SOP call-outs. More logic needed?

[A37575]

Sensory Illusions (SKYclip) - SKYbrary Aviation Safety (http://www.skybrary.aero/index.php/Sensory_Illusions_(SKYclip)?utm_source=SKYbrary&utm_campaign=31704dfeb2-509_Sensory_Illusions_15_03_2017&utm_medium=email&utm_term=0_e405169b04-31704dfeb2-276530305)

In this short animation of a jet transport go-around in IMC, the viewer is shown the PF pushing the nose down to almost zero body angle because he is experiencing a sensory illusion of climbing steeply.

The PM says nothing initially as he waits for the PF to correct but the aircraft PF is still descending instead of climbing. The PM then says "Pitch" but the PF fails to react since he is still under the impression the aircraft is climbing too steeply. End of animation.

SOP's vary of course but it seems to me that simply calling "Pitch" is useless. Far better calling the current pitch attitude such as "Pitch zero" or similar or "Pitch 25 nose high" should alert the PF of an impending problem and cause a change in the right direction. Similarly some company SOP have a requirement to call "Speed" if outside set parameters. Would it not make more sense to call "Speed high" or "Speed low" and in turn this would enable the crew to compare IAS indicators rather than a bald "Speed?"

Same with navaid tracking or height keeping. 'One dot left (or right)" makes more sense than "localiser" "150 feet high/low" makes more sense than the world "Altitude"

[safetypee]

A37575, the problem with attempting to better match procedures to a specific scenario is that there will always be situations where the changed procedure would not apply.
A major shortfall of the animation is that it implies that one pilot will be able to monitor the situation, unaffected by the illusion.
Illusions pose a significant hazard in aviation as both crew can suffer at the same time. In this instance the need to push nose down could be interpreted as correct for the situation by both crew. SOPs are unlikely to help; however having a strong understanding of what is normal for the situation may provide some indication of the crew suffering an illusion.
Learn for every flight, particularly to improving a wide ranging understanding of what is 'normal'.

In other non-illusion situations involving monitoring, I agree that careful thought is required in forming SOPs and calls.
The base line is what is being monitored, when, and against what datum.

Re recent CRJ accident involving an attitude failure. The PM called 'pitch' according to his EFIS reflecting an incorrect manoeuvre by the PF. However the PF flew according to his erroneous display, the 'PIT' comparator alert being inappropriately suppressed.
The situation required an automatic cross monitor alert (PIT), which would involve a procedure to refer to the st by instrument.
The PM 'pitch' alert assumes a serviceable aircraft and alerts the PF to an aircraft deviation, to be corrected by referring to the PF EFIS, which was being done, except that the IRS was at fault.
The crux of the accident is that without PIT annunciation then the two possible situations cannot be quickly resolved, and that a 'pitch' call can be more confusing or even taken to confirm correct PF action to resolve an apparent upset.

[A37575]

The PM called 'pitch' according to his EFIS reflecting an incorrect manoeuvre by the PF

Thanks SP. I saw that CRJ accident report today. It confirmed my assertion that calling "Pitch" meant little to the PF. if the PM had called "Pitch very low" it would have made sense and maybe (speculation perhaps) it would have had the effect of causing the captain to quickly compare the standby ADI to his own ADI and that of the co-pilot's, and then take appropriate action. That is of course if he had the required pure flying skill to do so.

From my experience in simulators, 95% of exercises are flown on full automatics and pilots come out of the simulator proud they have done a good job of flying. Boxes ticked and all that. They conveniently forget it was the autopilot that did the good job - not necessarily the programming pilot. Automation addiction is seductive.

It could be argued the captain of the CRJ lacked currency in manual raw data instrument flying and paid the ultimate price. 15 minutes of manual instrument flying in a simulator every six months means nothing in the real world.
Illusions pose a significant hazard in aviation as both crew can suffer at the same time.

Sensory illusions are often offered up as a possible cause of IMC accidents. A convenient catch-all maybe? Easy to say but impossible to prove. I believe lack of basic manual instrument flying skills are much more likely to be contributory causes of IMC accidents

[B737C525]

Alternatively...

The helicopter lifted from the helipad into a hover at 1924 hrs, and hover-taxied to the middle of the paddock where it came to the hover again; the pitch attitude in the hover was slightly nose-up. The helicopter then began to climb, almost vertically at first. At a height of approximately 32 ft agl, its pitch attitude changed from the slightly nose-up hover attitude, pitching nose-down, before the helicopter began picking up forward speed, continuing to climb.

At about 120 ft above the ground and with a nose-down pitch attitude of approximately 15°, the co-pilot said “nose down [commander’s first name]”. It could not be determined from the recorded voice whether this was to highlight the nose-down pitch attitude or a prompt for more nose-down pitch, but more forward cyclic input was applied. In the following second the helicopter crossed the boundary trees and started to descend with increasing ground speed, forward cyclic input and nose-down pitch attitude. Progressively more collective was also being applied, with the resultant increase in engine torques.

The co-pilot repeated the “nose down” words; again it could not be determined whether it was an observation or a request and again a nose-down input was made. The cyclic inputs, whilst still in the forward sense became more erratic, with one aft input recorded in the nal seconds. The nose-down pitch attitude started to reduce from the peak recorded value of 35° as the helicopter descended through 100 ft agl. The collective input progressed to 100%, the engine torques increased but the rotor speed could not be sustained at the nominal 102% and started to reduce. The last nose-down pitch attitude recorded by the combined voice and ight data recorder (CVFDR) was 25° with the helicopter 82 ft above the ground, descending at 2,400 ft/min, with a ground speed of 90 kt.

The helicopter impacted a line of large hay bales lying across a eld. The cabin structure was destroyed and all the occupants were fatally injured in the impact sequence.

http://https://assets.publishing.service.gov.uk/media/56162ac0e5274a625100000f/Agusta_Westland_AW139_G-LBAL_10-15.pdf

[Intruder]

"PULL UP! PULL UP!"

It works for GPWS, why not in the OP's scenario?

[safetypee]

why not in the OP (illusion) scenario.
Because EGPWS is able to identify terrain with a high level of certainty. Whereas with illusions it is not possible (no certainty) for the human to determine if an individual is suffering (which one), or manage situations where both pilots suffer illusion simultaneously.

[Intruder]

But, IF the human IS going to say something, why not something recognizable, like "PULL UP!"?

[safetypee]

say something, but how does the 'speaking' pilot know what and when to say something if they too are suffering the effects of illusion? Is the call correct or incorrect, who can decide?

If the issue concerns the content of the call, then I agree that only the inappropriate condition should be called, and that this should include information of how to recover.
This also raises the question of the experience required to judge non-annunciated situations requiring verbal intervention. Would a novice pilot's erroneous perception conflict with the more experienced pilot's correct perception; who is the better monitor.
Also see the problems of perception vs role, where with increasing calls and procedures the PM may be concentrating on the 'primary' task to check modes and make status calls, such that there is little flight path monitoring or spare mental resource to make an alerting calls.

Do You See What I See? Effects of Crew Position on Interpretation of Flight Problems. (http://homes.lmc.gatech.edu/~fischer/SortingTM.IH-014_Orasanu.pdf)

[Intruder]

Once more... IF the decision has been made - regardless of how/when/why - and the PM IS going to say something, why not something recognizable and valuable? "PULL UP!" is both, where "Pitch" is nebulous at best, especially if the PF is disoriented.

AFTER the nose-low attitude is corrected and they are flying away from the ground, they can discuss failed PFDs, failed PIT comparators, and sensory illusions ad nauseum. When time is critical, though, a clear, recognizable command with one logical response is best.

[RAT 5]

"PULL UP! PULL UP!"
It works for GPWS, why not in the OP's scenario?

IF the decision has been made - regardless of how/when/why - and the PM IS going to say something, why not something recognizable and valuable? "PULL UP!" is both, where "Pitch" is nebulous at best, especially if the PF is disoriented.

An interesting discussion; considering all the other topics about human behaviour.
Let's consider the comparison with EGPWS. "TERRAIN TERRAIN" is advisory, as is "TOO LOW FLAP/GEAR". i.e. there is an error in what you are doing so pay attention and correct it. If you don't do that it may evolve in to "WHOOP WHOOP PULL UP." That then becomes a command. So EGPWS has two levels; advice of an error and then a command.
It is very apposite to make the point that an advisory SOP call assumes a fully functioning a/c.
IMHO the philosophy of an advisory SOP is to alert PF to an error in a parameter that has gone AWOL from the scan, and to draw attention to it so the PF can correct it. The SOP wishes to stay clear of PM commanding PF what to do and possibly causing irritation to PF and perhaps a negative effect. If PF doesn't act to correct the erroneous parameter then PM can resort to a command, i.e. as EGPWS does. It is a real skill and to be used wisely. I see many PM cadets, in their TR courses, nagging "Speed" "HDG" worse than my wife. What they don't do is monitor Thrust and advise of that BEFORE the "Speed" call that is the inevitable result of incorrect thrust. I spent many moments urging them to shift their fixation from PFD. (Even the ND/MAP was lacking interest??)

One scenario that startled and disappointed me: line flying with a low experience F/O on approach. All going very nicely down the slope; flaps coming out, manual flight thrust at idle, landing flaps selected. PF then set an N1% that was too high for the conditions. Speed started to creep up, but as yet was within speed limits but trending to go beyond. Not what you want on short finals. I called "Thrust" and PF shoved the levers further forward without looking. I quickly followed up with a "SET X N1%." The discussion afterwards was: PM "I assumed you were warning me thrust was too low." I learnt something from that. Our SOP was to call the parameter error and Thrust was not one of them. According to SOP PM would wait until speed was outside limits and then call "Speed." That is reactive not proactive. Both have a role.
The root cause of the problem was because PF set a thrust N1% without looking at the gauge, just a guess with lever position. That's not uncommon in the sim as some 'smart SFI's' teach to set thrust via the green speed arrow. Yes, that can work, but it is reactive rather than being proactively in command.
It also discourages scanning away from PFD, but that's another issue for another day.

[zoigberg]

I worked for one company where, for example, PM would call 'bug plus 10' for speed deviations. I am now with a company where the speed deviation call is 'speed'. I agree totally with the thread starter that a prompt of the amount of deviation is far more useful, and far quicker to process.

[safetypee]

RAT 5 :ok:
Intruder, there is no disagreement with the suggestions of what to say, but it is essential to consider how the decision to say something is reached.

The OP questions the logic required. The conditions and assumptions before the 'logical' IF need to be well-considered and clearly stated before creating a procedure.
The SKYbrary video clip starts with the assumption that a head-up pilot will suffer an illusion, opposed to being head down - inaccurate. The most demanding illusions are those which challenge the belief in the instruments, i.e. suffering an illusion when head down. Instrument displays do not provide immunity, but are the basis of disciplined attention.

The procedure requiring a call assumes that the monitoring pilot is not suffering an illusion, which cannot be assured; as discussed in previous posts.

When the call is given, including pitch direction, who's view in correct. This cannot be established without an independent arbitrating view from someone not suffering an illusion - which cannot be assured.

Thus to mitigate the hazard of illusions we need to understand the circumstances when they could occur and the importance of believing the flight instruments, whatever we might feel.

When assessing the risk of illusion, it is also important to consider the frequency in addition to the potential outcomes. Somatogravic illusions depend on acceleration; in commercial operations the risk during take off might be higher than for a go-around because of greater speed change and rate.
GA procedures may only involve a 10 kt speed change, less than 3 kts/sec, which is much less than the example given elsewhere in SKYbrary. (also applies to deceleration)
A greater threat during GA might be the perception of pitch change based on aircraft pitching moment from thrust and configuration change thus inappropriate crew intervention; then again there is trim....
Each situation requires assessment according to aircraft type and operation; don't assume that someone's else's procedure will work in your operation.