RE: [sc] Highest software SIL possible



RE: [sc] Highest software SIL possible

From: Martyn Thomas <martyn_at_xxxxxx>
Date: Thu, 1 Mar 2007 10:56:42 -0000
Message-ID: <IJEALLAMCAJDABDBACKLMELLCPAA.martyn@xxxxxx>
I think the quote from Mike Ellims (below) mischaracterises the discussion.
I have not seen anyone arguing that mechanical systems are intrinsically
safe; but that mechanical systems are easier to analyse because they have
fewer discrete states that have to be considered separately.

Mike's examples illustrate that, even with mechanical systems, the operating
environment may present challenges that the designers had not fully taken
into account.

The real world is just as complex an environment for digital systems as it
is for mechanical systems - but we have the added problem of having to
consider the interaction of real-world complexity with a hugely greater
number of discrete system states.

That's why sampling the interactions, by testing the system, only gives a
very limited view of the system's safety, or reliability.

Regards

Martyn Thomas

-----Original Message-----
>
> I have to admit that I find the faith placed in the safety of
> mechanical
> systems quite touching, if possibly somewhat misplaced (drive my very
> old and fully mechanical car sometime).
>
Received on Thu 01 Mar 2007 - 10:56:06 GMT