Originally Posted by
RAT 5
The industry answer to a crash is to redesign the systems and include more back-ups, more warnings, more automatics, more..... more SOP's...
Well, to an extent that's understandable because while it's possible in today's data-driven world to provide verifiable improvements by making changes to systems, be they technological or procedural - it's not possible to do the same with the Mark 1
homo sapiens.
so the human doesn't go too near the edge...
But if you think about it, because most flights are automated between departure and approach/landing, is it not the case that the automatics are also capable of going too near the edge - and accidents result more often than not these days from their doing so?
As others allude to, do you not think that considering things in a respect of "human *vs.* automation" sort of sets things up to fail before we even look at the issue? For one thing, the spectre of the "pilotless airliner" has always only ever come from one place - the press - who, as we are well aware, by and large know diddly-squat about aviation or technology.
And then they tell us not to touch anything and let the automatics run the show....
Do they? I was under the impression that even the most gung-ho airlines had a policy of using automation "to the fullest extent possible". Ultimately that extent is at the discretion of the flight crew, is it not? And these are just the most extreme examples.
These are honest questions by the way - if you know differently, please let me know!
Additionally, sometimes the correct course of action *is* to leave the aircraft alone, regardless of the level of automation, because a correctly-trimmed aircraft should maintain a stable flightpath. Indeed, it seems that the broad consensus regarding the correct course of action in the AF447 scenario upon AP disconnect was to not make any immediate control inputs and observe (while of course being prepared to correct if necessary).
and then they tells us that humans, especially the over educated ones that airlines still insist in putting into flight decks, are bad monitors of automatics and how it is the interface of human-automatic gizmos that is the problem.
Again - which "they" are you referring to here? I've read the point on humans being bad monitors of automation, which is why there have been buzzers, lights, stick-shakers et al. to alert the pilot that something is amiss and usually give a clue as to where their attention should be directed.
Interfaces and ergonomics are usually a separate issue from this, and there has been a steady improvement over the decades that cannot be denied. Usually though, it's less often the case that the interface is bad than it is that the system has not been thoroughly understood and trained - not just at the pilot level, but all the way up to ops and management.
You can't have it both ways; you design the human out of the loop and then tell us we are not good at operating outside of the loop.
...
Surely it's on their plate how to reduce pilot error, but I haven't heard of or become aware of any shattering initiatives. Are there any?
For a start, there has never been an intent to "design the human out of the loop", simply to reduce the workload on the human by automating tasks that a machine tends to do better - i.e. usually the rote and repetitive ones.
I had the pleasure of visiting the Hiller Aviation Museum last week during a trip to San Carlos, and I marvelled at the size and complexity of the FE station and overhead panel on their 747-100 flight deck - I already had a ton of respect for the folks at that station and believe me it got kicked up a couple of notches seeing it in the flesh. But it did make me wonder just how much more complexity in terms of managing all that equipment the human brain could handle, and I could see why there was a drive to simplify that with technology as aircraft became even more complex.
So anyway - the intent, no matter what the press said, was not to take the human out of the loop
but to make being "in the loop" require less direct action on the part of the human. This becomes even more useful, in theory at least, given that the skies were only ever going to get more crowded and the workload in terms of flight management was going to continue to increase.
I think the term "pilot error" is beginning to fall out of use - by accident investigators at least - because the evidence has shown that it's too easy to provide a 'catch-all' answer when most incidents and accidents are far more complex in nature. The problem is that such complexity doesn't lend itself well to a TV soundbite or newspaper headline, so you tend to find that if pilot action is mentioned at any point during a briefing or press conference, then the press will report "pilot error", regardless of the contents of the rest of the briefing.
Ultimately there are no easy answers or quick fixes, but speaking for myself as a regular SLF, techie and lifelong aviation devotee I hear and respect your concerns. At the same time though I'm concerned that the adversarial take on technology and pilots does more harm than good. I know, because I've got the notes and documents here, that even the A320 (and by extension her bigger sisters) was designed around the pilot and with input from pilots - and if airline management are using the existence of technology to put the pilot in the role of "last-chance saloon", then that is a flagrant misuse of what the technology was designed to do.
Technology is nothing more than a tool to be used. A complex and powerful tool, certainly - but a tool nonetheless. Your non-crew management will have been sold on the "magic box" idea by the sales guys - same as always - and the onus unfortunately then falls on you to bring that perception to somewhere nearer reality. For what it's worth, if you need a techie in your corner to help you do so one day, I'm more than happy to assist in any way I can.