This is not about better stick and rudder skills.

[Northbeach]

Voss Says Pilots Must Back Up Automation

"Five years ago we passed the point where automation was there to back up pilots," said Flight Safety Foundation CEO Bill Voss at last week's Flight Safety Foundation Corporate Aviation Safety Seminar in San Antonio. "Clearly today, the pilot is there to be the backup to the automation." Voss told AIN "This is simply a realistic assessment of the world today,

Without being present to hear and see Mr. Voss deliver his remarks in context it is difficult to comment. Additionally who am I (just a 15K hour line pilot) compared to the CEO of a prestigious institution such as Flight Safety Foundation?

However I need to ask what happened 5 years ago that resulted in this monumental shift where the industry went from the automation's job was to back up the pilot,to now the pilot backing up the automation? I missed that memo (wouldn't be the first time that happened).

Clearly today, the pilot is there to be the backup to the automation

I do not ascribe to the view that my flying partner (Captain in training) and I are there to "back up the automation". My own myopic view of the industry worldwide is limited to my experience with my company; perhaps Mr.Voss has a difference audience or reference in mind.

"Clearly today, the pilot is there to be the backup to the automation." Voss told AIN "This is simply a realistic assessment of the world today, except we are not training pilots to be backups to automation. We have to own up to the fact that we need develop new kinds of pilot training," he said.

If I correctly understand Mr. Voss' comments he is advocating a new kind of pilot training aimed at producing pilots who backup the automation. If I understand his premise and intent correctly then I would have to disagree. I run the automation and the flight, I access all the resources available to me including the automation, but I am in control here; not the programmers, the software version or some remote drone operator buried in a corporate bunker somewhere.



Concerning raw flyin skills, I was never better than I was when I was flying singlepilot multi-engine hard IFR with no autopilot in and out of busy commercial hubs and into remote non-radar areas 6 days a week. My "stick & rudder" skill were honed. I was not able to move from 250 hours and a commercial license to the right seat of an Airbus or Boeing. In retrospect I am glad I had to put in the years flying on demand charter, freight and corporate before moving on to the airline job. That experience and training has made me a better pilot. Of course the extra 12 to 15 years of seniority, if Ihad skipped the intermediate steps, would be nice.

[Keylime]

Part of the problem facing this industry is the lack of skilled, experienced pilots. Worldwide, the shortage that is already upon us is being sold to operators as "You can get away with a less experienced pilot because the automation will save you". And, it will 90 something % of the time.

If you look at the majority of major accidents, many of them(not all) are in the entry level aircraft and in the majors in the Airbus 320 family and B-737 types. Why is this? This is the entry level aircraft. Because of this shortage, pilots that years ago would never have been in the left seat are now there. They may be able to go into the simulator and perform to a necessary level to pass the ride(which in many cases they already know the profile), but put them out on the line and they don't know how to "think outside of the box"(pun intended).

You can't put everything into the book, and when you give them something they have never seen before or "not what I was expecting" things get interesting. Some acquiantances of mine working over in Asia tell me about how the F/O's can flying the sh@t out of an ILS but give them a visual and it is all over. Unfortunately, they will try to fly the ILS all the way to the ground. Everything is done from rote. Turn off the flight director and it is all over. Good flight training is expensive. Many of the instructors are young pilots trying to build flight time. I am sure they do a good job, but there is no substitute for experience.

[A37575]

The simulator does not give the physiological effects of 'g' etc, so recovery from unusual attitudes needs to be practiced in a suitable real aeroplane.

Yes there is that limitation -but not insurmountable by any means; but in my opinion it is not the seat of the pants feeling that is the problem but more the lack of IMC basic manual instrument flying competency whereby the pilot lacks the rate of scan skills to react correctly to what the flight instruments are telling him.

When you see pilots as I have many times in the simulator `pulling through` from an inverted attitude in the 737 simulator instead of rolling level to the nearest horizon using the sky pointer for reference, you realise that the pilot has either not been trained properly - which often means most of his simulator sessions have been regulatory box ticking exercises - or he simply hasn't got a clue for whatever reasons.

[RadioSaigon]

A fascinating thread, on many levels!

Originally Posted by Capot

The aircraft was ferried empty, skin wrinkles and all, for sale as scrap...

hmmmm... to my way of thinking, this incident as reported would alone make a nonsense of Qantas' claim to have never suffered a hull-loss.

The thread in its entirety brings to mind the age-old axiom of: make something idiot-proof, watch somebody come along and make a better idiot.

[before landing check list]

And how, in the current climate, are they to acquire, hone and maintain these essential skills?

Just asking....

The airlines should hire pilots with more actual flying experience not clones that go to and leave some aviation academy. A lot of these pilots learn how to use a checklist in a C172, some twin time then move to a right seat of a jet with so much automation almost any average person can handle the aircraft in a benign/normal/canned (Same approach to same place with same altitude and speed restrictions to same runway). However get these pilots out of their canned environment (hold them a bit higher/closer/faster due to traffic) and they have NO IDEA how to make the aircraft go where it needs to go to make a safe approach and landing. The same pilots will start almost all the time to look towards the autopilot to fly them out of a dangerous situation not because it is the right thing to do (sometimes it is) BUT THEY CANNOT DO IT WITH THE HANDS.

What to do? Quit hiring these guys with the cookie cutter education, put these new guys with a regional airline and have them fly a lot of approaches with the hands with a good dose of automation in weather that is a challenge and runways less then 2 miles long.
Give them once a year 2-4 hours in an acrobatic aircraft or a real glider. That will actually cost less in the long run when there are fewer accidents.

However the new normal appears to be leaning toward the use of automation as the primary focus and hoping nothing serious happens. The sad thing is the pilots with actual flying experience (hours watching the water go by from a flight level is not actual flying experience) are retiring and there will be NOBODY up front with flying skill soon. Unless the bean counters turn this around I fear this will happen.

[bubbers44]

Unfortunately most of these hands on pilots haven't been hired much in the last 30 years. Diminish that number as you get closer to now. We have lost that quality pilot unfortunately thanks to the pilot mills and regional airlines aiming for the cheapest help. We will pay for this as more accidents happen due to airlines choosing the lowest qualified pilots to keep costs down.

[before landing check list]

Or am I missing something?

Not at all.

[FullWings]

I think some of the problems have come from the gradual removal of little bits of training, both initial and recurrent, over the last few decades, plus ever-tightening constraints over what you can do to keep current on the job. This has arisen for budgetary reasons but also, strangely, from safety analyses.

There was a large delay or lag in the effects of these policies because the average experience levels (not just flying hours but hours dealing with low-tech airframes and automation as well as EFIS/FBW, etc.) were pretty high. Now we have captains who have used nothing but autothrust for their entire airline careers, excepting a few scary minutes in the sim and have little/no experience of pitch/power couple or basic performance attitudes. This has been compounded by ever-reducing technical training, to the point that answers to questions are learnt individually rather than deduced from a broad knowledge base or even from first principles.

I remember when a large fleet in my airline went to autothrust only, on the back of safety concerns (probably more legal arse-covering, really). At that precise moment in time, yes, you could probably have made a case that it was slightly safer and you always had the pilot(s) to fall back on who were fully up-to-speed on manual thrust control. Fast forward 10 years or so and almost nobody is current or even ever *has* been current, yet it is allowable under the MEL to dispatch with A/T U/S...

As far as the pilots being the backup to automation, then I would agree that that is a fair assessment of much of what we do today. The difference being that today's pilot is much less of a backup than one with a few decades experience, simply because they have not had (and maybe never will have) the exposure and/or training to allow them to take control in an effective manner.

Technical advances such as EGPWS, TCAS, RAAS, OWM, etc. have made the skies much safer in their own respective corners. Unfortunately, I think we will see, percentage-wise, more accidents with prime causes in mode confusion, loss of SA and inadequate basic handling skills, to mention but a few areas.

[Tinribs]

Control during tricky moments

While I would never claim my first aviation employer, God Bless Her,
always employed wonder thinkers they were sometimes right.

As I recall we had a slightly similar problem with the Lightning in the 70s. It cruised close to Mach 1, was the first proper supersonic and so some concessions had been made for performance, it had very swept wings, it could get itself into some tricky situations because of the role.

We kept loosing them through loss of control in stall spin type situations, usually in combat exercises.

The solution was to take the pilots once a year and let then play with a JP, stalling spinning aeros. The losses ceased

We could not do that with Boeing/Airbus of course but maybe a session in a light aircraft while we wait for simulators to catch up.

[Microburst2002]

agree or disagree with the following statement?

"He who does not truly master hand flying skills cannot truly master automated flying skills"

Do you guys agree or disagree with the above statement?

[2EggOmelette]

Agreed. One compliments the other.

[before landing check list]

Agree, however we are not bean counters.

[FullWings]

I think I'll have to disagree a little.

You could teach someone from ab-initio to fly near-perfect profiles in the sim or on PC flight-simming software, using only automatic functions. At that point they would have never touched a real aircraft or even a stick/yoke/rudder. OK, they might have problems taking off or doing anything other than an autoland, but the concept is there.

Hand flying skills and automation skills are neither complementary nor coincident; they are overlapping. You don't *have* to be good at one to be competent in the other but it sure does help...

[Desert Dawg]

If you allow me a small thread drift.. but on the same meaning as the current topic:

I watched an episode of Top Gear where Jeremy Clarkson played a computer game, racing a Subaru around a well know European track. His "sim" track times were impressive.. up with the top times from professional racers.

However, when he took a real Subaru around the actual track in Europe, his times were crappy and he openly admitted that due to the lack of real feedback from the car, he could never achieve the same lap times as he had in the computer game.

I agree that an aircraft simulator has far better feedback than sitting in an armchair with the car computer game, however there is no substitute for the real thing - real feedback.

I agree 100% that it is imperative that ALL airline pilots should get some mandatory hours in a glider/acrobatic aircraft to get real feel for stalls and spins etc. to learn what it feels like -in your gut - to stall and spin an aircraft. Once that sensation is learnt, I am sure the number of incidences will decline.

Just my two pence worth...

[A37575]

"He who does not truly master hand flying skills cannot truly master automated flying skills"

Do you guys agree or disagree with the above statement?
Microburst2002 is offline Report Post Reply

I thought carefully about that statement and have my reservations. My experience is that automation operating skills are an entirely different animal to pure flying ability. I have seen pilots in the simulator operate the automatics with great aplomb with the result usually a splendidly flown ILS by the automatic pilot. OK so they may spear off the centre-line on to the grass after touch down sideways on a wet runway but hey - you can't expect perfection after a career on automatics.

Now disconnect the automatic goodies of FD and AT and ask the same pilot to hand fly an ILS with a mere 15 knot crosswind and you may as well expect him to send out a precautionary PAN call, such is the awful handling skills (?) displayed.

The situation becomes positively scary now that Pay-to Fly schemes are in vogue with a vengeance. We are seeing a significant number of brand newly qualified (?) pilots with 200 hours plus and only the minimum legislative command time for their licence, rushing to USA or Asia for a 737 NG rating then paying huge money to the purveyors of Pay-to-Fly schemes who place them in an Asian airline as trainee first officers for an agreed number of hours with a hapless local captain who is forced to fly single-handed.

Almost all of their training on the NG/A320 is autopilot controlled and this continues during their short stint of maybe 500 hours RH seat. Then they may be out on their necks until the next newby arrives for his "training" on revenue flights. All the while the passengers haven't a clue what is really happening up front...

Yet by dint of plenty of money generally supplied by adoring (or desperate) relatives, eventually some of these newby's will make it in short time (3-5 years?) to the left seat -all the while "monitoring" the automatics and with no serious flying skills. The old adage from when glass cockpit flying first came in 40 plus years ago, of "I can't fly for nuts but I can type at 80 words a minute", is as true today as it was then.

[Cornish Jack]

Did NOBODY else watch Tubby Linton's link! That and the rest of that series should be compulsory and REPETITIVE watching for anyone in this line of business!

[Rabski]

"...have my reservations. My experience is that automation operating skills are an entirely different animal to pure flying ability. I have seen pilots in the simulator operate the automatics with great aplomb with the result usually a splendidly flown ILS by the automatic pilot. OK so they may spear off the centre-line on to the grass after touch down sideways on a wet runway but hey - you can't expect perfection after a career on automatics"


Not sure how to react to that comment. I'm 55 years of age and now exclusively short-haul, but I still spend some of my 'play time' in gliders and single engine VFR, and helping teenagers acquire a PPL. My day job involves the ultimate in automation () but I still need to keep my eye on the ball.

Most of the time, I trust the electronics. However, I have a food-processor that sometimes bites me and a digital alarm clock that takes the p155 on occasion. Basically, anthing electrical or mechanical has the ability to go wrong, just as does anything human.

Automation is wonderful, but it's still an assistant, rather than a master. Basic flying skills will always triumph in the (admittedly very) few occsasions when it all goes totally t*ts up.

Frankly, the training of pilots to become slaves to automation is worrying in my opinion, but hey, I may be wrong. Still, in the very few cases where all the holes have lined up, it seems that the times it's all come right were when good old basic skills were employed. Far too often quoted, but Sullenberger did the job right by using glider-pilot skills, not by trusting electronics.

Rely on machinery by all means, but have the basic skills to intervene when the hounds of hell escape from their cage, or you'll just end up as a statistic.

[Bergerie1]

Desert Dawg,

I think all civil pilots, during their basic training, need to fly some hours on an aerobatic aircraft. They don't need necessarily to learn aerobatics, but the do need (1) to know how to recognise and recover from stalls; (2) to know how to recover from spins; and (3) even more important, to know how to recover from unusual attitudes on instruments. This latter exercise needs to be trained on a fully aerobatic aircraft where the instructor can throw the aircraft around to thoroughly disorientate the pupil before handing over so that he has to recover on limited panel instruments.

Having done this, it is likely that a reasonable tolerance of the 'g' forces and the basic knowledge of how to cope will remain with him for the rest of his career. Whether or not further practise is required is an open question. But I would guess that an airline pilot who also flies light aircraft in his spare time and who practises such manoeuvres is more likely be able both to avoid, and to recover from, a loss of control incident.

[Dustertoo]

Recovering from unusal attitudes and stall recovery needs to be learned while in intial training. In two cases we have read how pilots reached back...707 captain remembered his Tiger Moth days and more recently...Sully and his glider skills. Yes we will always need to be the masters of the basics if we want to manage automation when the unexpect presents itself.

[captplaystation]

Not just learned . . . but probably practiced too.

Where, in any syllabus for recurrent training, is this mandated by ANY aviation regulator, where might one find that ?

As usual, Bean-Counters rule, and the Regulators, even if empirical knowledge/accident report conclusions dictate otherwise. . will dither.

Bucks rule, not Buck Rogers (unfortunately)




Refer to post #8, some , if we are sufficiently interested, we can do ourselves, but some of the rest, is expensive. Do our employers/regulators get involved ? Nah . . a certain number of accidents ARE acceptable . . . . . for the Beanies, sure, but for us, no way !