Originally Posted by
langleybaston
Please for the ignorant may we have a definition of a Safety Case?
I doubt you will find an example of an aviation safety case in the public domain, but the [uk] requirement for them comes from standards like JSP-430 and Def Stan 00-56. The non-military HSE document R2P2 ("Reducing Risks and Protecting People") also has stuff to say about them.
As for a definition, 00-56 says:
"A Safety Case is a structured argument, supported by a body of evidence, that provides a compelling, comprehensible and valid case that a system is safe for a given application in a given environment".
That's about as succinct as it gets - the JSP version is much wordier and less clear as a result.
There is an inherent problem with safety cases - they are a verbal "argument". This means that for any complex system (anything much more complex than a penknife, actually) there are lots of words over lots of pages which need an expert to grasp, comprehend and critique. That was one of the issues in the nimrod case - the post-holder who contracted it knew exactly what should have been in it, but when it was delivered a few years later the current post-holder wasn't a safety specialist and could only assess it by weight ("thud-factor" as Haddon Cave called it).
To construct a meaningful safety case for a fast jet needs many thousands of hours of specialist engineering. Even just drafting it in any meaningful way involves a special symbolic language called "goal-structured notation" which is then fleshed out in text. Once it's created someone else then has to try to understand it to critique or approve it - arguably the only way to do this is to deconstruct it back to the symbolic form, and that takes another few thousand hours of specialist engineering effort. In the real world these skilled resources rarely exist in the quantities needed.
And of course for every slightest change in the system or its operating context the whole thing must be revisited to see if it invalidates (or just weakens) the arguments as presented.
Something like a bang-seat can't really have a safety case in its own right - it has it's contribution to the overall fast-jet training system safety case. Why? Well the acceptability of some safety mitigations can often depend on how and where it is used, stored & maintained.If the seat is sitting in stores it may be acceptable to mitigate a wear hazard under the seat pan by daily inspections, but this probably wouldn't be a viable argument when it's installed.
I know nothing about the Hawk case; the above is only a generic description of the items in question, in my personal opinion, as an engineer who happens to hold a safety engineering qualification.
E&OE,
YMMV,
PDR