Re: Cost of Cockpit Instruments

On Sep 23, 12:55 pm, "Neil Gould" <n...@xxxxxxxxxxxxxxxxx> wrote:
Recently, Le Chaud Lapin <jaibudu...@xxxxxxxxx> posted:

The dentist's XRay machine was either set up incorrectly (perhaps by an
uncertified person?), or operated improperly. As for Shuttles, by and
large they are experimental crafts with many possible points of failure.
The risks are known and willingly undertaken by the participants, just as
are the fliers of experimental aircraft and drivers of experimental
vehicals (racing comes to mind).

The point is that I that think that the "beware the danger of COTS"
attitude is too extreme for the actual risk involved. There is always
some risk.

The general public are not willing to take such risks, nor should they be
subjected to them unknowingly. The recent recall of toys and baby
furniture underscores this last point. If I'm not flying an experimental
plane, I don't want the same risk levels as those that do. And, if I *am*
flying an experimental plane, I'd want good knowledge of what makes it
experimental. In the case that you're creating, it would be uncertifiable
components in an otherwise normal aircraft. Like Blanche, that would make
me rather uncomfortable in some flight conditions.

The public is willing to take such risks. Ever technology advance
that posed some risk to human users initially was tested by guinea
pigs. The world is filled with them. Thy average automobile has
quite a few alone. We still use them because, attitudes change over
time, along with improvements in the technology. Instead of constanly
asking, "What if this fails..what if that fails...", reason takes
over, and people start looking at the likelihood of failure, along
with consequences. Eventually, we take for granted certain things
simply won't happen, even though they do occasionally.

Is it true that if COTS components were used, airplanes with fall from
the sky by the 1000's? 100's? The truth is that we do not know,
becuase few people are doing it. Incremental improvement, using
extremely expensive devices, is the alternative, devices that still
fail occasionally.

So I think the same thing could happen in aviation. There is a trade-
off between pain and pleasure of assumption. There is probably a
point where the cost would be so low from using (well-engineered) COTS
components that the risk of using them is superseded by the value that
they would bring.

How would one know if COTS units were "well-engineered", except by trial
and error? Take, for example, the recent problems with the certified G1000
(see other topics about this). The failures were attributed to some
supplied components in an otherwise "well-engineered" design. Now, IMO, if
the design was all that well-engineered, either those units would have
failed on final inspection (the preferable outcome) or the design would
have tolerated the components and operated properly without problems. Take
the certification process out of the equation, and who knows what one
would get?

Well, at least this goes to show that there is no guarantee. Even
certified componenets might fail.

Yes, trial an error tells quite a bit. There is a safe way to do
trial and error (planned testing with test pilot with safety measures)
and a wrong way (let the Wilsons have virgin flight).

Well, I have an idea of what one would get. As one who has bought and
built many computers over the years (I build them when my requirements are
more stringent than COTS can deliver), I can tell you that all brands of
either full systems or basic components are not of equal quality. The same
would be true for COTS cockpit instruments.

That's always the case for different manufacturers. That would not
bother me at all. If the more expensive components made me feel
safer, I would buy it. If I knew that the cheaper component would
likely material no material effect, I would buy it.

This is true for non-critical components in an aircraft, and might be
true for many critical components.

I seriously doubt it.

Most technologies gradually move toward commoditization and ability to
interconnect, mix and match, etc. The rate at which this happens
often has less to do with technical capability, but more toward
perspective and attitude of systems designers.

-Le Chaud Lapin-

.

Relevant Pages

  • Re: embedded Vs Realtime
    ... >>You need to assess the system over much wider risk areas than just time. ... > irrationally on time to the exclusion of other design issues. ... >>are correct that the failure criteria are for the customer to define. ... More importantly the engineer must understand the process that he is ...
    (comp.arch.embedded)
  • Re: Cost of Cockpit Instruments
    ... have about using COTS components (thanks, ... rigorous certification processes around. ... attitude is too extreme for the actual risk involved. ... supplied components in an otherwise "well-engineered" design. ...
    (rec.aviation.piloting)
  • Re: Saturn V
    ... for vehicle or mission failure depends on the combined risks for the ... The practice of risk analysis is not ... Managing and understanding risks for better safety has a long history ...
    (sci.space.history)
  • Re: AMD planning 45nm 12-Core Istanbul Processor ?
    ... Heard of ISO 9000? ... Your constant insisting on comparability of risk estimation in long term ... So please show me that catastrophic failure of risk estimation of software. ... failure of *risk analysis* software. ...
    (comp.sys.ibm.pc.hardware.chips)
  • Re: Cost of Cockpit Instruments
    ... attitude is too extreme for the actual risk involved. ... over, and people start looking at the likelihood of failure, along ... Sticking some COTS ... Certification means that the instrument will perform at a particular level ...
    (rec.aviation.piloting)