Mismanagement of automation
HC
I wasn't comparing hours flown, you were stating the above.
Chip? Don't have any old chap - just enjoying the banter - I hope your flying (diagnostic automation management and real manual flying) is more accurate than your statements on here!
we aren't at war that much these days
Chip? Don't have any old chap - just enjoying the banter - I hope your flying (diagnostic automation management and real manual flying) is more accurate than your statements on here!
HC - I can't give you the training system you need because you won't pay for it - the argument about going bust is the malaise of the helicopter industry from what I read on this forum - someone will always do the job cheaper and it will always be about short-term gain rather than long-term planning.
As long as the industry accepts this status quo, all the wailing and gnashing of teeth about training falling short of needs will never change anything.
It seems the industry can't pull together to raise standards (because they are statistically pretty safe and the bean counters won't pay the extra) and the regulators are reactive rather than proactive.
The reality is that we will continue to accept crashes like Norfolk and Sumburgh as an acceptable attrition rate for the money that customers are prepared to pay. You only get as much Flight Safety as you are prepared to pay for.
This thread started as a discussion about mismanagement of automation - the answer is clearly - more training ; the question is 'who will pay for it'? Until you solve that dichotomy then things will continue to steadily decline.
I agree that the luxury of the mil system being allowed to throw more hours at any similar problem - although not as easy to do nowadays (even the mil has budgetary constraints) - isn't likely to happen in the civil market but there doesn't seem any push to even try.
One thing that might help is if pilots designed AP systems rather than engineers, then we might have less complicated failure modes, degraded modes, diagnosis of system failures and an industry standard for naming things.
Big buttons and bright lights works for me
My nephew has just finished his MCC with a well-known budget airline and the AP system on a modern airbus has so many warnings, both visual and aural, even of minor system malfunctions, that it is easy to miss something important when wading through the niff naff.
As long as the industry accepts this status quo, all the wailing and gnashing of teeth about training falling short of needs will never change anything.
It seems the industry can't pull together to raise standards (because they are statistically pretty safe and the bean counters won't pay the extra) and the regulators are reactive rather than proactive.
The reality is that we will continue to accept crashes like Norfolk and Sumburgh as an acceptable attrition rate for the money that customers are prepared to pay. You only get as much Flight Safety as you are prepared to pay for.
This thread started as a discussion about mismanagement of automation - the answer is clearly - more training ; the question is 'who will pay for it'? Until you solve that dichotomy then things will continue to steadily decline.
I agree that the luxury of the mil system being allowed to throw more hours at any similar problem - although not as easy to do nowadays (even the mil has budgetary constraints) - isn't likely to happen in the civil market but there doesn't seem any push to even try.
One thing that might help is if pilots designed AP systems rather than engineers, then we might have less complicated failure modes, degraded modes, diagnosis of system failures and an industry standard for naming things.
Big buttons and bright lights works for me
My nephew has just finished his MCC with a well-known budget airline and the AP system on a modern airbus has so many warnings, both visual and aural, even of minor system malfunctions, that it is easy to miss something important when wading through the niff naff.
The big advantage of flight automation is the savings in training and continual practice. There will be a time where a basic helicopter course followed by a Flight Management Systems course will enable one to be qualified to fly, sorry, operate helicopters on the North Sea. Then as individual costs go down then there will be glut of pilots available. The costs will then be reduced yet again because they can be offered peanuts for salaries as they are only automated bus drivers.
Until something that the pundits haven't dreamed up in the checklist happens.
Until something that the pundits haven't dreamed up in the checklist happens.
Crab - so your line is "we (in civvy street) can't have a similar amount of training as the mil do, so might as well gives up and go home". Seems rather defeatist to me. In my (old) company at least, we did significantly increase recurrent and conversion training time in response to the new technology. But also that new time tended to get filled with knee-jerk extra items following crashery (eg simultaneous double engine failure in the cruise) that were put in for, predominantly, political reasons.
We may not be able to increase training time to match the mil, but there is a huge amount of scope to make better use of what time we do have. We need to dump extra stable-door-closing political rubbish, we need to dump historically important but no longer relevant stuff. In fact we need to look at the task, the failure modes (human and technical) and the reasons why we still crash, and devise training plans accordingly. Oops, I do believe I am sounding like Geoffers! Well, nothing wrong with that!
Anyway, it's a shame you want to be defeatist rather than being part of that debate.
We may not be able to increase training time to match the mil, but there is a huge amount of scope to make better use of what time we do have. We need to dump extra stable-door-closing political rubbish, we need to dump historically important but no longer relevant stuff. In fact we need to look at the task, the failure modes (human and technical) and the reasons why we still crash, and devise training plans accordingly. Oops, I do believe I am sounding like Geoffers! Well, nothing wrong with that!
Anyway, it's a shame you want to be defeatist rather than being part of that debate.
HC - I suggest you re-read your own posts before you declare who is being defeatist in this matter.
I have no power to change what happens in the offshore world - you were in it for a long time as a very experienced operator and instructor I believe - how much did you manage to change it?
I don't mean to sound defeatist, I am just reflecting the comments made here by you and others - as you never tire of telling me, I know little of the offshore job.
You have lots of good ideas but who is going to implement them?
I have no power to change what happens in the offshore world - you were in it for a long time as a very experienced operator and instructor I believe - how much did you manage to change it?
I don't mean to sound defeatist, I am just reflecting the comments made here by you and others - as you never tire of telling me, I know little of the offshore job.
You have lots of good ideas but who is going to implement them?
Crab - well my time is done (a little prematurely) but in my time I think I made some differences to both the usage of automation and the training for it. It was certainly an uphill struggle with resistance to change from many quarters including some crusty trainers, the older generation of pilots, and the CAA.
But for example we ceased doing a manual ILS on every check, instead doing approaches with partial automation to tick the "manual" box. That being infinitely more useful. Engine failure on takeoff after DP was presumed to be flown hands off by the automation. We started flying overlay approaches (ie coupling to FMS to fly a VOR or NDB approach down to DA (and through the missed approach)). And lots more.
It was quite interesting to see the slow but inexorable shift in line pilots' attitude from "can't I just fly it manually, it's so much easier" to loving the automation, to getting a bit close to "automation dependancy". And that of course brings up another point which is that optimising training depends amongst other things on where the company is in its cultural transformation from steam to automation, and the provenance of each pilot. So for example converting a crusty old line pilot onto the 225, nearly all the emphasis was on the systems and automation. It was presumed that they had basic flying skills. However doing a type rating on a baby CPL, one had to include lots more manual flying so that the trainee would gain confidence and skill in flying something much bigger than a Hughes 300.
So looking back on the period 2005 to 2012 which was when we got the 225 to when I had to retire, I'd say the culture regarding automation and the training for it changed dramatically in our company. Jobs like that are never complete though, but I suspect it is continuing.
But for example we ceased doing a manual ILS on every check, instead doing approaches with partial automation to tick the "manual" box. That being infinitely more useful. Engine failure on takeoff after DP was presumed to be flown hands off by the automation. We started flying overlay approaches (ie coupling to FMS to fly a VOR or NDB approach down to DA (and through the missed approach)). And lots more.
It was quite interesting to see the slow but inexorable shift in line pilots' attitude from "can't I just fly it manually, it's so much easier" to loving the automation, to getting a bit close to "automation dependancy". And that of course brings up another point which is that optimising training depends amongst other things on where the company is in its cultural transformation from steam to automation, and the provenance of each pilot. So for example converting a crusty old line pilot onto the 225, nearly all the emphasis was on the systems and automation. It was presumed that they had basic flying skills. However doing a type rating on a baby CPL, one had to include lots more manual flying so that the trainee would gain confidence and skill in flying something much bigger than a Hughes 300.
So looking back on the period 2005 to 2012 which was when we got the 225 to when I had to retire, I'd say the culture regarding automation and the training for it changed dramatically in our company. Jobs like that are never complete though, but I suspect it is continuing.
But for example we ceased doing a manual ILS on every check, instead doing approaches with partial automation to tick the "manual" box. That being infinitely more useful. Engine failure on takeoff after DP was presumed to be flown hands off by the automation. We started flying overlay approaches (ie coupling to FMS to fly a VOR or NDB approach down to DA (and through the missed approach)). And lots more.
FED - with you completely, it stops being flying and starts being watching - not the same thing at all
stops being flying and starts being watching - not the same thing at all
I must be lacking in imagination to not grasp the benefit of being able to conduct an auto ILS, as opposed to being able to fly the aircraft
Thread Starter
Join Date: Aug 2001
Location: Cornwall
Age: 75
Posts: 1,307
Likes: 0
Received 0 Likes
on
0 Posts
CRAB et al
It may have escaped your notice but CAT is about flying A to B and delivering the goods as safely as possible with maximum customers satisfaction. It is not meant to be nor can it deliver happy smiley pilots grateful for the 'workout' and opportunity to broaden or polish their skills. If you want to do that then go find a job that requires near constant application of those skills, fire bombing, SAR, police ops, etc. CAT is what it is. The only saving grace is that so far we have no Cat 3 Autoland so at least we get to do that part of the flight and in the offshore world that can be quite stimulating, especially at night and/or in bad weather.
It of course is a shame that nobody seems to recognise that hand flying skills DO require practice as they are easily lost with the passage of time. Quite what the regulators imagine that a generic series of OEI exercises that have remained virtually unchanged for 45 years will do to help I don't know. We need to get the word out to the rest of the world. The current model for pilot training in helicopters is not working as I continue to see pilots who by no stretch of the imagination can be described as truly competent but they continue to make it through the system because the 'tick-box' approach is dysfunctional.
I am fascinated by HC's ability to twist the CAA into agreeing that there is no need for a manual raw data ILS. I wonder how widely known that little wrinkle is? Such enlightenment is most uncharacteristic not to say out of step with the rest of EASA-land. Of course, if we were using an Evidence Based Training model there would be time and opportunity for both manual and 'modified' AP based approaches but don't get me on that. (i'll just give myself a slap for being so 21st century).
G.
It of course is a shame that nobody seems to recognise that hand flying skills DO require practice as they are easily lost with the passage of time. Quite what the regulators imagine that a generic series of OEI exercises that have remained virtually unchanged for 45 years will do to help I don't know. We need to get the word out to the rest of the world. The current model for pilot training in helicopters is not working as I continue to see pilots who by no stretch of the imagination can be described as truly competent but they continue to make it through the system because the 'tick-box' approach is dysfunctional.
I am fascinated by HC's ability to twist the CAA into agreeing that there is no need for a manual raw data ILS. I wonder how widely known that little wrinkle is? Such enlightenment is most uncharacteristic not to say out of step with the rest of EASA-land. Of course, if we were using an Evidence Based Training model there would be time and opportunity for both manual and 'modified' AP based approaches but don't get me on that. (i'll just give myself a slap for being so 21st century).
G.
Thread Starter
Join Date: Aug 2001
Location: Cornwall
Age: 75
Posts: 1,307
Likes: 0
Received 0 Likes
on
0 Posts
Why 'Tick-Box' doesn't work
For months I have been searching for a suitable analogy to explain why the current (tick-Box) approach to License Skill Tests (LSK) and LPC/OPC's is dysfunctional.
Imagine that you are seeking a to qualify as an expert in English Literature. The test (LSK) is to recite verses from various Shakespeare plays as required by your examiner (TRE). The verses are detailed on a test sheet and there are 55 of them in two sections, Comedy & Tragedy, In each section the examiner will select twenty verses and ask you to perform them.
Knowing this you will receive instruction and practice at all 55 verses and chances are that with a bit of hard work the day before the test you can get things together enough to get a pass.
Now to my mind this doesn't even make you an expert in Shakespeare let alone English Literature.
If you can think of a better analogy please let me know for I will use it in place of Shakespeare if it will do the job and explain to those that need to know why we need to change the way we do things.
G
Imagine that you are seeking a to qualify as an expert in English Literature. The test (LSK) is to recite verses from various Shakespeare plays as required by your examiner (TRE). The verses are detailed on a test sheet and there are 55 of them in two sections, Comedy & Tragedy, In each section the examiner will select twenty verses and ask you to perform them.
Knowing this you will receive instruction and practice at all 55 verses and chances are that with a bit of hard work the day before the test you can get things together enough to get a pass.
Now to my mind this doesn't even make you an expert in Shakespeare let alone English Literature.
If you can think of a better analogy please let me know for I will use it in place of Shakespeare if it will do the job and explain to those that need to know why we need to change the way we do things.
G
It may have escaped your notice but CAT is about flying A to B and delivering the goods as safely as possible with maximum customers satisfaction.
The CAT industry seems to want bus drivers and it appears they have trained exactly what they want.
Thread Starter
Join Date: Aug 2001
Location: Cornwall
Age: 75
Posts: 1,307
Likes: 0
Received 0 Likes
on
0 Posts
CRAB
You're not going to get any arguments from me on that score but it is a little harsh to imply that giving our SLF a free bath is a habit. The occasional reality check maybe but don't forget the millions of flight hours we have delivered safely since the 1970's (1960's for a few).
It is a desire not to be complacent that drives us to do it better but you don't have to be a rocket scientist to recognise the huge difference between the dear old S61 and the EC225/AW139/AW189/S92 yet little has changed in the way we go about training and testing save that simulators are IN and have a huge potential to make a (cost effective) difference as part of a revised, updated, modernised, clear thinking, context driven pilot training system.
Keep up the good work but try to be a bit more constructive or understanding of our dilemma if you can find it in your heart to do so.
I read this morning that Cranwell's 11 week terms are being chopped to 8 weeks in the interest of ....... god forbid ...... brevity (not money of course)! The Light Blues will never be the same. They'll be taking away those nice brown leather gloves soon.
G.
It is a desire not to be complacent that drives us to do it better but you don't have to be a rocket scientist to recognise the huge difference between the dear old S61 and the EC225/AW139/AW189/S92 yet little has changed in the way we go about training and testing save that simulators are IN and have a huge potential to make a (cost effective) difference as part of a revised, updated, modernised, clear thinking, context driven pilot training system.
Keep up the good work but try to be a bit more constructive or understanding of our dilemma if you can find it in your heart to do so.
I read this morning that Cranwell's 11 week terms are being chopped to 8 weeks in the interest of ....... god forbid ...... brevity (not money of course)! The Light Blues will never be the same. They'll be taking away those nice brown leather gloves soon.
G.
Join Date: Dec 2006
Location: UK and MALTA
Age: 61
Posts: 1,297
Likes: 0
Received 18 Likes
on
4 Posts
Crab - All incidents of dumping passengers into the sea have been caused directly by lack of automation, lack of automation skills and/or lack of adequate automation policies.
It is for precisely this reason that in offshore land ( no paradox intended) much progress is being made in providing automation, training properly how to use it and mandating when it should be used.
Helicomparitor is correct that in some types a manually flown ILS on a PC is rather pointless as this mode of AP operation is not a possible failure mode. It is better to fly this requirement in "Degraded" mode, whatever that might be and in some types there are numerous degraded modes to practice and test.
In my small world I often see poor HMI skills when these modern AP aircraft are flown in ATT mode such that the pilot does not fully benefit from the stabilisation properties inherent in the system, in my opinion this is a symptom of automation dependency and can be managed by knowledge and training followed by sensible automation policies during line flights. ATT mode is the mode used at the very start and the very end of all flights even when automation has been mandated to maximum application. Therefore the time we spend in ATT mode can be minimal. No surprise then that many of the latest accidents/ incidents have occurred with the AP, or some axis of the AP in ATT mode.
Crab I understand your comments that the pilot is not "flying" but reduced to "watching". We call this "Monitoring" and it is a vital skill that needs strategy and in depth training to ensure safe flight trajectories. However, the pilots require in depth systems based training to understand the deployment phase of AP higher modes in respect to; Indications, Prioritisations and AP Behaviours otherwise not only are these monitoring phases rendered ineffective but lack of understanding may lead to interference in the flight control system by the PF.
Flight by automation requires significant and important skill sets if safe operations are to be assured. Theses skillsets are not divorced from hand from hand flying but they are different.
DB
It is for precisely this reason that in offshore land ( no paradox intended) much progress is being made in providing automation, training properly how to use it and mandating when it should be used.
Helicomparitor is correct that in some types a manually flown ILS on a PC is rather pointless as this mode of AP operation is not a possible failure mode. It is better to fly this requirement in "Degraded" mode, whatever that might be and in some types there are numerous degraded modes to practice and test.
In my small world I often see poor HMI skills when these modern AP aircraft are flown in ATT mode such that the pilot does not fully benefit from the stabilisation properties inherent in the system, in my opinion this is a symptom of automation dependency and can be managed by knowledge and training followed by sensible automation policies during line flights. ATT mode is the mode used at the very start and the very end of all flights even when automation has been mandated to maximum application. Therefore the time we spend in ATT mode can be minimal. No surprise then that many of the latest accidents/ incidents have occurred with the AP, or some axis of the AP in ATT mode.
Crab I understand your comments that the pilot is not "flying" but reduced to "watching". We call this "Monitoring" and it is a vital skill that needs strategy and in depth training to ensure safe flight trajectories. However, the pilots require in depth systems based training to understand the deployment phase of AP higher modes in respect to; Indications, Prioritisations and AP Behaviours otherwise not only are these monitoring phases rendered ineffective but lack of understanding may lead to interference in the flight control system by the PF.
Flight by automation requires significant and important skill sets if safe operations are to be assured. Theses skillsets are not divorced from hand from hand flying but they are different.
DB
Anyway that isn't the argument we used - we had an enlightened FOI who also flew the type and could see the point. Part of which is that having to fly an ILS in ATT mode is not a feasible failure. And he soon realised that partial automation could be a can of worms and thus best trained for thoroughly.
Although tacit acceptance was probably as far as it got! With some of the Mesozoic folk at CAA having retired such flexibility is easier.
Yes I agree and it's a good way to keep / check "proper" manual flying skills. It is just the routine flying of ILSs in ATT mode (basic AP) during LPC etc that I rile against, that being just a mindless continuation of AS332L practices.
Avoid imitations
Join Date: Nov 2000
Location: Wandering the FIR and cyberspace often at highly unsociable times
Posts: 14,573
Received 415 Likes
on
218 Posts
Back in the eighties the UK CAA mandated autopilots for light twins working night ops in the police segment of helicopter aviation.
The accident mentioned about the police helicopter (Scottish EC135) was caused mainly by the pilot not understanding the autopilot and fighting it all the way to the ground, rather than allowing it to help him as intended. Surely that accident was caused by a lack of initial TR training if ever there was one.
It's only in the last few years that I've been checked on my ability to fly a coupled ILS. I think this was because some of the dinosaurs at the CAA didn't understand the systems themselves and somehow saw use of them as "cheating"!
Thankfully, the folks now in the CAA have a more modern mindset and an understanding of modern aircraft and how they need to be flown.
Join Date: Feb 2009
Location: Do I come here often?
Posts: 898
Likes: 0
Received 0 Likes
on
0 Posts
Try this one........
At my home base we have a nice shiny ILS. Due to a problem with the glideslope the UKAIP (and now the charts) says "No auto-coupled approaches to be flown" On the smaller end of our fleet, that basically means hand fly the bugger to minima, at the larger end it means really knowing the VNAV set up, but still be able to do it the old fashioned way, basically as the wildly out of date LPC requires.
However, we are a small company, we fly together often enough to be able to notice slips in skill sets. One area those slips formed a pattern was in very basic tracking and the mental arithmetic around it. Three years ago one of our trainers suggested that when aircraft were positioning, and the pilot had time then drop the upper modes and hand fly the old fashioned way. It has made a noticable change to the IF skills of all the pilots, especially those new to SP/IFR on-shore.
As a result we have seen a change in priorities with a move away from importance of the "set piece" LPC/OPC/IRR to Line Training becoming the more important area.
While an ability to fly the profiles and numbers to the required standard still exists, the ability to manage automation, and then seamlessly cope when it takes a walk has a higher priority here. My last major mechanical snag in an aircraft happened in 1995, since then I have lost count of the times the auto-pilot has decided to let me down in some way, in aircraft varying from the 332 to the 355 and quite a few types in between, from letting go each axis at a time on a 332, to a hard over on a 355, to the whole lot taking a duvet day on an S76 as the glideslope came in at Gatwick.
One exercise we carry out in the sim is to hand fly a precision approach just on the standby instruments. We do it because I once had to do it for real. It is unnassesed as far as final reports go as it isn't required anywhere on any of the CAA/EASA forms. However, it is a massive confidence booster for the pilots, and quite often results in the handing over of dollars from the sim instructors at FSI to the pilot on training as depending on confidence levels a $50 dollar bet is fairly usual.
Another area that I believe is letting pilots down is lack of robust SOP's. I have watched JAR replace CAP 360 and a level of SOP's disappeared, to compare a current EASA OM with a CAP 360 OM reveals a shocking reduction in both guidance and operational instruction to pilots. I certainly believe that a couple of the more recent off-shore accidents in the UK could have been avoided had the crews had the SOP's that existed under CAP 360 rather than JAR.
I have now turned into all the captains I used to get so frustrated with. Bald, greying, reading glasses and banging on about how much better it was in the old days. All I need now is a crew room chat about pension entitlements and golf.
A happy and safe New Year to all RotorHeads.
SND
At my home base we have a nice shiny ILS. Due to a problem with the glideslope the UKAIP (and now the charts) says "No auto-coupled approaches to be flown" On the smaller end of our fleet, that basically means hand fly the bugger to minima, at the larger end it means really knowing the VNAV set up, but still be able to do it the old fashioned way, basically as the wildly out of date LPC requires.
However, we are a small company, we fly together often enough to be able to notice slips in skill sets. One area those slips formed a pattern was in very basic tracking and the mental arithmetic around it. Three years ago one of our trainers suggested that when aircraft were positioning, and the pilot had time then drop the upper modes and hand fly the old fashioned way. It has made a noticable change to the IF skills of all the pilots, especially those new to SP/IFR on-shore.
As a result we have seen a change in priorities with a move away from importance of the "set piece" LPC/OPC/IRR to Line Training becoming the more important area.
While an ability to fly the profiles and numbers to the required standard still exists, the ability to manage automation, and then seamlessly cope when it takes a walk has a higher priority here. My last major mechanical snag in an aircraft happened in 1995, since then I have lost count of the times the auto-pilot has decided to let me down in some way, in aircraft varying from the 332 to the 355 and quite a few types in between, from letting go each axis at a time on a 332, to a hard over on a 355, to the whole lot taking a duvet day on an S76 as the glideslope came in at Gatwick.
One exercise we carry out in the sim is to hand fly a precision approach just on the standby instruments. We do it because I once had to do it for real. It is unnassesed as far as final reports go as it isn't required anywhere on any of the CAA/EASA forms. However, it is a massive confidence booster for the pilots, and quite often results in the handing over of dollars from the sim instructors at FSI to the pilot on training as depending on confidence levels a $50 dollar bet is fairly usual.
Another area that I believe is letting pilots down is lack of robust SOP's. I have watched JAR replace CAP 360 and a level of SOP's disappeared, to compare a current EASA OM with a CAP 360 OM reveals a shocking reduction in both guidance and operational instruction to pilots. I certainly believe that a couple of the more recent off-shore accidents in the UK could have been avoided had the crews had the SOP's that existed under CAP 360 rather than JAR.
I have now turned into all the captains I used to get so frustrated with. Bald, greying, reading glasses and banging on about how much better it was in the old days. All I need now is a crew room chat about pension entitlements and golf.
A happy and safe New Year to all RotorHeads.
SND