[SystemSafety] Comparing reliability predictions with reality

[Prof. Dr. Peter Bernard Ladkin]

On 2025-02-24 16:20 , Derek M Jones wrote:
> The fact that some companies have produced reliable software
> provides evidence that it can be done (at some cost).  If these
> companies were to publish details of the processes they use
> and associated reliability, then we would have some evidence.

OK. So by "evidence" you mean publicly available evidence. I suggest it is better to use those three 
words when that is what you mean.

I have a colleague whose company produces, amongst other things, software-based sensors for 
industrial processes (chemical plants and so forth). In well over a decade of use, they have never 
had a software error manifest in any piece of this one particular kit. I'd say that's evidence. But 
it is also publicly available evidence.

>
>> And of course it is not the only example. Modern civil aerospace is full of very-highly-reliable 
>> software-based kit, developed according to evolutionary company practices following DO-178C and 
>> DO-333 (or EUROCAE ED-12C and ED-216). Evidence, again, in the operational histories of all this 
>> kit.
>
> I think you need to distinguish between claims made and
> evidence presented.
>
Sure there can be discrepancies between claims made on the basis of operational history, and public 
data as to that history, as with the example I gave above. And as with almost every civil aerospace 
example. That does not mean such claims are not based on evidence.

PBL

Prof. Dr. Peter Bernard Ladkin
Causalis Limited/Causalis IngenieurGmbH, Bielefeld, Germany
Tel: +49 (0)521 3 29 31 00