Wikiposts
Search
Safety, CRM, QA & Emergency Response Planning A wide ranging forum for issues facing Aviation Professionals and Academics

Monitoring & Intervention

Thread Tools
 
Search this Thread
 
Old 26th Feb 2012, 17:28
  #1 (permalink)  
Thread Starter
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
Monitoring & Intervention

An aspect of the India Air Express B738 overrun (http://www.pprune.org/rumours-news/4...-crash-27.html) and similar, involves the apparent failure of the monitoring pilot to provide timely or effective alerts, or intervention to avoid a fatal outcome.
A ‘failure’ of the monitoring pilot to intervene is often cited as a factor in accidents. If considered in isolation and particularly with hindsight, this may be labeled as a root cause - as deviant behavior, which is not necessarily appropriate.

The frequency of accidents involving ‘failure to intervene/take control’ in comparison with normal ‘safe’ operations is very low. This could suggest that there should be many successful interventions in normal operations (saves vs failures), including some ‘taking control’. However, there is little, if any evidence of such extreme interventions.
The safety callouts made by the monitoring pilot in the India Air accident and subsequent behaviour was relatively normal, albeit in hindsight, late, and not as required.
This behaviour, and apparently that of monitoring pilots across industry in general, is not as expected by the concept of monitoring; this suggests that all operations have similar weaknesses.

The interventions which I have experienced (or witnessed) fell into three categories:

Type 1. There is a very small number of pilots who experience increasing apprehension of situations which they have not previously encountered, these may be feared due to inadequate knowledge, e.g. stalling, unusual attitudes. Perhaps this class of person should not be a pilot, as a failure to control surprise / fear can be debilitating. (Those which I encountered were predominantly during demonstration / training flights).

Type 2. In normal operations, a low experience ‘novice’ monitoring pilot can provide alerting, but usually late in a developing adverse situation. There appears to be an inability to comprehend the outcome of events (projection - situation awareness). Instead there is reliance on the PF for ‘education’ or a belief that the situation can (will be) corrected, e.g. not appreciating that a late descent and continued use of normal procedures will result in a high/fast approach.
This represents a skills discrepancy in awareness – what and when to scan (why) and how to project and asses an outcome which has not been experienced / considered previously. This is a classic ‘getting behind the game’ behaviour, or being somewhere where you brain hasn’t been first. Any intervention usually states the obvious – what has happened, not what may evolve.
In extreme situations, the inability to keep-up results in an apparent failure to elevate the level of intervention or intervene in time.

Type 3. Where both pilots are highly experienced, the monitoring pilot alerts a disparity very early during an evolving situation. This represents good awareness and projection ability based on experience or visualisation and anticipation. Adverse situations rarely develop as the monitor’s intervention is both timely and with a helpful suggestive attitude – ‘consider this, or this could occur’, i.e. thinking ahead.

In general the interventions (2 and 3) are not driven by SOP deviance, instead they orrigionate from the developing situation – situation assessment.

I wonder what the industry’s experience of monitoring / intervention is:
Is the categorisation of pilots / experience as above, a common feature across the industry?
Are there many significant interventions in normal operations (no incident), vice routine helpful comment / crew interaction?
Is situation assessment the dominant aspect for triggering an intervention, or is SOP ‘mantra’ used (checks by numbers/rote)?
What are the general views of situation assessment and projection skills and training?

Do monitoring / intervention work, or is the concept flawed?
safetypee is offline  
Old 26th Feb 2012, 18:59
  #2 (permalink)  
PJ2
 
Join Date: Mar 2003
Location: BC
Age: 76
Posts: 2,484
Received 0 Likes on 0 Posts
safetypee, an excellent thread.

I wonder if the nature of the emotion generated in the AIE thread doesn't, in some way, show us one aspect of why, and how, interventions may not occur, or may not be timely?

I truly don't mean this unkindly - I'm making an observation and thinking about it...there almost certainly isn't one among us who has at some point in his/her career, not kept his/her mouth shut so as not to offend/insult/embarrass, or hasn't spoken up out of fear of retribution, dismissal of one's "silly" thoughts, embarrassment about possibly being wrong, or has disagreed vehemently which tends to re-direct the discussion into emotive responses rather than dispassionate replies? For example, we see just a hint of it here with new posters who know the reputation of PPRuNe..."I'm new here, please don't be hard on me, but...".

This isn't about "just growing thick skin". This is about fundamental human performance factors which come with emotions whether we wish it or not. CRM training and techniques are designed to take these factors out of the cockpit conversation and I think your characterization of the industry is correct in that CRM interventions work almost 100% of the time but not quite.

Such interventions are not always delayed because of ego, intimidation or an otherwise emotionally-charged atmosphere in the cockpit. Sometimes one is a highly-respected, skillful pilot with a stellar reputation and when a mistake or error in judgement occurs, it is not always a straightforward matter to point it out. Now, such a pilot would perhaps have already addressed the issue by making it "safe" to speak out and even make it a requirement of the operation in initial briefings, especially with unfamiliar or new crew members.

Aside from the examples brought up by aterpster I'm wondering now about the Safety Pilot on board the THY B737 at Amsterdam. What occurred there that nothing was said until it was too late? What about the complete absence of SOPs in the immediate response to the loss of airspeed information on AF447? Why no intervention? Cockpit gradient is only part of these discussions I think.

The piloting community is, despite occasional yet characteristically strong disagreements on techniques, aircraft types and so on, a deeply loyal community to its own. Part of that loyalty may be a misplaced sense of respect which may make one hesitate slightly before speaking out. In extreme cases of course, this can be a psychological and even economic matter in some cultures where fear of loss-of-face, or even loss-of-job is the driving factor and those kinds of things are well beyond airlines' capacity or even willingness to resolve.

Just some intiial, off-the-cuff thoughts on the question.
PJ2 is offline  
Old 29th Feb 2012, 01:05
  #3 (permalink)  
Thread Starter
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
PJ, emotion and bias can contribute to a reluctance to comment, similarly there may be problems with highly respected / experienced crews, or due to cockpit gradient.
However, before these problems are encountered, the PM must have formed an understanding of the situation. This understanding would trigger an assessment (decision) to intervene or not, and if so, how.
The significant problems of intervention appear to be at this stage of the monitoring process; either the situation is misunderstood and it is concluded that intervention is not required, or with understanding, the need to intervene is withheld due to other reasons including emotion/personality/cultural issues.

My interest is with any weaknesses in situation assessment and experience as it is these which could affect both PF and PM.
In the worst case the PM may not understand the developing situation, neither the PF; in this instance the concept of independent dual assessment (cross monitoring) fails.
A more likely circumstance is that both PM and PF have a partial understanding, which with added complexity of time pressure and low experience level, results in a late and weak intervention.
Furthermore, the increased workload demand in modern operations, particularly on the PM, reduces the opportunity to monitor.
A further aspect is that monitoring skills may be poorly taught or understood.

In many discussions, e.g. AMS, awareness is often overlooked, instead favouring the more emotive personality/cultural aspects. However, if this type of accident is reviewed avoiding hindsight bias, situation awareness, experience, and timing are often dominant contributors. Also, they are issues which evolve relatively early in the chain of events, e.g. AMS approach vectoring, high / fast, late GS capture.
safetypee is offline  
Old 2nd Apr 2012, 13:47
  #4 (permalink)  
Thread Starter
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
With continuing interests and without challenge to the categories in #1 (nor any supporting evidence of them being the norm), I have found an alternative view. This involves ideas of surprise within the concept of resilience.

Surprise can be divided into situational and fundamental. (Ref)
Situational surprise is what we normally define as a surprising, unexpected situation. We know how things work, what might fail, and thus what the consequences might be. Thus even when a sudden event occurs - a surprise - we have sufficient background knowledge to enable understanding and the selection of a course of action, e.g. engine failure. Also we can anticipate, monitor and provide defenses/responses to these unexpected but foreseeable situations.

Fundamental surprise involves those situations which refute our basic beliefs about how things work, and thus are “inconceivable”. These situations either cannot be foreseen (comprehended), or we choose not to see them beforehand, i.e. a rare, obscure, or multiple system failure (AF447, 737 AMS).
The lack of foresight or understanding depends on an individual’s level of knowledge and experience (how to use that knowledge). Due to these limitations, situations involving fundamental surprise cannot be defined in advance and thus cannot be monitored or have preplanned actions (SOPs).

The deficiencies are not in a system design or circumstance, but with the level knowledge which is available for use before and during an event. Could we have foreseen a triple speed sensor failure in an aircraft which has a triple redundant control philosophy – we are naturally biased to conclude that such failures (or the result) are unforeseeable (not the same as impossible, cf Black Swans).

The ideas above appear to be very similar to the categories in #1. An inexperienced pilot (case 2) might be more likely to interpret a surprising situation as one of fundamental surprise; opposed to an experienced pilot (case 3) who manages to comprehend the surprising situation due to greater knowledge / experience.

These ideas may enable some defensive aspects for the inexperienced. Whilst we cannot anticipate or monitor all situational conditions, we can be prepared for surprise by monitoring our resources of workload and capacity to respond – open mindedness, questioning, and considering other alternatives.

The reference also provides interesting views on how we learn from surprise – in our post-accident analysis do we inadvertently convert fundamental surprise to situational surprise, and thus the learning is not as good as it should be - cf discussion on recent accidents – the effect of hindsight bias.

Ref – intro: http://www.resilience-engineering-as.../Papers/33.pdf
safetypee is offline  
Old 12th Apr 2012, 07:05
  #5 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Monitoring & Intervention for Graceful Degradation

Monitoring & Intervention is important in promoting Graceful Degradation of "effective aircraft" (System + crew), so it's necessary. But to "work", depends on many factors.

You need capable PM. Sometimes very capable PM. A possible flaw is too much savings in PM preparation. The concept is valid, necessary. The requirements are sometimes, high.

In "Black Swan" scenarios the concept likely will not work. Despite to be valid.

Will think first based in what we have (factual info) on AF447.
RR_NDB is offline  
Old 9th May 2012, 11:39
  #6 (permalink)  
 
Join Date: Jun 2006
Location: Australia
Posts: 1,186
Likes: 0
Received 0 Likes on 0 Posts
Do monitoring / intervention work, or is the concept flawed?
It might be the ideal but in my experience it is more often than not clouded by ethnic culture. This is a culture bred into some where seniority, respect for age and perceived experience - and finally a marked fear for job security - strongly influences against intervention. A first officer in these cultures may well see an adverse situation developing - whether caused by the captain's incompetence or something external such as bad weather ahead. Ideally, and no doubt having been taught about CRM, he will say something. It may be an inaudible murmur on the intercom so as not to create perceived offence, or it may something be stronger. Or he may say nothing thinking the captain always knows best because he is the captain.

Only a few months ago this writer was giving simulator refresher training to an airline crew from China. During the pre-flight briefing on the dangers posed by an unstable approach, I posed the scenario of an unstable approach that was so bad it was must be obvious to the first officer that the captain was fixated on continuing the approach and that no doubt his mind-set was "it can't happen to me" so he continues to barrel down the glide slope at high speed.

Everything is set for a disaster if things continue as they are.

I asked what options are open to the first officer to prevent a certain accident if the captain continues ? The same question was asked of other crews of the airline on the same course of refresher training. The first officers hours varied between 300 and 2000 hours total time.

All agreed that advising the captain of being too fast, or too high, or too low would be the first action. A few said they would suggest to the captain he should go around. Not all agreed with that - indicated perhaps by their refusal to comment in front of the others.

Asked what would be their options if the captain ignored their calls and pressed on regardless, everyone went to ground and were silent. Asked if they would consider taking over control from the captain and going around, all except one remained silent. .
That first officer hesitated then slowly put up his hand to reply and I am sure his words echoed in the minds of the others who had said nothing.

He said "I would never consider taking over control from the captain because he is the captain of the aircraft".

Their culture dictates absolute respect for their elders, no matter what private thoughts they may have about their technical or mental competency. Loss of face is a characteristic of many Eastern cultures. The thought of causing loss of face to the captain if his first officer dared to take over control, is something too unbearable to think about. That means in some societies, rather than take decisive action, there are first officers who would prefer to stoically accept whatever fate awaited them and their unfortunate passengers..

Last edited by Tee Emm; 9th May 2012 at 12:09.
Tee Emm is offline  
Old 9th May 2012, 13:08
  #7 (permalink)  
Thread Starter
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
There is a good ‘walk-around’ monitoring and culture aspects in the paper below.
It’s also relevant to the CRM thread.
Industry CRM Developers - Situational Awareness Management Course Outline
safetypee is offline  
Old 10th May 2012, 21:35
  #8 (permalink)  
 
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
MONITORING

Im sorry, but I dont get it.One guy flys (PF) and the other guy monitors him flying (PM).I fail to see the need for yet another system to catch/stop pilots from running off the rails into danger territory.Standard callouts and advisories from the PM ("SPEED!!") and from the back up GPWS callouts when things get a little hairy are enough.If the response/correction from the PF is not timely enough, these industry-standard callouts can be said a lot louder and repeated if deemed necessary ("SPEED SPEED SPEED"!!) but are enough in my opinion.
piratepete is offline  
Old 11th May 2012, 13:42
  #9 (permalink)  
 
Join Date: Mar 2006
Location: England
Posts: 995
Likes: 0
Received 4 Likes on 2 Posts
Tee Emm, the point being questioned appears to be the awareness and realisation that something is not as required. Speaking up, challenging, and intervention are options of what to do after the ‘monitor’ has been triggered.

piratepete, you assume that all situations will be covered by an SOP or that a suitable SOP is self-evident and its purpose and limits are known by the monitor.
Your example of SPEED means what ? Look at the airspeed, an error from the required value, a limit exceedance, or a failure. A better SOP might include information in the callout; ‘ Vref + 15’, for use on the approach below 1000ft.
Consider a circling procedure where the PF has briefed a slightly higher than standard turn speed (+5) for turbulence and a small addition for the increased bank angle. Is the approach callout still valid – yes, but it is more for information than alerting and might only change between the two modes at 300ft. Where is this defined, who judges that, who calls it – and what if the flight requires that monitor to look out to assist judgement of the turn, which monitoring function is more important?

There are deficiencies in monitoring; it appears to work when things are routine and progressing reasonably normally, but if not, then monitoring can fail. This is a human failure which may depend on what they are being asked to do, but also not aided, or even confused by, poor SOPs and assumptions.
PEI_3721 is online now  
Old 11th May 2012, 23:19
  #10 (permalink)  
 
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
SPEED!!

DESCENDING
Speed requirement by ATC is 250, indicated speed is 280 so a call of "SPEED" would be appropriate by PM

APPROACH
ATC require 180 due to traffic behind/ahead but ias is 200, so a PM call of "SPEED" would be appropriate

2000 feet aal (IMC).IAS is very close to the current flap setting limit speed and is say accelerating due to eg a tailwind, so a PM call of "SPEED" would be appropriate to avoid damaging the flap drives etc.

1000 feet aal (IMC).Landing flap is selected, Vref plus 15 (say 135) is our TARGET speed due to gusts at the runway surface but ias is 146 (more than target plus 10- a common industry-wide trigger for the "SPEED" call), PM should call "SPEED".

300 feet aal Due to a drop off in headwind, ias drops below Vref, say its now 118, requiring the PM to call out "SPEED"

Does this answer your question? Pete.
piratepete is offline  
Old 12th May 2012, 02:24
  #11 (permalink)  
 
Join Date: Jun 2010
Location: USA
Posts: 245
Likes: 0
Received 0 Likes on 0 Posts
Tee Emm, the point being questioned appears to be the awareness and realisation that something is not as required. Speaking up, challenging, and intervention are options of what to do after the ‘monitor’ has been triggered.
Exactly. This is the same flaw he makes in the "CRM is useless" debate so it's not surprising he repeats it here. In the first instance CRM is not about changing anyone's behavior; it's about changing their mindset. Whether that change in mindset changes the flight crew's behavior all depends on the specific circumstances. In many case, even the majority of cases, it won't change the behavior of the flight crew at all. That doesn't mean CRM isn't working; it precisely means it is working.

The same thing with the idea of monitoring. The difference safetypee is trying to articulate, if I understand him correctly, is the difference between active and passive monitoring. The difference between these two states isn't the difference between whether the PM intervenes or not; it's what is going on in the PM's head. In many, even most cases, the difference between these two mental states will have zero impact on the flight crew's behavior. But when things start to go awry the difference between active and passive monitoring is stark.

Safety doesn't start with the hands; it doesn't start with interpersonal interaction. It starts with what is going on in the minds of the crew. The hands and the interpersonal interactions flow from the mind. That's the key point of CRM and a key reason there is even a PM to begin with.
MountainBear is offline  
Old 12th May 2012, 04:00
  #12 (permalink)  
 
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
PYSCHO BABBLE

iM SURE YOU MEAN WELL WITH WHAT YOU ARE SAYING, BUT TO ME NOTHING YOU SAY MAKES ANY SENSE IN A PRACTICAL WAY........SORRY.CRM is all about BEHAVIOUR and ATTITUDE
piratepete is offline  
Old 14th May 2012, 15:43
  #13 (permalink)  
 
Join Date: Mar 2006
Location: England
Posts: 995
Likes: 0
Received 4 Likes on 2 Posts
piratepete, I had hoped that the question would generate some debate; not IMHO, result in a statement of a poorly constructed SOP (#10).

The isolated (unqualified) ‘SPEED’ call does not help to clarify the situation and conveys little explanatory information: Speed high or low?
The point which may have been missed, and I reiterate, is that ‘SPEED’ requires that both crew members understand which context it is being used in, and thus what the call means in that situation.
However, one purpose of monitoring is to alert the handling pilot to an error which could be due to a poor understanding of the situation, but understanding the situation is a requirement for using the SOP. Thus the call could be puzzling – generating confusion; this adds workload. The reverse is also true, if the PM originates the error - mishearing the ATC request or the approach briefing; the inappropriate SPEED call could also be confusing.

A better method of operating / monitoring (the ‘SOP’), could be an alert formulated by the situation, e.g. “ATC has requested 250 (180)”. This would direct attention to airspeed, and also crosscheck the crew’s joint understanding of the target speed for the situation at that time; this would involve less workload and opportunity for confusion, and provide an error check from wherever it originates.

Last edited by PEI_3721; 14th May 2012 at 15:44.
PEI_3721 is online now  
Old 14th May 2012, 21:47
  #14 (permalink)  
 
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
BOEING FCTM?

PEI
Have you looked at a Boeing FCTM lately? (any model).Have you studied the standard callouts section? What does your airline SOP(standard callouts section) say with respect to speed excursions? Every airline im aware of has these standard calls, and if the PF is confused when he is 30 knots fast and hears the PM call SPEED then he needs to be re-educated about S.A.You are advocating too much talking....its very distracting especially at a critical stage of flight like at 300 feet say on final.You have your own ideas but they wont get traction anywhere.I have had the priviledge of writing/designing the SOPs for several airlines and use these flight safety foundation recommended calls in them.
When im in the SIM as TRE and hear trainees doing what you suggest giving too much feedback as PM ie excessive talk talk talk, it will certainly be brought up in the debrief as a criticism (within reason).
piratepete is offline  
Old 14th May 2012, 22:43
  #15 (permalink)  
 
Join Date: May 2012
Location: Earth
Posts: 88
Likes: 0
Received 0 Likes on 0 Posts
Hiring practices are such that FOs these days are unlikely to catch the problem that the Captain missed...so typical is a nice quiet cockpit, both flying unaware to the scene of the accident.

In the rare cases where the FO actually caught the problem that the captain did...many times they just argued to the scene of the accident.

I will say this about CRM...empowering the FO to call for a go around, with out retribution is/was a good thing. I'd be curious how many accidents were averted because the FO knew he could hit the panic button will no ill effect to his career.
Sillypeoples is offline  
Old 14th May 2012, 23:41
  #16 (permalink)  
 
Join Date: Oct 2007
Location: fort sheridan, il
Posts: 1,656
Likes: 0
Received 0 Likes on 0 Posts
hiring practices...
hmmmm


back in the good old days when you had to actually have a few hours (more than 1500) to even get an interview, may pilots had to INSTRUCT to get flying time.

And those instructors learned (sometimes the hard way) how to speak up when things were going badly...and to take control in order to stay alive.

those skills transferred well to the airline cockpit. now, many captain fly just fine and never need someone to speak up...but once in awhile we all screw up a bit and the diplomatic, former instructor, knew how to get things back with the right turn of a phrase.

You didn't keep students(and therefore monetary income) when you slapped their hands, or yelled at them...but a quiet...hey frank, you're a little slow there, usually would bring things back to the good. Lindbergh himself said (paraphrase) that you never really knew HOW to fly, untill you could teach someone else how to fly.


Sadly, many have not had the experience of being an instructor.
sevenstrokeroll is offline  
Old 14th May 2012, 23:53
  #17 (permalink)  
 
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
Excellent!

SSR thats a very good post and very very true actually.I find it hard to believe that some of the quite silly posts here are being made by pilots, they are either silly or dont make sense and this is the giveaway.........too many wannabees who dont know their onions.
piratepete is offline  
Old 15th May 2012, 03:59
  #18 (permalink)  
 
Join Date: Oct 2007
Location: fort sheridan, il
Posts: 1,656
Likes: 0
Received 0 Likes on 0 Posts
dear piratepete

thank you .


I remember when I had 250 hours, commerical single/multi, and cfi. How easy my life would have been if I could have gotten on as a jet copilot somewhere. Wow...and a flight attendant to boot! (expression as in icing on the cake...not to kick).

I barely could find a job instructing.

These posters don't know and can't even imagine what it takes to make a really good airline pilot...and it takes more imagination than these bloaks have.

good luck to you piratepete and happy landings
sevenstrokeroll is offline  
Old 15th May 2012, 13:10
  #19 (permalink)  
 
Join Date: Mar 2006
Location: England
Posts: 995
Likes: 0
Received 4 Likes on 2 Posts
piratepete, we appear to have different backgrounds and views.
Monitoring – actually the intervention part, isn’t about talking, it’s about communicating relevant information, as and when the situation demands.
Monitoring requires good situation awareness; intervention also requires awareness including the importance of any difference between the existing situation and that required, i.e. when to intervene (speak) .. we don't need an SOP for that.

Many SOPs result in clipped meaningless alerts which can be lost in the routine of operation; monitoring and intervention require a focussed exchange of information alerting or warning of a serious problem, to effect a change in activity / situation.

We need to be re-educated about SA”, on this I think we can agree.
PEI_3721 is online now  
Old 15th May 2012, 14:13
  #20 (permalink)  
 
Join Date: Jun 2000
Location: Australia
Posts: 4,188
Likes: 0
Received 14 Likes on 5 Posts
callouts can be said a lot louder and repeated if deemed necessary ("SPEED SPEED SPEED"!!)
"SPEED" by itself is an utterly meaningless call. The call should be amplified by saying if the speed seen on the PM's ASI is high or low. If the speed is low for the configuration for example and the PF's ASI reading does not agree then it points to a potentially serious discrepancy between the two ASI's.

In other words the PM says "Speed high" and presumably or at least hopefully the PF will make the necessary correction. But if the speed is not high by the view of the PF then he can say "Crap" there is something wrong with your (or mine) ASI.

Last edited by Centaurus; 15th May 2012 at 14:15.
Centaurus is offline  


Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.