"Derek Lyons" wrote in message
...
"Greg D. Moore \(Strider\)" wrote:

Of course it doesn't work that way in practice.

Person 1 looks and says, "hmm, looks good, besides, if I'm wrong, #2 will
catch it."
Person 2 looks and says, "well I'm in a hurry, I'm sure #1 caught
anything,
besides, that's why #3 is there."
Person 3 looks and says, "well, I don't want to be the guy that calls out
#1
and #2 for being wrong, so I won't say anything. I'm sure if this was
really a problem, they'd have caught it."

Adding people like this does not necessarily make things any safer and in
fact can make things less safe.

What's the cause and effect here? Having too many people, or poor
implementation/allocation of responsibility, or poor training and
supervision?

I'll be honest here, these types of issues have been of interest to me, but
I don't really keep track of wher I've read stuff.

My simple recollection is it's just a matter of human nature and people not
taking responsibility. i.e. not really a matter of training so much.





The way to properly due this is to focus on failsafe procedures. For one,
an interlock that doesn't permit the clock to go past X time unless the
door
is registered as closed (and I'd be surprised if they don't have this).

On submarines we have a switch that operates a remote display in the
control room indicating whether or not a hatch is shut - but when
shutting the hatch we still had people verify the hatch shut and
locked. The indicator was used to verify that conditions had not
changed, but eyeballs and hands (two pairs of each) were used to
establish that condition.

Which is sort of what I was getting at without being real clear. Note you
mention 2 people, which seem to be about the right number.

And note I'm not arguing against the Mark I eyeball checking important
things like the hatch being open or closed. Merely that adding more people
doesn't necessarily help and ideally they are confirming what the system is
already telling you.



Think about when you fly. The flight attendants are told to check the
doors.
They don't have all 3 or 5 or whatever check every door. And for the main
loading door, they have a procedure to make sure the door is locked and
the
slide enabled.

When's the last time you heard of a door on an airliner not being sealed
properly?

I suspect that if on takeoff an airliner was committed irrevocably to
a course where minor door failure was extremely dangerous, even if an
abort was laid on, and the door absolutely had to perform for as much
as two weeks... They just _might_ treat them differently.

Might, but probably not too much.


That they have had decades of experience and millions of operational
cycles might also have something to do with it.

That is probably by far the bigger factor.





--
Greg Moore
Ask me about lily, an RPI based CMC.