GA200 ~ Just reread the report.

A question I raised on the PIA 8303 thread ~ the lack of involvement by the Aerodrome Controller.

A recent UK AAIB report on an unstable approach at Bristol revealed that the Aerodrome Controller instructed the B737 to Go Around as he assessed that the ‘profile’ on final was unsafe.

No mention in the GA200 report of ATC saying anything to prevent the impending disaster, even though eye witnesses on page 41 comments on the excessive speed [210kts]. The final hole in the ‘Swiss Cheese’?

Is / was the role of ATC at a legal variance in Indonesia
or is it a ‘Cultural’ matter that you don’t question a pilot even when you know that it “doesn’t look right”.

The ATC involvement with IX 1344 might also be revealing....

(my bold)
A general point, since we don't yet know what went wrong in this crash:-

I have often said that taking control needs to be actually trained. It is written in the books that the F/O is supposed to challenge the Captain three times and then take control. But is this ever actually taught or practised. Flying a plane or landing a plane is not left to a written description in the FCOM. We spend weeks having it explained, being examined on it, having our understanding checked and our ability to do it by practising it in a simulator, and finally having to pass a 4 hour LPC/OPC exam in a FFS.

Yet one of the most important safety barriers to accidents - taking control from an errant Captain - is reduced to merely a few lines written in the books somewhere and rarely referred to, and never actually practised for real or tested, (in my 20yr commercial flying experience with 4 UK airlines).

Taking control from a Captain is a big deal - especially if they are also management or very senior. It therefore needs to be taught and play-acted. By everybody. F/Os need to be coached and practise saying the words to a Captain in a 'real' situation. Captains need to experience being taken over from. That way, all members on a flight deck will have been through that process. The F/Os will have more confidence to take control because they will have practised doing so. Captains will have experienced the feeling of losing face and having their authority taken from them. So if it ever has to happen for real, it won't be the first time for anyone.

This will improve F/Os confidence and make Captains better Captains because they will know that if they stray and don't correct errant decisions, they will have their control taken away.

It's not limited to some cultures either. I have sat on (UK) flight decks thinking this guy is doing something questionable. Should I take over or will it be alright.

In present case it's a bit early too say anything because it's not yet confirmed whether the aircraft overran during landing or during attempted late GA. Culture alone is not the cause. How well the FO was trained has a lot to do with the confidence he exudes in the execution of his duties. In the same airline in the Mangalore accident the FO who himself was upgrading to Capt(not in that flight) didn't provide any input about badly out of profile descent and approach till the RW was right under them but couldn't enforce the GA. The airline did make some changes to their CRM training because one year later a repeat of similar act after touchdown was overridden by the FO. In Australia in an A320 incident the FO was PF had more experience on type and when he messed up an approach the Capt due to his lesser experience on type felt diffident to interfere that led to a GA. In the present case FO could have interfered if approach was unstable in any parameter or the float was appreciable. But that could only happen if FLD was examined critically taking into consideration the margin if any they had with TW and worse than wet condition on a RW that would not permit overrun.

Readers may recall the crash of Silk Air Flight 185 in 1997. See: https://en.wikipedia.org/wiki/SilkAir_Flight_185

On an earlier flight some weeks before that accident the same captain was trying to lose height on short final by applying full aileron in both directions to activate the spoilers. The F/O who was not culturally bound called for the captain to go around as the 737 was badly unstable. The gyrations of the aircraft were alarming but despite this the captain was intent on landing.

When the captain ignored all the usual SOP warnings, as a last resort the F/O pushed the throttles open under the captains hand leaving the captain with no choice but to go-around. The actions by the F/O saved a sure accident. In this case the F/O did not try to take over control from the captain as he knew this would only exacerbate an already dangerous situation.


So true. In the following instance both students were undergoing 737 type rating training in the simulator. The PM was a powerfully built chap from SE Asia with an aggressive personality.
The briefing included a multiple bird strike immediately after airborne causing the failure of one engine. The aim being to return to land at the same airport.

The engine failure was well handled by the PF. After all applicable checklist items had been completed the PF joined the circuit. A fire warning then occurred on the remaining engine. The aim of that exercise was to continue with the planned landing and deal with the fire warning once on the ground, In other words how to deal with priorities.

The PM immediately called for the engine with the fire warning to be shut down; which of course would have been disastrous. The PF wisely disagreed for obvious reasons. This is where culture came to play. The PM flourished the QRH in front of the PF who already had his hands full, demanding the PF follow its instructions which was to close down the engine. Despite protestations by the PF the PM closed down the live engine and fired the bottle. The result was predictable. Afterwards the PM was suitably debriefed by the instructor in no uncertain terms.

Thanks for pm, parkfell. Perhaps we are slowly moving in the right direction.

In the SIM we practise pilot incapacitation, which is easy because the other pilot does not interfere or argue. CAA should add an exercise to the D3 LPC/OPC form where, instead of passing out, Cap is secretly told to deliberately fly over limits or recklessly to see what F/O does. This is where F/O should question Cap and eventually take control, and would be a very useful exercise - in fact much better than doing it in a CRM classroom.

In one airline I was with we had an excitable checkie with a short fuse. The simulator session included captain incapacitation at some point in the session. Just before one takeoff the checkie whispered to the captain to be incapacitated during the takeoff run. The F/O was unaware of the plan,

Half way down the runway, the captain suddenly gasped loudly and fell over the controls with admirable acting. The F/O was shocked and genuinely alarmed and reached over to try and help the captain whom he thought was having a real "event." With no one on the controls the aircraft veered off the runway at full power heading towards the control tower. The checkie immediately froze the simulator. The captain miraculously came alive while the F/O wondered what the hell was going on as he thought the captain was a goner real time.

The checke got stuck into the F/O for not taking over control from the simulated incapacitated captain. The F/O protested and said he thought the captain had suffered a heart attack for real. The checkie then ticked the FAIL box on the F/O's records. Naturally this upset the F/O who was a former RAF pilot with real life operational experience during the Cold War days.

After the tea break the F/O became PF. Keep in mind in the company training syllabus incapacitation only ever applied to the captain. Half way down the take off run the F/O without warning suddenly collapsed with a loud groan at the same time stiffening one leg to full rudder.

This was not a briefed exercise and the captain was immediately alarmed and leaned over the pedestal to try and comfort the F/O who by now was making anguished cries. Meanwhile under the influence of full rudder the aircraft headed towards the control tower. The checkie clicked what had happened and froze the simulator and berated the F/O calling him a smart-arse for doing something not in the syllabus. The F/O showed admirable restraint in not snapping back at his unfair treatment.

Interestingly the checkie chose not to criticise the captain for not taking over control when the F/O had his "event.". Rank obviously has its privileges..

Wise thoughts and insightful stories above. Need to point out all of them for the plan B. Still appreciated, really.

Back to plan A? For the PIC to be lucid enough and discontinue on his/her own? That is what I had in mind when mentioning the responsibilities of the investigation body.

As mentioned above, I personally see quite some sense in a punitive rule for landing de-stabilized.

E. g. for continuously hot or high profiles, assuming 1000 ft OM-A policy:

• g/a before 1000 ft - ignored
• descend through 1000 ft: FDM event
• descend through 500: SOP violation that goes onto the file, this can be avoided by a g/a below 1000 but before 500.
• ​​​​​​land from unstable at 500: demotion, this can be avoided by a g/a below 500 but before the touchdown
• landing past the markers or Vapp+25 - termination

Surely it can be tuned, possibly adding the DA as a gate. These days pilots are properly trained to take the landing decision (evaluation) at that moment.

Suggestions for other methods greatly welcome.

In 2019 Indian DGCA suspended 40 pilot's licenses for overruns and other incidents for a period of 6 to 12 months.

'In 2019 Indian DGCA suspended 40 pilot's licenses for overruns and other incidents for a period of 6 to 12 months.'

That action does not appear to have improved safety in 2020; what next ?

Its difficult to change the human condition, but you can influence behaviour by changing the place of work (James Reason); place of work includes the operational situation, runways, overrun areas, weather and runway condition reporting, …

I can only speak for the operator I work for but continuing an unstable approach in IFR below 1000’ is a career altering experience. One incident will get you tea and bickies and a please explain, or worse if you didn’t self report. A second will get you busted back to the right hand seat.
QARs have been around for years. There really are no excuses. Repeat offences in a particular environment represent a comprehensive failure of safety culture and regulatory oversight.
As result of personal experience there some countries in which I simply will not fly in domestically.

From 500ft if the flight path and thrust is maintained the speed should not change. The problem is if at 1000ft if speed is+15 and GS half a dot above it fits into stable approach but can end up in overrun if not handled correctly.

I don't know how true this is but understand that a few years ago China Airlines (Taiwan) captains were continually getting the blame for QAR reports of unstable approaches into Taipei when the PF's were inexperienced cadet F/O's. The QAR's were used as a Big Brother punishment tool. That trend changed when the captains refused to give their F/O's any landings.

I think you’ve got to have some disciplinary bottom lines for reckless behaviour, but at the same time you need some “carrot” as well as some “stick”. Actively praising/rewarding discontinuing an approach that is getting out of hand is very important important part of safety culture, IMO. This should be emphasised in training, checking, line operations and FOQA.

When the chimp is sitting on your shoulder giving directions, if it understands that one path leads to a reward and the other to a smack, it might lean more towards the reward.

Agreed. The aircraft state needs to be continuously assessed until you stop/vacate. I would differ in that not everything undesirable happens because of poor technique; gusts/shears, etc. at critical moments are part of normal operation but you do have to have a clear idea about what you can/can’t accept and what you’re going to do if it gets out of limits.

Quote:

---

.......Actively praising/rewarding discontinuing an approach that is getting out of hand is very important important part of safety culture, IMO. This should be emphasised in training, checking, line operations and FOQA.

---

Reminds me of the Lenny Henry sketch, where he says that traffic police always pull you over if you have done something wrong, but never if you have done something well...."the way you took that corner, man, that was wicked...." ! :)


Perhaps one snag is the 1000'/500' thing.

Pilots all know they should be stable by 1000', but they also know that if they are nearly stable at 1000' there is leeway to continue to 500' (technically only if positively correcting and visual, but definitions vary).

This tells a pilot that if they are near enough at the correct speed and correcting, they can continue through 1000' - but each pilot will have their own personal definition of 'near enough', and 'correcting'. Having continued through 1000' too fast, when they get to 500', the subconscious thinks, 'well I am still near enough, and anyway, look, there is the runway; plenty of length, so we might as well land, and I won't have to write a report or waste 15mins more fuel and the chief pilot will never know'.

Perhaps the 1000' criteria needs to be tightened, so as not to allow any over-speed at that gate?

I am wondering how FDM software would automatically pick up a long landing ? I guess you could set a flag for the time from closing the thrust levers to WoW? Or perhaps from a RA of, say, 50' to WoW.

FDM systems vary considerably depending upon their age & complexity (cost!) - and also the capabilities (mainly amount & rate of data capture from the recorder). A simple one will indeed measure time from point A to point B. It can also then work with the g/speed to reach a fairly accurate t/down point. A very modern sophisticated system will use GPS to ascertain the exact t/down point (very accurately) and if then associated with the runway used (& airport) can build up an extremely useful database about long landings vs airports. If you factor in the wind as well it can be a really, really useful safety tool - just depends upon how much the airline really wants such a system......

George, #333, as per the issue above #332, the QAR view in isolation might at best only provide a small improvement, but never a guarantee.

Again from James Reason ' even the very best people make the worst mistakes ', which has to be considered in context - the actual situations; accidents generally have several contributing factors, which come together at a particular time. '… the chance and largely unforeseeable concatenation of many different causal factors, none of them sufficient or even especially remarkable by themselves, but each necessary to bring about the outcome.'

The need of defence in depth.

Operators should reduce landing distance available by the amount of an unsuitable overrun area.

Airports should inspect runways more frequently in monsoon conditions; have time delay landing restrictions for poorly drained surfaces, check and publish rubber contamination (slippery when wet).

National authorities should require grooved runways, for those airfields without then reduce landing distance available to compensate for reduced braking / drainage.

Governments should allocate funding for improvements, act to instigate the above.

"Should, could, must, but who does"

The recruitment criteria for most Western airlines is strongly biased towards not just technical competence but also ability to operate without direct supervision, acceptance of ,and compliance with , a rules based operational environment and high personal standards. Plus an ability to work with and listen to others. The ability to send a crew off on a multi day mission in charge of a multi million dollar asset is an achievement that I still am awe struck by. How you replicate that is the problem.

vilas, #335 '… if at 1000ft if speed is +15 and GS half a dot above it fits into stable approach but can end up in overrun if not handled correctly.' :ok:

Landing distance required must be calculated (considered) for foreseeable error situations, planned for situations - approach speed increase for wind, or 'near enough'.
The approach speed must be adjusted during final approach to achieve the expected landing performance, but that would be 'unstable' - the clash of reality with procedural requirement, requiring a balanced judgement influenced by rules, interpretation, safety culture.

Uplinker, GizmO, re touch down position from FDR.
This is relatively easy to measure, even the most basic recordings provides a reasonable indication.
A key issue is the use of the data; view the human as the threat, or learn from aggregated data about influencing conditions, specific runways, airport restrictions.

I recall one analysis, many years ago, which identified a correlation between the point of touchdown and runway length; the study ranged many airports in a large operational system, and aircraft types according to long haul / short haul.
Thus with all other things being equal, if pilots can touchdown within a suitable distance on a short runway, then why not on a longer runway; a lengthy and repetitive discussion - because its human.

The operator published the findings for awareness and recurrent training comparison - the training results on all runway lengths was acceptable - we can do it, we can choose to do it, but not always.

Yes, it does not have to be a big deal, we only have to change pilot's attitude and judgement - a bit like changing drivers' attitudes to wearing seat belts or drink-driving.

Most pilots already accept the real need to be stable on approach - the days of 240kts to 4dme etc are well behind us, Being unstable should lead to a go-around without punitive consequences, and this message is slowly getting through.

On landing, none of us want to slam the aircraft in every time, but a long float is a bad habit. (I am sometimes guilty). Chief pilots could mandate that a long float should lead to a baulked landing. A baulked landing should be seen as a lesson learned and a positive contribution to safety, in the same way that other go-arounds are seen by most enlightened chief pilots.

If it was freely publicised that FDM would be used to flag touchdowns outside the touchdown zone, standards would gradually improve.

Another way to introduce incentives to change behaviour would be to publish league tables, (or even offer modest bonus payments), for stable approaches, touchdowns within the touchdown zone, and, say, three manually flown approaches every six months. Pilots' standards would improve !

League tables - of any sort - are a highly contentious subject in just about every airline. Even if they are, nominally, in the name of safety. Someone will always be "put out" or feel unfairly treated and that will ultimately lead to trouble.........

Uplinker,

An interesting idea, but when incentivising behaviour, beware of the unintended consequences which can sometimes be totally unexpected.

Definitely. I remember when they first brought in excess fuel carriage tables in my company. You had two groups who tried to get to opposite ends of the table, treating it like a competition, while the rest of us just took what we deemed appropriate on the day. The dangerous bit, IMO, was when some individuals vying for the top spot started returning negative excesses - they were using get-outs in our fuel policy designed to be used on rare occasions every time they flew...

As the CEO of a failing multinational (I'm not), I always get​​ bonus payments for, erm, doing my job. And, as it happens, invariably when I fail to do my job as well. I thought we had this bonus payment sector tied down, but it seems there are whispers that pilots should have a slice of the pie too. Not on my watch, no siree.

Seriously though, where does this stop? Should our cloakroom attendant be rewarded for not losing any coats? Or draping them properly on a hanger? If you find it impossible to track every coat you get, or can't hang them properly, you can have some more training. If that doesn't work, perhaps cloakrooms aren't for you.

"Stable approaches, touchdowns within the touchdown zone, and, say, three manually flown approaches every six months" shouldn't these be minimum requirements? If this isn't happening, or it's not being measured and recorded, surely something has gone seriously wrong?

@rb14, yes, I know it sounds counterintuitive, but humans need incentives. In the 1970's we should have all worn seatbelts, but we didn't. Now we do, because it became socially unacceptable not to.

Pilots should do everything right, but quite clearly they don't. You could use a stick and prosecute all errors. Or you could use a carrot and reward, or otherwise incentivise safe behaviour.

League tables for fuel is frankly xxxxxx dangerous. I have heard of really stupid and dangerous behaviour, such as one moron who flew back to a partially SNOWTAM ed UK with just 300kg extra fuel..............:mad:

So, never for fuel, I agree.

But if we had league tables or part of our pay was, say, £2.50 for every approach we flew that was fully stable at 1000', and a touchdown within the touchdown zone, it might start to change attitudes to risk taking and provide an incentive to encourage safe behaviour; it could work.

And how do you measure safe behaviour in a fair and objective way?

FDR readouts? CVR readouts? Cockpit videos? Fellow cockpit crew member reports? ATC reports? Cabin crew reports? Check rides? Passenger satisfaction?

No thank you.

Uplinker

It may start a trend to duck under. Financial incentives may not be what they appear. That's why salaries are rarely have one fixed figure per month. Basic salary can be reduced and then made up by giving a little for this and a little for that. Basically it's motivation to do a job well. It could be financial or appreciation of sincere effort some other way.

On the other hand of spectrum, you have people taking more than half an hour of extra fuel to an almost deserted 2-runway airport in severe CAVOK, just because 15 years ago they had to fly one turn in a hold one day. There's extreme cases on both sides.

If you're not stable at your gate and have not touched down in the touchdown zone or at the point that you should have for a safe landing (those two might not co-incide), you should go-around. We shouldn't reward people for doing what's expected of them.

What's next, a fiver for lifting your bottom off the seat and doing a walkaround when it's raining?

Where this might come into its own would be on training days, where the ‘spread’ of touchdowns were shown.
The analogy are the graphics shown as to where the ball pitches on the wicket during Test Matches etc. Analysed into various ‘zones’ as percentages and groupings. [Apologies to our American friends talking about a cricket match that can last 5 days]

AND / OR

Anonymous monthly publications of the ‘spread’, together with CP tea and biscuits, or sometimes without, as necessary.
Perhaps even individual performances sent confidentially. The difficulty then comes as to the impact caused by ‘inexperienced’ FOs learning their trade.

On balance UPLINKER’s idea is worth a closer look.

Some interesting reactions !

How can we incentivise pilots not to continue unstable approaches or long floats?

"Well pilots are paid to do their jobs properly". Yes, of course, but we have recently had two fatal crashes where unstable approaches, and in one case, a very long float ended in disaster. And this is in today's modern piloting, with CRM and SEPs etc. I am merely thinking aloud about possible ways chief pilots might push the balance towards a safer mindset.

Being both properly stable at 1000' and touching within the touchdown zone, would apparently be easy to program into FDM, and some way of incentivising these twin goals might start to reinforce the message to all?

Be wary of league tables; they can be a two-edged sword. Many years ago, a colleague on an annual assessment. "fewer days off sick; that's good" then "You tend to carry more fuel than most". His reply; "Yep, more fuel, less stress, fewer days off sick"

I was under impression most sensible airlines already have that in their FDM, and for me personally, not having to go and see the chief pilot for tea with no biscuits, is a great incentive to not bust the gate or land long.

Whilst the side discussion on monitoring, margins, etc, covers many aspects, how would this be applied to the accident situation. What additional distance would be expected to be added, and what is the origin of the baseline data.

Most regulatory agencies and operators require an additional distance safety margin to over baseline landing data; e.g. using OLD the minimum recommended addition is 15% (FAA, EASA), in some manuals this is listed as FOLD Factored Operational Distances (Airbus) - a significantly different concept from 'actual' distances or AFM distances with other factors.

What landing distance safety margin keeps you out of the office ?

What is this operators policy, national policy, is it recommended or mandated ?

So was I...Long landing/excessive time in the flare/Float, whatever you want to call it, is certainly analysed in our FDM and occasionally the more extreme events receive some suitably redacted publicity..

While "real men don't go around" or loss of face is ingrained in the brains of some culturally afflicted pilots, clearly this sort of insane mindset is here to stay. All the lectures on CRM, TEM or whatever is the latest ICAO buzzword (invariably paid lip service by the company), won't change a thing.

We have to live with it because history has shown us these sort of accident contributory causes will continue to feature in accident reports. Unfortunately the strong possibility of a crash being caused by a cultural bent is never mentioned in accident reports possibly due to a subtle application of political correctness. Also, the nature of this type of mindset usually means it cannot be proved beyond doubt.

Flight safety technical advances may reduce the number of accidents but they cannot erase 2000 years of culture.

'Long landing/excessive time in the flare/Float, … is certainly analysed in our FDM' # 357

That is an important aspect, but not the only one. Of greater importance is where the aircraft stopped - sufficiently slow, vs where it could have stopped; from this a comparison between the planned safety margin and actual available in the conditions, a check both on human contribution and the reported runway conditions.

Re several contributing factors - how many in this accident ?
The info-graphic on slide 20, 'Overrun Characteristics' https://www.icao.int/EURNAT/Other%20...FR%20PPT01.pdf (slow loading file) shows the issues, indicating the relative frequency or overall significance.

* GRF - Global Reporting Format - for runway condition; to be introduced this year together with revised landing distance performance which can be related to the reported conditions.
Also:- https://www.icao.int/EURNAT/Other%20...hp%20PPT06.pdf
and
https://www.icao.int/Meetings/grf201...-%20Airbus.pdf

a lot of thread creep here. A great discussion for elsewhere!

Anyone have any updates on the subject of this thread?
Any developments in the investigation?

"On Aug 25th 2020 the NTSB announced, they have appointed an accredited representative to assist the AAIB of India with the investigation according to ICAO Annex 13."

India Express B738 at Kozhikode on Aug 7th 2020, overran runway and fell into valley

Before being misunderstood, misquoted, called an idiot etc... Calculated, briefed and carried out within limits there should be (Is?) no issue with a tailwind landing. Fact... Moving on...

... there appears to be a relatively high percentage of over runs during this period of massively reduced traffic in the global skies. I am currently camping out waiting for the aircraft to arrive at Curitiba in the South south of Brazil, the terminal is tucked up at the beginning of runway 33. the tendency here during this period of fewer movements is to take off on 33 and land on 15 to reduce taxiing times. Does anybody else play the following thought game: that we are spending more time these days performing tailwind landing (takeoffs too) to save on taxi time as it is much more likely for reciprocal operations to be granted and as such, there is still the same percentage of pilots willing to push the limits which is contributing to this?

or am I overthonking this and talking utter muffjank?

overthonking or utter muffjank
 

In case I am mistaken, the issue with tailwind landings involves flare judgement at higher ground speeds. The aircraft feels the same - same airspeed, but visual feed back differs due to the 'unusual' ground and vertical speeds, thus can be misjudged, such that the variability in flare time and distance are much larger (certification speed allowance is 150% wind speed), i.e. the aircraft is more likely to float - a long landing.
Many pilots will have operated in moderate tailwinds, the industry seems to depend on the capability re changing runways or not. However, with increasing tailwinds, beyond normal ATC changeover limits, pilots lack experience and practice.

The risks in tailwinds above 10 kts increase rapidly.

Many years ago the recommended maximum was 10 kts (ICAO), this has been systematically allowed to drift to up to 15 kts. e.g. aircraft had 10 kt tail limit, customer request for 15 kts (night noise abatement) was reluctantly agreed providing that the AFM specifically limited to that operator and airport, and with additional training; 10 yrs later the AFM clearance for 15 kts was standard without limits or warnings - we forget, then fail to relearn hard lessons from accidents.

Risk during landing has increased - drift into failure - for convenience, noise, weather (particularly if wet or gusting wind), without due regard to mitigations - runway grooving, overrun area, runway condition reporting, accuracy / interpretation of landing data.

Neither 'overthonking or utter muffjank'; the reality of operations, which continually surprise us.

"What we observe is not nature herself, but nature exposed to our method of questioning."

FlyingStone

Which ‘extreme’ would you rather have when flying as a passenger? I know which one I’d prefer.