SLF/attorney venturing a question here but only for the purpose of gaining better understanding of ongoing discussion.

Reference has been made to the PSM + ICR report. Accessing it (and adding it to lists of reports of various kinds that might be added as background in academic course and program outlines, for public and private international air law) I noticed it had roots in the mid-1990s and was issued in late 1998. Since that time, have there been significant changes in the way Propulsion System Malfunctions present themselves as occurrences to flight crews for response?

Stated differently, is there a significant caveat to be attached to that report relating or referring to advances or changes in cockpit instruments, automation, or procedures? (This is not to suggest the report was lacking in any way when issued or that it is lacking now - only that it seems reasonable to think the interaction between a malfunctioning propulsion system, and the pilots, could be different in some significant ways now, and such differences could in turn be relevant to keep in focus as responses to the accident under discussion are evaluated.)

The problem is the 737 is a design from the 60's and the presentation of a systems malfunction in that aircraft has not significantly changed from the thinking of that era. There have been significant changes but in aircraft like the 787 and A350 that are not hamstrung by a legacy design.

WillowRun 6-3

When we drafted that report we did so with the participation of all stake holder parties including pilots, Operators, aircraft manufacturers, engine manufacturers, simulator manufactuers and FAA regulator types from , propulsion, aircraft and Training directorates ( i may have left somebody out)

We spent most of the time reviewing all the historical data in all the regimes before making suggestion for discussion. We ran into problems with our pet ideas for modified simulators when it became obvious that its cost with all the possible cues to the pilot. was prohibited leading to breaking simulators trying to generate motion and sound cues. In the end the engine guys suggested adding better understanding and cues regarding the engines, but this was dropped in favor of keeping the current training requirements in the reg (predict costs and time) and offering anything beyond the nice to know if you have the time

Lookleft

So if that is the case, why are we now seeing problems with crews dealing with the aircraft? the early 737 models did not fall out of the skies, can it be an over reliance on Auto everything rather than on basic flying skills?

Longtimer I refer to my post at #642. What we are seeing is an aircraft designed where experience and skill were a given to where the limitations of that design are exposed by a lack of skills and experience. Just to be clear, it is a world wide problem and not limited to a few countries.

If you sit in the command seat of a B737 and watch
- a thrust lever split
- the autopilot rolling on aileron
-the trim wheel in motion
-asymmetric engine indications
-asymmetric rudder pedals
and you cant figure something is happening then you shouldn’t be there.

Just for a moment consider the millions and millions of flight hours flown safely over the life of this airframe.

Its not the aircraft.

Longtimer

I totally agree with Lookleft and George Glass. The big difference is the background of the crews flying the aircraft. The 737 was not, before the boom in low cost travel, an entry level aircraft for pilots with limited experience. The first jet I flew had a fairly limited autopilot and no autothrottles, and the SOP was that if we used the A/P in climb or descent we disconnected it to level off, retrimmed and then re-engaged the autopilot. The next two types were better, but as a result of flying them by the time I got to fly the 737 classic it was second nature to have hands on the control column and thrust levers at the one to go call, and all the time below 10,000 feet, so you quickly recognised if something wasn’t going right. This was reinforced by the instructors on the type rating course who were very familiar with one of the historic incidents referenced further up this thread, as it had occurred at the airline where I was training, and we were told about it in detail.
Now it is not uncommon to fly with first officers ( I’m not saying it doesn’t apply to captains as well, I just don’t fly with them), who will let the aircraft do all sorts of things without bothering to guard the controls, even sometimes to the extent of letting it fly the approach with their hands in their laps. Reminding them of SOPs and the reasons that we guard the auto pilot / auto throttle shows they understand it, but the fact that they’re not bothering would indicate that Captains they fly with aren't bothering as well, so the chances are that when they get promoted they will pass this lack of monitoring on to the F/Os they fly with.
But I don’t think it’s about not about not being able to fly, it’s about not being properly trained and mentored because of the much shorter “apprenticeship” which they now serve before getting into a reasonable sized jet compared to what used to happen in the 1960s when the 737 classic was designed.

Nailed it.

Possible distraction scenario
 

Pre-flight aircraft log check not thorough.
In climb out Wx avoidance taking a lot of attention.
Insidious slow thrust reduction goes unnoticed while radar and outside conditions being checked and heading changed.
Further climb clearance given and set.
Possible further distraction if cc rings
Neither pilot monitoring engine panel or control positions.
Too little attention being applied to basic flight conditions.
Suddenly all hell let loose - AP off - sharp roll - loss of height - one engine in idle. Change of attention. Change of priorities. Confusion.

All just an imaginary scenario to demonstrate possible distractions - not intended to accuse or condone the crew in the actual cockpit conditions, which we don’t know.

OK, but I know, that if I had an upside down airplane at 5000 feet all of a sudden I'd most likely be able to recover. Unless you sit on your hands baffled waiting for the autoplilot to sort the mess out. What altitude were this lot at?

Sounds a bit inexcusable to me. We get battered in the sim with upset recovery training 4 times a year. It's doing my head in, but after all these cock ups over the last 10 or 20 years it's no wonder.

Thank god I'm leaving this industry. It's gone to the dogs.

P.s. - I just scrolled up a hundred or so posts and it appears it was 10,600 feet when the loss of control happened. Give me strength. I give up.

Willow ‘… have there been significant changes in the way Propulsion System Malfunctions present themselves as occurrences to flight crews for response?’
Not fundamentally from a technical viewpoint; noting that this accident appears more related to an automatic control system and not the engines themselves.

lomapaseo provides an excellent summary ‘nice to know if you have the time’.
Whilst the review had a common objective - improved safety, the processes of achieving this depended on the stake-holder viewpoint.

The FAA - safety by regulation, was satisfied by having the review and report.
Operators - safety by doing, would meet requirements by having the report for reference (but not mandated), - then placed on the shelf alongside the Wind-shear Training Aid, CFIT, and RTO training aids.
Each agency influenced by a common factor - cost / time benefit; ‘we are safe enough, someone else's problem’. Added to which was the underlying belief that the problem was a training issue, and then because training was already conducted it was good enough.

Although most aircraft systems have evolved, the human contribution appears static. In regulation - a belief that the human element can be regulated; and in operation - that the human can be trained to conform. Both views assuming that the situations and human reactions will be as imagined, as regulated, as trained for.

These issues are not a normal management problem where a solution can be found. Aviation is a ‘complex adaptive system’ (anything which involves humans) without specific solution for perceived ‘issues’.
Perhaps a range of small adjustments seeking small improvement, but not to disturb the current high level of safety, nor introduce new hazards.

Change is required in the manner which the industry views incidents, questions assumptions, and uses the valued human resource.
One approach would require proactive reductions in exposure to hazardous situations, but the inability to foresee every situation still requires reactive ability to manage unknown outcomes. Yet again being unable to foresee or constrain outcome, crews require abilities to mange the uncertainties of the situation, which starts with situation awareness and making sense of these (airmanship, expertise.

Most upsets are caused by pilots. Definitely so on Airbus. In normal law it simply cannot be put into and in alternate law the pitch is same but you got to keep wings level. UPRT training teaches you how to recover from upset. But how about training what causes upset and how not to cause an upset? Because if you don't cause it then it doesn't. Present case is an example. The crew set themselves up for a perfect upset but didn't know how to recover. AF447, QZ8501 no different.

It’s a mindset thing. I am sure I could recover too - did it a few times in service training etc.
But the difference to the sim is that there, you come into it from a scenario of expectation.
These guys (not saying they weren’t at fault...) probably had it in their face, while they were deep in thought about something else. Hesitate and the bird gets away and then, you’ve got to be on the ball and concentrate.
Easier to monitor properly in the first place, not to say safer.

Lookleft got it right in post 642. Airbus dumbed down their new A320 in anticipation of the skill level of the new generation of pilots who would be flying them. This enabled the airlines to employ a lower standard and experience level of pilot at a lower cost. The automatics and protections keep the right side up, pilots just need to know the SOPs and follow the ECAM.

The older generation of Boeing’s still need a real pilot who can actually FLY the aircraft instead of today’s generation of automation managers following the magenta line. The training needed to safely operate older aircraft needs to be adjusted to account for the greater level of monitoring required and degree of intervention needed when things go wrong.

Longtimer

Yes they did, at a truly alarming rate.

Of the 737-200, around 1000 were built, and over 120 of them suffered hull-loss accidents, although not all of them fatal.

Roughly 2000 Classics (like this one) were built, of which around 130 were lost.

The NGs did somewhat better, around 7000 built, and a similar total number of losses. Hopefully in the long run the MAX will be even better.

Mods, I resubmit this (lost) post as being relevant to side discussions on human-automation interaction, and of the perception of industry experience relating to this accident.
Or if for no other reason to provide alternative views of safety thinking.

Longtimer, ‘… can it be an over reliance on Auto everything rather than on basic flying skills …’
First it is necessary to show that flying skill is an issue, vs not understanding the situation;- http://www.pacdeff.com/pdfs/Errors%2...n%20Making.pdf
Second, consider why the vast majority of normal operations are safer - probably with a positive contribution from automation.

Lookleft ‘…lack of skills and experience.’
Experience must be qualified with context - the situation.
The industry has a very high level of experience for normal tasks, everyday flying is very safe.
However, many accidents suggest reduced levels of experience, but to be qualified by - ‘in the situations encountered’.

https://www.newscientist.com/article/mg24933200-300-how-to-be-an-expert-what-does-it-really-take-to-master-your-trade
Normal ops - journeymen and masters
Abnormal events - novice apprentices

What are the chances of finding CVR after all this time? Pretty slim?

They found AF 447 after 2 years - in orders of magnitude deeper water.

It rather depends on how badly they want to find it (and how much money they're willing to spend).

The biggest problem is just who the "they" are with deep pockets. Don't expect many to budget for unknown costs.

We tried to get some agreement with this in ICAO but it got tabled

Aviation is a global industry. The induction requirements, training standards, operating standards, cultural effects on operation of machines that were the product of the rational scientific World is all going to be different in different parts of the world. Even the best Aircraft training is meant for operating an aircraft. But what happens inside the aircraft is also the result of the human that remains outside the aircraft. Otherwise in emergency we would not see people praying instead of applying the procedure. In many incidents the effects are subtle. Only a machine can be made that operates in a certain way any where in the world and if it doesn't then suitable changes can be made in a short time. That's why more and more automation is brought in. Whether for doing normal approach or unreliable speed or Emergency Descent. Machines are trusted more than humans. It's easier to forbid use of outdated technology Aircraft than change global standards of training and skill. If you look at human factors from technology point of view discovery of myriads of human factors is not a friendly but a damning statement against human presence in the cockpit. Because humans cannot be changed but machines can be. Pilots are in the cockpit because their skills are supposed to be better than computers. But if that's not so then they won't be there. World is not going to go back to dynasaur age of manual flying.