[SystemSafety] Units for Functional Safety variables ? - A Friday afternoon question on a Wednesday.

[Peter Bishop]

In measurement theory
- a nominal scale are just labels
- an ordinal scale has a rank order (1st, 2nd, 3rd, etc., or A,B,C,D ) - 
but you cannot do maths on the rank values
- a ratio scale is used for expressing physical quantities relative to 
some base unit (where you can do maths)

HFT could be represented on an ordinal scale, as there is implied 
ordering, i.e. 2 better than 1 {as more component failures can be 
tolerated) - but we do not know how much better.



Andrew Rae wrote:
> Jon,
> Not everything has a dimension. (Although I suppose even dimensionless 
> constants could be considered dimensionless ratios).
> Hardware Fault Tolerance (assuming you mean in the 61508 sense or 
> equivalent) is of the same "type" as a Safety Integrity Level - they are 
> assigned labels.
> This is indicated in the fact that assignment and manipulation is 
> according to arbitrary rules, not normal arithmetic.
> 
> The other way of looking at it, if you reject the idea that HFT is a 
> "label", is that it is a count. The dimension of a count is whatever you 
> are counting, in this case
> "levels of redundancy". This lets you perform dimension maths - you can 
> weigh your system, and come up with "Levels of redundancy per kilogram", 
> or cost your system and determine
> "$$ per level of redundancy".
> 
>  
> 
> My system safety podcast: http://disastercast.co.uk
> My phone number: +44 (0) 7783 446 814
> University of York disclaimer: 
> http://www.york.ac.uk/docs/disclaimer/email.htm
> 
> 
> On 2 October 2013 15:49, Jon Hind <jon at jonhind.co.uk 
> <mailto:jon at jonhind.co.uk>> wrote:
> 
>     I posted this (only partially) frivolous question on out internal
>     system, without any reply.
> 
>     So trying here for some erudition...
> 
>     What dimension and or units does Hardware Fault Tolerance (HFT) have ?
>     - A Friday Poser on Functional Safety Terminology (FST).
> 
>     I remember being taught at university in the last century that
>     everything has a dimension in terms of Length (L) , Mass (M) and or
>     Time (T), or is a ratio (factor). A good check of the validity of an
>     equation is that it must balance dimensionally.
> 
>     Availability Common Cause Factor (β, CCF) and Safe Failure Fractions
>     (SFF) are dimensionless ratios, failures rates (λ) are T^-1, Proof
>     Test intervals are T etc.
> 
>     And HFT ? HFT is an integer - In our (oil & gas process) world, 0 1 or
>     2 ; is it part of a case statement, or heaven forbid a pointer like in
>     the C language ?
> 
>     I don't want to call it a factor as we already have that in CCF and SFF.
> 
>     Answers on a postcard please....
> 
>     Jon Hind
>     jon at jonhind.co.uk <mailto:jon at jonhind.co.uk>
>     _______________________________________________
>     The System Safety Mailing List
>     systemsafety at TechFak.Uni-Bielefeld.DE
>     <mailto:systemsafety at TechFak.Uni-Bielefeld.DE>
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> The System Safety Mailing List
> systemsafety at TechFak.Uni-Bielefeld.DE

-- 

Peter Bishop
Chief Scientist
Adelard LLP
Exmouth House, 3-11 Pine Street, London,EC1R 0JH
http://www.adelard.com
Recep:  +44-(0)20-7832 5850
Direct: +44-(0)20-7832 5855