAAs or ICAO to say "enough cost cutting already, it's getting dangerous"?
And time for the XAAs or ICAO to say "enough cost cutting already, it's getting dangerous"?