NTSB update on Asiana 214

[Volume]

2) the pilot flying was inadequately and/or incompletely trained in manual operation the airplane with or without assistance provided by whatever automatic systems

I think the issue is about the "with or without assistance provided" part of this sentence. The pilot thought he would fly the aircraft "with assistance provided" while in fact he was flying it "without assistance provided" (because he unintentionally and unnoticed asked the "assistant" to quit).
All the other issues (adequacy of the instrumentation etc.) are just secondary, the main issue is that the pilot thought he would do something different from what he was actually doing. It was about understanding how to work together with systems. Obviously he was not trained to question what the systems are doing. Obviously he was not trained to monitor the systems. He was trained to trust in the systems.
It is hard to blame any single aspect, but it is easy to identify where everybody can improve (airplane designers, training organisations, pilots, airlines, authorities....), and that is just what we should do now.

[Gretchenfrage]

Reports of pilots about any kind of inadequacies are often brought up, mostly by citing them as absent, so apparently everything is ok.

We have to get away from that concept.

Every report about any inadequacy points, voluntarily or not, to somebody to blame. In today's environment this is highly critical and not welcome.
Pilots are often discouraged to maintain such reports, for many reasons, but mainly because they would cost money or incite lawsuits. Friends working at the big one in the ME even experience open harassment if filing safety reports!

In the case of the airspeed tapes, there is basically no big surprise. Everyone know that a round display is easier to discern in stress moment than any tape. Pilots can get used to tapes, but will never perform as good as with round instruments.

However: The underlying problem with this accident is the switch between moving throttles and fixed ones. The same findings apply: Pilots can get used to fixed ones, but perform better with moving ones. The real problem is when they were trained on one and then switch to the other design. It takes a long time to get the new one under your skin, especially when moving from fixed to moving.

That is what should be addressed, because a change of any display or system will be "discouraged" by the lawyers and bean-counters, often leaving the pilots with the lesser adequate solution.
Either give pilots more training when changing, or leave them on the system they have been trained initially.
But again, this might involve cost and will therefore be decried.

[safetypee]

AirRabbit, assuming that I have interpreted your commentary correctly (#1101), then there may not be any significant difference in our views. However, there are some interesting points within the differences and in the viewpoint.

The training views appear to follow a linear cause-effect-fix path, identifying and improving an aspect which has an acceptable, but not guaranteed probability of success (the human).
The alternative is an attempt to understand a dynamic interaction of factors, like a flywheel which cannot be totally fixed (stopped) because then nothing would be achieved, thus the task is to apply a brake to maintain an acceptable speed range (influence human behaviour with small changes in the overall operating system – man, machine, and environment).

Rather than go into detail, there may be better explanation in the documents:
‘How to be safe.’
‘Myths in safety’.
‘A Tale of Two Safeties.’

[DozyWannabe]

Originally Posted by FGD135

how poorly the airspeed is communicated via the tape.

Just a minute here...

Originally Posted by DozyWannabe

What was interesting about the original 1949 Grether study was that the experiments used a control group of non-pilots - so if we look at the summary:



You can see that round dial type "D" and tape display type "G" are roughly comparable in terms of percentage error and interpretation time (in fact the tape display seems superior to the dial in terms of percentage error). The interpretation time is interesting to compare because the non-pilots interpreted both in the same amount of time, but the AAF pilots interpreted the dial marginally faster. From this one could argue that the pilots' prior experience with the dial might have skewed the results.

I'll say that again.

Round Dial "D":
Percent error - 0.7 (pilots), 0.7 (non-pilots)
Interpretation Time - 1.7s (pilots), 1.9s (non-pilots)

Tape Display "G":
Percent Error - 0.3 (pilots), 0.4 (non-pilots) [Note both figures are superior to round dial "D"]
Interpretation Time - 2.3s (pilots), 1.9s (non-pilots) [Pilots take slightly longer to interpret this one by 0.6s, but non-pilots interpret as quickly as they did the dial].

How is the tape display poor or inferior?

With the overspeed, alpha and stall zones clearly marked alongside the tape as well as the speed bugs, this satisfies the "proximity" perception as well, plus with the PFD you can use colour changes to alert pilots to developing abnormalities in a way you cannot with analogue gauges.

I'm well aware of the psychological studies you're talking about, but as far as I can see all of the potential issues have been addressed. You may prefer the old gauges, but to say they're objectively "better" is incorrect.

Originally Posted by Gretchenfrage

However: The underlying problem with this accident is the switch between moving throttles and fixed ones. The same findings apply: Pilots can get used to fixed ones, but perform better with moving ones. The real problem is when they were trained on one and then switch to the other design. It takes a long time to get the new one under your skin, especially when moving from fixed to moving.

Turkish 1951 says otherwise (i.e. three pilots who'd flown with nothing but moving thrust levers missed them rolling back too far). As with FGD - it's absolutely fine to have your own personal preference, but it's intellectually dishonest to state that your preference is fundamentally better with no significant proof. "Everyone knows..." just doesn't cut it.

[barit1]

There are a couple common features between Turkish 1951 and Asiana 214:

o Boeing moving throttles, with inadvertently disabled autothrottle system
o Approached the g/s from above
o Three people in the cockpit - one of them a 'safety' pilot
o Despite six eyeballs, very late recognition of airspeed/altitude awareness

This seems to be a witches' brew of factors, none of which individually has been a problem over the decades. Rather than isolating them, let's look at the synergy they create when together.

[DozyWannabe]

Originally Posted by barit1

There are a couple common features between Turkish 1951 and Asiana 214:

Indeed, but I don't think the A/THR was "disabled" in terms of being off or not functioning. The Asiana A/THR was inadvertently set to "HOLD" by the crew, and the Turkish A/THR went into "RETARD" (i.e. rolled back too much and too early) because of a faulty RA input. The crew moved the levers forward the first time, but did not disengage A/THR, so it simply pulled the levers back again.

[barit1]

You're right of course, but the net effect was the a/t didn't provide the protection the crew had assumed.

In Asiana's case it was software behaving "correctly" (and contrary to the crews' expectations); in the other case it was hardware in a failure-intolerant system.

[Machinbird]

Round Dial "D":
Percent error - 0.7 (pilots), 0.7 (non-pilots)
Interpretation Time - 1.7s (pilots), 1.9s (non-pilots)

There is something misleading with this data.

The recognition time might be true for an initial scan where you note the significant digits that you are interested in, but a rescan of the same instrument, once you have found the landmarks, takes only a fraction of a second.

I operate in re-scan mode when I'm flying instruments. At least most of the time.

[FGD135]

How is the tape display poor or inferior?

Dozy, Dozy, Dozy,

That Grether study is almost completely useless to this discussion - because the way the subjects were assessing the dials/tapes is completely different to how a pilot uses his airspeed display.

Given that Grether was measuring how accurately the person could read the instrument, and how quickly, I imagine that the method was to simply flash an instrument in front of a subject, and then time how long it took him to discern the value - then record how accurate his readings were.

So what the subjects were doing was very, very, very different to what a pilot's brain is doing when using instruments to fly an aircraft.

And to really underline this difference, look at the scores for instrument 'I', which is just a purely digital reading. According to the scores, this form of presentation is the best of all!

But if you seriously tried to give pilots just a purely digital value (instrument 'I') for airspeed you would be laughed out of town!

That instrument 'I' scored so well shows that this test is not at all relevant to pilots and aircraft instruments.

To properly test tapes vs dials - as used by pilots - you would have to use a flight simulator, with the pilot trying to accurately fly the virtual aircraft. The pilot would also have to be in a high workload, high stress situation.

Do this with two simulators - one with tape for airspeed, the other with dials, then look at how accurately the aircraft was flown.

From that study you linked to earlier, it sounds like testing along these lines has already been done. Here are the relevant quotes:

Testing in a Link simulator found the tape display to be workable, but pointers resulted in a superior flight performance. Further experimentation with expanded scales and more training was recommended (Mengelkoch & Houston, 1958).

and:

A reminder that tape displays are also not optimum when a pointer can cover the required range was seen in testing of several formats for an F-16 vertical velocity indicator (Cone & Hassoun, 1991).

and:

An indication of the limitations of tape displays in dynamic flight environments is seen in the midseventies when the USAF moved away from tape displays for heads down primary flight displays

It seems that you, and a few others on this thread don't like these quotes, but I believe we should be attaching considerable weight to them, as they relate specifically to how pilots use instruments when flying.

... this satisfies the "proximity" perception as well, ...

Dozy, I think you misunderstand what "proximity compatibility" is. The "proximity" is not so much about physical separation, but more about "similarity". See the following:

"The Proximity Compatibility Principle: Its Psychological Foundation and Relevance to Display Design" by Wickens, Christopher D.; Carswell, C. Melody - Human Factors, Vol. 37, Issue 3, September 1995 | Questia, Your Online Research Library

Human?computer interaction - Wikipedia, the free encyclopedia

You may prefer the old gauges, but to say they're objectively "better" is incorrect.

This is a stunning statement, in light of what that list of quotes had to say. I think you need to have another read, slowly and properly, of all those quotes. That post here:


http://www.pprune.org/8607329-post1043.html

... it's intellectually dishonest to state that your preference is fundamentally better with no significant proof.

Dozy, Dozy, Dozy. That list of quotes contains some very significant proof. Making it more significant is the absence of anything saying that tapes are superior. You and some others seem to have blinkers on when it comes to the message carried by those quotes.

[Oakape]

You're right of course, but the net effect was the a/t didn't provide the protection the crew had assumed.

Protection is there to try to avoid an accident if the crew make a mistake or miss something. It is not there to become the normal way of operating the aircraft.

If both of these crews had been 'flying' the aircraft, as their duties require them to, both accidents would not have happened. Crews are required to monitor airspeed at all times & to ensure the power is set correctly to maintain the required airspeed. If this basic requirement of flying had been done, neither aircraft would have been involved in an accident.

Not so many years ago, the concern among aviation professionals was that that the manufacturers were designing too much automation into their aircraft & that that was a threat to both safety & pilot careers. It seems a little ironic that these days a lot of pilots are concerned that there is not enough automation on the flight deck & that it does not fly the aircraft 100% accurately, all by itself. Perhaps pilotless aircraft are closer than I used to think & that part of the reason for that is that some pilots seem to want to design themselves out of the flight deck.

[Oakape]

Here are the relevant quotes:


Quote:
Testing in a Link simulator found the tape display to be workable, but pointers resulted in a superior flight performance. Further experimentation with expanded scales and more training was recommended (Mengelkoch & Houston, 1958).
and:

Quote:
A reminder that tape displays are also not optimum when a pointer can cover the required range was seen in testing of several formats for an F-16 vertical velocity indicator (Cone & Hassoun, 1991).
It seems that you, and a few others on this thread don't like these quotes, but I believe we should be attaching considerable weight to them, as they relate specifically to how pilots use instruments when flying.

Perhaps pointers resulted in superior flight performance in 1958, but was that more to do with the fact that tape displays were relatively unknown & therefore the test subjects were significantly more familiar with the round dial & pointer presentation?

Also, with regard to the second quote, vertical speed is somewhat different to airspeed & the best indication for each is not necessarily the same.

[olasek]

This is a stunning statement, in light of what that list of quotes had to say.

No, it is not a stunning statement, perhaps you and others "fell" for the quotes but they hardly represent objective truth, they are fragments of someone's opinion, they aren't even a full opinion. To call them "proof" for anything is a folly. You would have to dig a lot deeper to provide a 'smoking' gun because so far there ain't one. By the way, the discussion is getting highly repetitive, some clearly have a lot of time typing voluminous arguments, thousands of words, quotes, fancy editing, etc. that actually show or prove nothing. This is just another forum with a lot amateurs, some pilots who flew something and by the way with no single 777 pilot as I can tell.

[Capn Bloggs]

Originally Posted by Olasek

This is just another forum with a lot amateurs, some pilots who flew something

You can say that again. Your simplistic "it was the idiotic, incompetent pilots fault" view shows just how much you don't understand the situation with Asiana 214.

[olasek]

Could be simplistic statement on my part but is is actually simplistic to focus entirely on the speed tape. There is zero evidence they even glanced in the direction of the speed tape or had capacity to process whatever data was there in whatever format. Since they were incapable of processing high angle of attack visual cues outside the window or 80 lbs force on the yoke it is even more unlikely they had capacity to process any instrument readout, analog or digital. But if you 'understand' Asiana 214 better than anybody else perhaps you can explain what happened then...

[FGD135]

... is actually simplistic to focus entirely on the speed tape

olasek,

I have not focused entirely on the speed tape. If you had been reading my posts properly you would know that I said that "an habitual reliance on automation" was also a factor in this crash.

Read the posts properly.

... they are fragments of someone's opinion, they aren't even a full opinion.

Perhaps you are not familiar with research papers and how they sometimes make references and summaries to other, similar research.

Those quotes were all one or two line SUMMARIES of earlier research/studies. That is not OPINION. That is nothing like OPINION.

[olasek]

Those quotes were all one or two line SUMMARIES of earlier research/studies.

Ok, so..?? I don't see any consensus there, anyone reading those "quotes" of yours will have impression that there research in progress, some seem skeptical, some see some faults (some see faults in the round dial too), see need for my work but no one blasts tape as unacceptable, NASA is clearly favorable and so are others. Situation is more like 50-50, undefined. Also these are all fairly old quotes, anything new, anything recent? Clearly if some would find the tape so objectionable that would be a flurry of research, studies, simulations, culminating with some definite outcome - nothing of the sort is present.

[Gretchenfrage]

@ Dozy

You are right, "everyone knows .." does not cut it. I should have gone the more scientific way, as FGD135 does perfectly.
Read his contribution very carefully and you might grasp what we meant.

A part from these data, let me state that i have flown 6 airliners of three manufacturers, 3 with round instruments and 3 with tapes. I can only sustain the above mentioned quotes and data.

What airliner with what dials have you flown??

I don't want to belittle your knowledge, but please acknowledge that the pros out there might have some more experience before you call them intellectually dishonest when they put their opinion up front. Especially when backed up by scientific data.

[Clandestino]

Congratulations Gretchenfrage, FGD135 and Machinbird. You have discovered that the whole aeronautical world is wrong and that replacing round dials with electronic speed tapes in every airliner and increasing number of GA types is grave mistake. Now, this valuable piece of information should be brought to attention of every aeroplane certifying authority in whole wide world until they mend their wrong ways and I'm afraid that just posting your valuable opinions here won't help that cause. Gentlemen, you should be more proactive towards the well-meaning but obviously misinformed folks that design and certify our cockpit instruments and go straight to them with your amazing discovery. Your unquestionable expertise on cockpit design has realy overgrown the bounds of PPRuNe.

let's look at the synergy they create when together.

Maybe to add one ingredient more - idea that having instructor in cockpit somehow abolishes "spooled up" part of stabilized approach criteria.

[Gretchenfrage]

Congratulations Gretchenfrage, FGD135 and Machinbird. You have discovered that the whole aeronautical world is wrong and that replacing round dials with electronic speed tapes in every airliner and increasing number of GA types is grave mistake. Now, this valuable piece of information should be brought to attention of every aeroplane certifying authority in whole wide world until they mend their wrong ways and I'm afraid that just posting your valuable opinions here won't help that cause. Gentlemen, you should be more proactive towards the well-meaning but obviously misinformed folks that design and certify our cockpit instruments and go straight to them with your amazing discovery.

Very valuable contribution not, quite in line with most of what you write.

1. Not one of the aforementioned has stated that the "whole aeronautical world" is wrong. It's your paparazzi-style journalism. We stated that one design is better than the other, which is a very different statement, at least for the educated.

2. This and other statements about design have been brought to attention of manufacturers and authorities, you might just haven't noticed or read. These bodies didn't ignore such statements, but deemed the existing systems adequately enough. Both standpoints can be accepted, there is no black and white, not right or wrong here, at least for the educated. Even remaining on one's opinion is no crime.

3. That this problem suddenly came back, after it was discussed abundantly in the early 90ies with the advent of the MD11 and A330, has to do with a unexpected rise of accidents in airliners lately with loss of speed and subsequently loss of control. It is a privilege of the free and educated world, and paramount to our all's professional strive for the highest level of safety, to reconsider design deficiencies almost forgotten and to re-discuss them.

[DozyWannabe]

Originally Posted by Gretchenfrage

... in the early 90ies with the advent of the MD11 and A330, has to do with a unexpected rise of accidents in airliners lately with loss of speed and subsequently loss of control.

Including the Birgenair B757 and WCA MD-82, both of which have round-dial analogue airspeed gauges, if I'm not mistaken.

FGD's summaries are of papers (one of which was a commercial study for Boeing in 1985) - *not* scientific studies. As such, of course it's possible that those papers would have an inherent bias towards the round dials, if that was what the author of the paper was trying to prove. In fact a significant number of those quotations come from a paper I linked to, and I've already said that the particular paper doers not seem to be scientific or sufficiently rigorous. Much of it does indeed read like an opinion piece.