Pilot handling skills under threat, says Airbus

[JW411]

It seems to me that a lot of this manual flying versus automatics has to do with the philosophy being taught at the time.

When I did my DC-10 course, we were expected to hand-fly an ILS, with flight director to 100 feet and raw data to 200 feet otherwise you didn't have a job. The thinking at the time was that if you could hack an approach down to 100 feet then you would not have any problem flying down to 200 feet should the sh*t hit the fan in anger.

It was often said to me that "we know the automatics can fly the aeroplane but we want to know if YOU can fly the aeroplane".

Nowadays it is the case that very little hand-flying is actually required during the course of an LPC/OPC.

Just watch this space; it is all about to change again for such organisations as the UK CAA are becoming rather alarmed at the lack of manual skills being displayed by some of our brethern.

[Ladusvala]

Autopilots have limits

Tee Emm, do you mean that the autopilot can land the aircraft in gusty x-wind up to 35 kt? It canīt in the 737īs that I fly and thatīs one of the reasons I fly manual departures and approaches, practically all the time.

[rooaaiast]

Appropriate automation

Having flown "iron dial" and current fly-by-wire aircraft, imho pilots should be free to use "appropriate automation". Bad vis, long day, etc are good reasons to use the automatics. Otherwise it should be manual/manual below 3000'.
How else can you know your aircraft, and fully ready for the critical max x-wind landing?

[p51guy]

Flying with full automation all the time if required by your airline but having to handfly a landing if xwinds and gusts are above autoland maxes doesn't sound safe. A good feel for your aircraft is a must in those conditions and the only way you can get a good feel is to frequently hand fly the approach and landing. Disconnecting AP at 50 ft gives you little time to feel the aircraft.

[By George]

I agree with P51 the Aircraft I fly has a A/P limit of 25kts and a manual limit of 35kts dry. Thus the need to hand fly to stay in touch. Automatics are for when the weather or traffic conditions are bad. On a nice quiet sunny day its being lazy not having a manual play.

[aguadalte]

When was the last deep-landing incident from an autoland? for example; just wondering.

RC

Turkish Airlines in Amsterdam, comes to mind...

[Tee Emm]

Just watch this space; it is all about to change again for such organisations as the UK CAA are becoming rather alarmed at the lack of manual skills being displayed by some of our brethern.

Seems that way. Flight International 6-12 October has an article by David Learmont which states that Thomson Airways (previously Thomsonfly) has carried out eye-tracking tests on crews. The tests have discovered that a few pilots' instrument scans are seriously deficient, even when their performance would have been judged, by an examiner on the flight deck, to have been good. The article goes on to say the implication is that some airline crews, possibly at all airlines, are surviving because nothing goes wrong on their watch.

That is nothing new. No doubt most airlines have their fair share of pilots who scrape through proficiency and instrument rating tests simply because most of the tests are on automatic pilot. With probably at least 95 percent of all jet airline ops flown with full use of automatics, and very few significant aircraft technical defects requiring superior flying skill, the dodgy pilots get away with it.

The Thomson Airways "eye-tracking" only proves something that most simulator instructors have known for years. And that is there are a few seriously incompetent captains and first officers flying the airways quietly protected by first class, reliable, and almost fool-proof automatic pilots.

[DA50driver]

Flight Directors

An examiner I have a lot of respect for made sure we called it a Flight Suggester. Consider what it is telling you, but verify that it is what you want to do.

Since I don't follow the FD blindly does this make me a "Cowboy"? I don't think so, but I have seen enough of the new entrants in the business that would fly the aircraft into the ground if the FD told them to. They scare the crap out of me.

[CI 9999]

Interesting thread.

Some great posts here. There are less and less aviators in the business. We have taken in a lot of people onto the flightdeck that actually don't enjoy flying, do not take care to maintain their skills and may actually find it stressful to fly.

I truely enjoy flying, and as a longhaul 744 skipper I take every opportunity I get to practise my handling skills and scan. Flying manually most departures, arrivals and approaches either with or without F/D. The jumbo is a absolute pleasure to handle and a great joy to fly.

Enjoy your flying, pitch and power ain't king, it is religon, please practise it.

CI 4x9

[RAT 5]

Couldn't agree more, but there is a blockage. We've read many comments on here where pilots disagree and say that if you're slow add power and vice versa. To me that is following the a/c and not leading it. You can tell by the power setting that you are going to become slow or fast BEFORE it happens. And that includes the PM, who should be aware of this as much as PF. Or, if there is a unexpected power setting you can consider why; tailwind, headwind, thermals etc. You are controlling the a/c. There may be a dog in the cockpit, but I like to wag the tail.
From that CAA Review a rather damning observationver 10 years or more
"75% of a/c fatal accidents have had a human causal factor." [LIST][*]omission of action/inappropriate action 38%[*]flight handling 29%[*]lack of situational awareness 27%[*]poor professional judgement/airmanship

Relating to another thread about pilot FTL's and tiredness. The same CAA medical department has said that deprived of regular quality sleep a pilot can be worse off than a couple of drinks. The causal factors above will all be effected, markedly, by lack of sharpness and calm relaxed attitude. Yet this same CAA, and others, allows the degredation of FTL's. Getting out of bed at 04.00 5 mornings consecutively and then working 13 hours is not a recipe to reduce the incident/accident rate; including on the roads afterwards. Remember, these statistics are only the fatal crashes. What about the near incidents that never go reported. The list would be longer than Warren Beaty's girl friends.

CTC make the comment, very correctly, that training just to the minimum standard is not good enough. I've flown for many airlines who reduced their training to this level and increased FTL's to maximum. In all cases their excuse was "it's legal". I spoke to the JAA FCL and their argument was they established a minimum level, but expected the airlines to adopt a level necessary for their operation and in-house philospohy. That also included having a sensible buffer to FTL's and schedules. We all know what happened: cheapest legal requirement. It was a hopeless thought. 30 years ago my C.P said "I want my pilots to be excellent handlers on the line". Indeed, amongst the Greek islands on dark stormy nights, it was abolsutely necessary. The captains were generally excellent and demonstrated/ encouraged F/O's to follow suit. Command was after 5000hrs. Now 3000hrs seems to be acceptable and some airlines positively discourage visual approaches because there have been so many screw ups that time & money has been wasted. That is a very strange solution to the problem. Instead of improving training they encourage staying away from the problem. In some cases that is a valid response, but not in something as basic as a visual approach. There will be captains who are not proficient at this most basic of manoeuvre.
When I've seen pilots program a visual approach in the FMC and then fly F.D & LNAV, allbeit manually, I cringe at their thinking.
LPC's should be about a/c handling proficiencies. OPC's & LIne Checks can be about procedures and SOP's. Thus 2 different ypes of check. No where in the checking syllabus is there real handling checking. The 3 year un-usual attitude recovery recovery is pathetic. A couple of simulator induced upsets, training only, then move on with the rest of the program to more automation. A V1 cut & SE G/A should be child's play. That's all there is. Circling is not an LPC item, and even that is flown on A/P.
In new a/c there is a massive amount of information to make handling the a/c more accurate than needles & dials, but sadly the skills have been diluted. This has to be laid at the door of training departments. the LPC's haven't changed, so the CAA's have not changed their hoops to jump through. It has to have been driven by training departments, answerable to the financial dept's. It will have to be reversed by the same. I don't believe the CAA's can devise a program suitable for all operations and operators, but 2 different types of check might be a start. However, whatever change is introduced it has be long term and not a transient modification.
I see that some airlines have reduced the 2 day sim training/checking to 1 day combined. Legally!! That means even less time in the sim to practice 'what if' scenarios; including simple non-normal handling. During training sessions, if there had been a recent survivable crash, but it failed due to wrong analysis, mis-management, mis-handling, I tried to include the elements into the session. If one crew had got it wrong, why should we not learn from how they did that and learn what would have been better. It may not happen again, but finding out what would have been more successful will always have spin offs into other scenarios.
Is it really true that new fully integrated airline pilots' course are now conducted with much less handling skills and much more in simulators? I can understand the cost reduction incentive, but has this been the start point of training a systems opertaor and not a pilot. Has this sowed the seed in the cadet's mind that big jets can't be handled like spam cans? There is an argument to train for what you are going to do. A Super-tanker skipper doesn't need to be able to sail a dinghy, but he still needs a healthy respect of the elements and be able to manage when the sutomatics have gone AWOL. Has the 'airline pilots' training course been diluted too much solely to reduce cost. Why else? The focus of MPA/MCC could be introduced as a separate module at a later stage once the basic solid foundations of piloting have been established. CTC had an airline indoctrination course as a stand alone module before type rating. It's similar to medical students; they have a strong basic foundation of medical training and then some years to specialise. Why not us the same? Perhaps the career path has caused some of this; 150hrs straight into a jet. The ladder, via G/A, commuters, military, taught some of these basic skills and they could be carried forward to airlines.
I do think we've progressed backwards with pilot training.
Sadly, like the FTL's question I expect this one to revolve at ever increasing speed and disappear down some black hole until there is a smoking one. Only then will the bean counters stare the problem in the face, but it won't be an overnight fix. It has to be a culture within the airline, and that comes from the top.

[Tee Emm]

Thus 2 different ypes of check. No where in the checking syllabus is there real handling checking

First of all, unless there is a series of horrendous crashes involving automation complacency then nothing is going to change. The occasional crash like Turkish Airlines at Amsterdam will stir the ants nest a little but already that is old news; so blind reliance on automation driven by manufacturers and airline ops departments is here to stay.

One solution is for the regulatory authority to legislate the proficiency and instrument rating test/renewals must consist of equal simulator time spent between automatics operation and raw data manual flying without the aid of MAP and FMC. Normally the combined proficiency/instrument rating renewal covers two hours for the PF. Of this at present, roughly 80 percent is full or partial automatics. The pure flying is thus relegated to maybe one hand flown approach but FD allowed. Quite useless in terms of pure flying currency.

Instead the two hour session should be shared equally between each discipline with (for example), the first hour allocated to automatics competency. In basic terms, the candidate's button pushing skills. The second hour should be similar flight path and navigation manoeuvres modified where necessary, but hand flown raw data no automatics, no FMC , no MAP. This tests the candidates pure flying skills.

Since "practice" hand flying is frowned upon in a growing number of airlines because of a perceived danger to flight safety and passenger comfort, then it is logical to ensure pure flying skills are maintained in the only thing left - the simulator. No coffee spilt, no passenger complaints and no extra cost. A compromise betwen commercial issues and flight safety.

[JW411]

I retired from commercial flying some three years ago. Not long before I hung my boots up, I was discussing with our Training Manager the lack of manual flying required in order to complete an LPC as per JARs compared with the old days.

He made the statement that it was perfectly possible under JARs for a captain to pass an LPC without ever touching the controls!

Since LPCs are nowadays supposed to be conducted in a quasi-LOFT scenario, then it is perfectly reasonable for a captain, just for example, to get the F/O to fly the 3-engine ILS and G/A since that is what the company SOPs demand (RVR less than 1500 metres = a monitored approach).

However, when I asked my Training Manager if he would actually sign a captain's licence who had never touched the controls he admitted that he would not. As a TRI/TRE, I sure as hell would not!

So it is that we have to "cheat" by failing autopilots and F/O's instruments etc just to ensure that we can see whether a cptain can still fly or not.

Under the Old Queen, we had a four hour LOFT slot (which was invaluable). Then we had a Base Check/IR which was an intensive handling exercise full of unrelated failures and was therefore, a pretty unreal exercise but was great for assessing ability to deal with problems.

Now we have the same four hour LOFT exercise (the contents of which are published three years in advance) and then we have an LPC which is supposed to be LOFT-related, but because of the problems as related above, ends up by being pretty well about as realistic as its predecessor.

I cannot wait to see what EASA comes up with.

[Vode]

passing it on

When I started in the latter part of 80's, we still had a lot of guys who had flown with the real pioneers - both with military aces and line flying pioneers who had started in open cockpits when accidents could be fatal even if no-one had made a mistake.
There was a tradition with the pilots to try to make their flying as efficient and quick as possible. This meant idle descent and short visual whenever possible. Operating in a country with very little traffic at the regional airports, this was easy to accomplish. They were gently pushing the envelope with skills learned from their preceding generation. They passed on this skill to us.

Sometimes we got to show off at big airports, as we had a reputation among the ATC that we will be flexible. There is a true story of AMS ATC asking our plane if they can make 3 mile final. The COP (now Fleet Chief B757) responded: 'MY captain can make any final' and the ATC replied: 'OK, cleared for 2 mile final'. Which they did, with margins to spare.
I got to do a low 360 over the field in MAN, when ATC had cleared a departing BA in front of us. As we landed she commented: 'that was impressive, you don't see that every day!'

And I think this is the core of the problem. We don't get to do this and further more we don't get to pass it on to the next generation. If we always fly ILS and make a 7 mile final, they'll never get to learn how agressive flying can be made safely. I don't mean reckless but close to the limits of the envelope. The young are limited by SOP and aircraft design to flying in the middle of the road.
After serious malfunctions and pulling some breakers on a A320 simulator, we managed to get the plane behave like a Diesel 9 on a normal day. That seems to upset the youth very much - not because they couldn't handle it but because such a big deal was made about the 'loss of protections'.

I know it is a thin line between reckless and safe when you go away from the inherent safety of the SOPs, but there is something to be said about seeing and doing real flying. Our airline has a reputation for safety and I think it was partly based on the airmanship of our pilots, not on superior SOP or strickt following of the FOM.

[tocamak]

I know it is a thin line between reckless and safe when you go away from the inherent safety of the SOPs, but there is something to be said about seeing and doing real flying

I'm not sure if you meant it that way but testing out the thinness of the line with pax onboard doesn't sound like a good idea.

This thread looks in danger of degenerating into a fond look back at the "good old days" of flying when SOP's were down to the individual and CRM was the Captain telling the "boy" what to do. Exageration for effect here before you all jump down my throat!

Times move on and you reasonably would not expect to be doing a "low 360 over MAN" although you might do an orbit on final at a less busy place and quite rightly you should be able to do it without trying to remember how to actually fly. The accident records surely show that it is the decision making process which is at the heart of most accidents rather than raw flying skill.

[Microburst2002]

Hand flying skills and decision making ability are both mental skills.
We should have mental resources to do both simultaneously, shouldn't we?

Imagine that your hand flying skill is degraded due to lack of training and practice.
If the time comes that you have to hand fly and make a decision, your brain resources for decision making will be greatly reduced because you have to fly the damned airplane.
Or, God forbid it! you will be making a decision proccess with the D.E.C.I.D.E. or any other nice achronym while the airplane flies you. And doing all checklists and call outs until the CVR stops recording.

A skilled hand flyer has a better situational awareness than a non skilled hand flyer, all other characteristics being equal. So he is safer.

A TRE I know says that he often sees crews crashing with only an Auto Thrust inop. Sometimes with only one channel inop!
I have seen people in sims cancelling the master caution every five seconds and wondering in anger why the f*ck it keeps being triggered... (only 320 pilots can understand, maybe?). I have even seen some to reluctantly hold the thrust levers as if God himself had abandoned them.
If this is happening, then something is wrong in the system.

[RAT 5]

Interesting comments, and memories of AMS ATC. I flew in/out of there for many years. Sometimes it was like LHR; a convoy with ATC flying the a/c, sometimes it was, "we've a heavy at 12nm. Will you accept 3nm to be No.1? Wx 3km's & 500'." Hell, we had a triple autopilot, so why not. Any one of them would do. "turn right, base leg , descend 1200'." I know many airlines who's SOP's would say NO. Why? Equally, on a clear day/night a visual circuit power off was always approved. One of the smallest big airports I knew. (I know it has changed now.) That's something that's gone AWOL, a descending low drag visual circuit. I offered it to an F/O and his comment was, "how do I know how long to time pass the threshold?" First it was visual, 2nd there was a DME on the ILS. What more do you need. Back to lack of traning and line practice. There was no 'profile' for it.
The other comment about 'the envelope'. I flew for an airline that said, 'max UP speed at 15nm', 'gear down at G/S alive.' Going into USA B767, ATC "give me 250kts on the glide at 25nm". They were 'pushing tin' and needed speed control to establish separation between variety of types. 210kts at 12nm and he lets us slow down at 10nm. The F/O's eyes were on stalks, but it worked. The gear is a wonderful piece of drag when needed. The F/O had never put the gear down until mid-range flaps, and certainly never before the flaps. He didn't think it was allowed. That's what the SOP said, so that's the only way to do it; as you quietly make a glide landing.
I still say any captain should be able, when given the a/c at 4000agl, 5nm from the airfield, 90 degrees to the Rwy, clean speed, to make a visual circuit low drag onto either Rwy with no G/P indicators, daytme. It's possible to use level D sims for this and would be a simple element of a CAA annual handling check, but at the very least a command check. Sure we should all be capable of the most basic of aviation's manoeuvres, the visual circuit. If so, should it not be tested? You learn a great deal about a pilot's abilities watching someone fly a visual descending circuit.

[CapitainKirk]

RAT 5

"I still say any captain should be able, when given the a/c at 4000agl, 5nm from the airfield, 90 degrees to the Rwy, clean speed, to make a visual circuit low drag onto either Rwy with no G/P indicators, daytme. It's possible to use level D sims for this and would be a simple element of a CAA annual handling check, but at the very least a command check. Sure we should all be capable of the most basic of aviation's manoeuvres, the visual circuit. If so, should it not be tested? You learn a great deal about a pilot's abilities watching someone fly a visual descending circuit."

I could not agree more but change your should to could and your sadly in a whole new ball game.

Sadly the modern generation are not pilots but system managers - there was a thread on here a while ago - probably still here hiding somewhere - where a simmer who used the 767PIC program asked if anyone thought he could land a real 767. From reading what he wrote (some pretty technical on the ball stuff) i was fairly convinced that he 'probably' could - using all the systems - which actually was his question - however i am 100% sure that he could not hand fly that plane in anything other than SLF.

I wanted to fly.............now it seems that people want to be airline pilots for many other reasons that have little to do with - "flying is what i love doing". One of my friends wanted to be a pilot but for many reasons could not be - so he lived his life through his very clever son who at a speed faster than Concorde (how i hate BA/AF/BAE for what was the biggest piece of aeronautical vandalism since the TSR2) is now a 777 Captain. i really wonder how many hours he has when he was actually hand flying the plane on his way to becoming a 777 Captain. How many miles out were they on the 777 (BA038) that scraped over the fence at LHR when they dumped the automatics - and when does your typical C/FO take over and complete the landing in normal everyday landings?

We are moving into a gereration where kids can be super heroes on things like Gran Turismo on the XBox PS2 Playstation etc but what are we finding is that put them in a real car on real roads and they are killing not only themselves and others including their mates as well. But are we addressing that in any sensible way - by real life training - nope. I know the analogy is not directly compatible to flying BUT it feels like it is getting a bit too close to that (a computer game) for my liking. Sadly i dont see how that this is fixable - in this - lets fix it with a sticking plaster - modern world we live in. With the frequency and level of serious aircraft accidents we are having nowadays i am sure statisticians airlines and airline manufacturers will be able to prove we are 1000% safer with the NG of planes and pilots - (non existent) problem solved.

[AnthonyGA]

Sadly the modern generation are not pilots but system managers - there was a thread on here a while ago - probably still here hiding somewhere - where a simmer who used the 767PIC program asked if anyone thought he could land a real 767. From reading what he wrote (some pretty technical on the ball stuff) i was fairly convinced that he 'probably' could - using all the systems - which actually was his question - however i am 100% sure that he could not hand fly that plane in anything other than SLF.

I've been reading this thread, and I keep thinking about a strange parallel between the topic under discussion and the world of desktop flight simulation.

In flight simulation, the objective is to be as realistic as possible in virtually every aspect of flying, and that objective is approached much more closely than I suspect most people here realize. (Unless you are a serious simmer, you probably think that flight simulation at home stops with Flight Simulator or some overpriced, special-purpose desktop sim.) There are, however, some curious differences between simmers and real pilots (even when the simmers also happen to be pilots in real life, which is common enough).

One interesting difference is that simmers usually start their sim aviation careers in airliners, and then "work their way down" to small aircraft. Why? Well, flying an airliner is generally perceived as more glamorous, of course. But another very important reason for this is that airliners are easier to fly, because almost everything is automated.

The newbie simmer chooses a 747-400 as his first airplane in part because he need never touch the flight controls (almost) in order to fly it. All he has to do is pull the yoke back a bit at take-off, and then push a few buttons to let the FMS do the rest. And he can autoland (in the sim universe, autolanding is safe at any airport with a functional ILS), so he need not touch the controls at all after take-off.

Now, the hard reality is that anyone can push buttons and turn knobs, and anyone can learn how to do this from a book. No real-world practice is required, because these actions don't require any special motor skills, coordination, or motor memory. Most people have pushed buttons and turned knobs all their lives, and once they learn which ones to push and turn (which they can do by self-study), they can operate anything that requires only buttons and knobs … and that includes flying a modern airliner.

So the newbie pilot steps into his simulated 747 or A380, loads a flight plan into the FMS, and he's ready to go. He can press buttons and turn knobs and fantasize that he's an airline pilot.

As the serious simmer "graduates" from one level of skill to another, he moves down rather than up. Smaller aircraft don't have a FMS, so he has to learn how to use the autopilot. And he has to learn how to navigate, since an autopilot won't follow his flight profile for him. Lazy simmers rely on a GPS for navigation, though, and thus delay their progression.

As the simmer moves to still smaller aircraft, he loses the GPS, and has to start looking at charts. Then he loses the autopilot, and things get really tough, as now he must finally grab the flight controls and fly the aircraft by hand. Just as in real life, flying by hand takes a lot of practice, and so it may well be the last thing that a simmer learns—the crowning achievement in his sim career.

The net result is that the 747 captains on virtual flight networks are mostly 14-year-old newbies, and the Cessna 152 pilots are mostly 40-year-old veterans of simulation.

All of this reflects the fact that large commercial transport aircraft are increasingly flown by computers, with pilots only as attendants. Not only is it possible to avoid flying by hand, but it's expected, and it's by design. The ultimate goal is to eliminate the pilot, but in the meantime, he can be reduced to an observer on normal flights. It makes good sense for airlines, but it does not bode well for pilots.

The reality is that the only thing that requires practice in a real aircraft (or a full-motion sim) is hand-flying. Like riding a bicycle, it's something that you have to do in order to learn, because of the motor skills and undefinable "feel" that you need to succeed. But all the other aspects of flying do not require this, and they can be learned by rote without going near a real aircraft, and they can be carried out by someone with very little training.

I don't know of any way to reverse the current trend. Airliners are designed for airlines, not pilots. As long as all the systems on board work (and they get more reliable all the time), hand-flying is not required. Right now hand-flying skills are subject to atrophy; eventually, they won't even be required for pilots to begin with, once the systems are considered sufficiently reliable. In the meantime, it's important to have those skills if something goes wrong, but since things hardly ever go wrong, airline accountants don't understand the utility of paying for pilots to maintain those skills beyond whatever air regulations legally require. And pilot unions are often afraid of anything that might challenge a pilot's ability and rock the boat of the status quo.

Which reminds me: Is it harder to hand-fly a 747 than it is to hand-fly a Cessna 182? Or is it just different? Although I guess these days the question might be: Does anyone remember hand-flying a 747?

[CI 9999]

To answear your question, many of us do handfly the 747 as much as practical, I'm certainly one of those. The aircraft is very nice to handle, and as long as you stay ahead of her there is no danger there.

I also fly SEP IFR without autopilot, and can confirm that it is much easier to handfly the B747 IFR, but in either case the safe outcome is down to maintaining a solid scan and staying ahead of the aircraft at all times.

In terms of PC simming and simulators, these are powerful tools to practise instrument scan, and the latter to practise normal and non-normal procedures and CRM, but these are "computer games" and will never be exactly like the real thing, just for that matter.

Automation on Airliners is a very good thing, one must understand how it works and stay on the ball and stay ahead of the aircraft regardless. Autolands are great, but have their limits, and guess who can operate to higher wind limits, autopilot or a human pilot?

Enjoy your flying

CI 4x9

[have another coffee]

The post of VODE is very recommendable

I may have had the sheer luck of having a flightinstructor during initial (military) flight training who made me go to the limit every single time. The guy showed me every single sortie where the (safe) limit could be. It teached me the limit of operating an aircraft would be different every single day, be it weather, fatigue, aircraft limits, weakest crew member or hand flying skills etc...

The message where to draw the line is still there. In order to operate an aircraft "in the middle of the road" it is still important to know where the edge of the road is. Disconnnecting all sorts of automatics and doing a manual job is an important aspect in discovering how wide the road actually is and where on the road we are operating. As a personal observation, the road definately became smaller in the last 10 years concerning manual skills.