Posts about: "Human Factors" [Posts: 58 Pages: 3]

I raised this on one of the other threads, they are supposed to be isolated systems, and I am sure they are according to electromechanical tradition, but for both human factors and electromechanical reasons, should\x92nt they be off to each side of the cockpit to be considered isolated?

That isn't the human factors question. All it does is tell us that one of the flight deck crew was unaware of the issue as it was initiated. It doesn't tell us which, and it doesn't tell us why.

to add


In my airline (we don\x92t fly the 787 but our engine masters are in a near identical position on our jet) we have had *multiple* incidents of engine masters being manipulated accidentally in flight. This has involved both flight deck and cabin crew. This has meant a re-emphasis on SOPs regarding the centre pedestal but you still routinely see this broken on the line in minor and major ways from time to time

That seems to me to be a very feasible possibility. Noticing, then accidentally hittting the wrong thing in a quick reaction, especially during a possibly bumpy take-off, then only realsing as power starts to die away... A very simple human error possibility - and the switch location makes it horribly easy, especially ofthe switch locking is dubious. Ergonomics and human factors in design? In a very different environment many years ago I was involved in investigating somethimg rather similar (though not deadly). Taught me a lot about operational human factors and our design assumptions...

It's possible, damn it. How dreadful if correct.

That is also a worrying place to put vital switches. Handy place to rest your arm. I have one person I fly with who tends to rest his hand right on the flap lever (light aircraft) if he is not handling the aircraft. I really don't like that. It's bit like my current car, which has the handbrake 'switch'(!) on the central console, behind the gear lever. Just right for accidental operation - and it has been done.

Having studied human factors, that's not what I would call a switch guard. This is what I would call a guard:

Having studied HF, how many flights since the beginning of time have BOTH of those switches been inadvertently moved during takeoff? If that number is 1 or less you were probably better off studying statistics.

Depending on manufactures design principles, or certification standard, then the choice of switch may be determined by the result of normal operation; i.e. a reversible or irreversible outcome.

A 'unlocking' (guarded) switch, as per 787 fuel, is normally associated with normal use, but requiring a higher level of protection angainst inadvertent operation (two actions), but rarely, if at all, considers simultaneous dual operation - 2 switches.

Compare with covered / uncovered push button selections in other aircraft, and those requiring push button 'unlocking'.
A range of interchangeable words, meaning, interpretation; but the root principle is that form follows function.

Beware hindsight bias.

I think we need to be really careful about how we think about human factors.

To all of you who have talked about ‘muscle memory’ or ‘brain farts’ may I refer you all to post No. 183 by MaybeItIs. In it, he writes about the interactions between the cerebellum and the cerebrum – in effect, we have two brains; the cerebrum being the ‘conscious thinking part’ and the cerebellum being the ‘unconscious automatic part’ of our brains.

All of us have done very strange inexplicable things at some time or another in our lives, you only have to read about human factors to know how common such things are – even by highly trained flight crews . I know I have made mistakes like these, and I ask you all to examine your own lives and acknowledge that you may have done the same.

In my own experience, when I was a very new co-pilot, I was involved in a very nasty incident in a VC10 in which we all did inappropriate things, with the best of intentions . We very nearly ended up crashing. After that experience, I began to read books on psychology, brain structure and human behaviour. You will be surprised how easy it is to do the wrong thing and then, afterwards, to wonder why.

Later, I spent much of my career as a training pilot and as a training manager. Despite all the safeguards built in by designers and manufacturers, despite all the safeguards included in our crew procedures, and despite all the training we undergo during our careers, we are all still vulnerable to error, including the so-called ‘sky-gods’.

James Reason wrote about the Swiss Cheese model. Sometimes, one of the holes in the cheese can be our own totally improbable actions. What MaybeItIs has written in his post may help to explain such seemingly inexplicable events.

We all need to have some humility – There but for the grace of God go I.

This is a little over-simplified, but broadly true - and I haven't deleted much of your post for disagreement so much as simply that it was well said enough as you wrote it.

Human factors in the context of behavioural analysis of an event of this sort, is as I was once told, like trying to juggle 18,903 golf balls at the same time, while keeping precise track of which one is which.

We are capable of hugely unpredictable and improbable actions, even as we undertake complex technical tasks with total precision, and what breaks or holds together is often determined by factors which are wholly unrelated to the task in hand - which is of course why pilots (and others) train rigorously, to move essential responses from voluntary to autonomic status in our individual headspaces.

When you add high stress levels and intense anxiety, as probably being faced in this case, trying to apply a form of rationalised 'anyone would do this ' analysis is hardly likely to apply. Even a deliberate act by one of the crew may not represent what we, in our armchairs, assume.

Psychologists, often referred to as 'human factors' analysts/advisors/experts have been part of investigatory teams for decades.

Indeed. The inclusion of a psychologist does not necessarily mean that the human factor being considered relates to mental health. Psychologists study, and might contribute, on issues relating to human-computer interactions, perception, workload, decision-making, cognitive biases etc.

The first thing that occurred to me, after reading the report was, how human error (if that is the case), could be designed out of the system. Two such critical switches right under the engine controls almost looks like asking for trouble, if, for example, a brain fade activated those, instead of a gear up command. One way around that would be to have 2 factor authentication for safety critical functions, an are you sure ?. For example, two switches at the opposite ends of the panel, or similar.

I was editing my response while you replied, I get what you are saying, we are after all deep into figuring out possible rare scenarios, but I don\x92t think it is particularly unlikely(in all of the unlikeliness) that a common failure has a common resolution (I used a misplaced wire strand as an example, cycling pushing it loose).
I also think the time fits, sampling margins taken into accounts.


I don\x92t know what happened, but I think something in this vein of common mode failure in closely placed switches, connectors and looms, though completely dismissed by some, is in the realm of possible, and does not nearly demand the the impossible set up some claims it does.

I talk from experience with large broadcast installs, some made for posterity, some temporary, in which I would dare say the number of connectors looms and switches dwarfs any aircraft, in a fraction of the time, and I have seen some weird s***. At the same time, doesn\x92t nearly has to answer to the same life or death standards, and there is certainly a much larger number of much better paid people involved with the aircraft design.

I agree there is a bit of hamsterwheeling, especially with people thinking they know what was said on the flight deck, and about some of the systems.

But I find there has been some good points made, especially regarding human factors, interaction design, robustness of old and tried designs.

I have seen input from several experts in tangential fields that I find very interesting just in a general sense.

Well, basically, this would be putting a second factor in front of the pedals you have in your car.

Do you want to brake sharply on the freeway (Y/N)?__

I have rebuilt the site that organises this thread by subject here: https://paulross.github.io/pprune-th...171/index.html

It studies 3703 posts (and discards 1043 of them).

Changes:

- Add subjects: "Action slip", "AvHerald", "Human Factors", 'Pilot "Why did you cut off"'.
- Build threads up to July 13, 2025, 10:57:00 (built on July 13, 2025, 12:02:17).

Project is here: https://github.com/paulross/pprune-threads
Raise issues here https://github.com/paulross/pprune-threads/issues or PM me.

You're quoting me out of the context in which you contended a media report with no evidential background could prove a point you were making.

Having added your newer arguments too, you appear to have ignored what I've said since. But to respond to your direct question, my time in an airline cockpit is that which related to participating (as a human factors SME) in an aircraft accident investigation. Not much, but not actually relevant to the point I made relating to your post, or this subsequently - unless you contend that investigators have no place in investigating an accident.

What a relief and about time for PPRUNE commenters to become just a tad more serious. Technical issues bla bla bla for thousands of comments and no mention of the most important factor in modern aviation - human errors. Because that is tabu. The preliminary accidentreport is so clear that we cannot avoid discussing it. I know most here are not professional pilots here as the mods have mentioned and many of those who are - are probably children of the magenta line, but there are also serious aviators here and they know. The fact is that most accidents today are caused by mismanagement of the cockpit and by pilots who do not know how to fly an aircraft outside of the dailyday routines because they were sent from scratch direct into a modern airliner. And it is only going to be worse. We should be brave enough to discuss this - also from the beginning of discussions of an accident. Human factors, cultural factors, authority gradient, startling factors and many other. These matters are the cause for a much higher percentage of accidents than before - because technical issues have become scarse and flying has become much more safe. Keep going...

The captain reportedly took bereavement leave in 2022 after the death of his mother. The only other source for any claims about his medical history were made by a person named Mohan Ranganathan, a former Pilot who had until then publicly refuted claims of Pilot Suicide. He had supposedly "heard [that] from several Air India pilots". I have no reason to believe that random Air India Pilots would be privy to the reasons for any type of leave another pilot was taking. I have good reason to believe that a man like Captain Sabharwal would have been a great topic for baseless gossip.

So everything beyond the bereavement leave is based on statements by one of the many "experts" opining publicly for clout or money based on rumors from people who have no way of knowing what they are talking about. The list of baseless false statements by those experts is getting a bit long. If there was reason to believe Pilot Suicide we would expect Law Enforcement to be involved and the homes of both pilots to have been searched. Either this was done completely evading public notice, which I find hard to believe in such a high profile case, or it was not done at all. I find the latter option the more believable of the two.

There is no need to put any kind of emphasis on the "apparently" part. Obviously we are only looking in from the outside. But there is zero evidence for mental health issues and several indicators for this to not be the case. Taking time off to grief for a parent is a healthy and normal way to deal with a tragic life event like this. It shows that the Captain was both emotionally and financially stable enough to assess and prioritize his personal needs.

Again: Human Error is quite obviously a leading theory right now based on the preliminary report. Ascribing any kind of intent not based on factual information is not a good idea.

From post #918 (number as at the time of this post):

I've desperately grasped at technical rather than human factors as the explanation for this tragedy, but have had to abandon that hope.

For Dr Bru and Engineless:
I think that there is a thread about hacking of airliners somewhere on PPRuNe, might want to take that discussion there.
(Here is one, there are others)

A short response to you both:
1. Note that the article was from 2019
2. I am very doubtful that this occurred, beyond estimating the efficacy of any protections Boeing and Honeywell will have come up with since that article was published.
I doubt that either company sat on their hands after that Black Hat conference.
3. Caveat: yes, hackers never sleep .
= So some kind of sabotage is supposedly done, you think? = (I'll put the speculation into the spoiler)



But I think that you are both grasping at straws, for a variety of reasons, among which are:
1. No evidence to date.
2. Nothing in the report (but then, it's a prelim report)
3. If that kind of thing was going on, I don't think that the report would appear to lean so hard toward the human factors piece.
4. Unless that is part of the deception plan!
5. Yeah, right, we are back to the Hollywood thriller that neither of you have gotten a producer to try and get filmed.
6. Fear
7. Surprise
8. A fanatical devotion to Saint Bernoulli.

Please take any further discussion of this line of inquiry to a thread involving hacking.
Thank you all in advance.
T28B

(As an aside: if you took a look at the debris field, and the fact of the post impact fire, finding any evidence of something like what you are alluding to would be tough unless there's a way to parse EAFR information to detect the intrusion into the system of spurious / outside signals).