[SystemSafety] Data on Proof effectiveness from real projects

[paul_e.bennett at topmail.co.uk]

On 02/04/2016 at 8:22 PM, "Derek M Jones" <derek at knosof.co.uk> wrote:
>
>Steve,
>
>> What's your source for the 44% number? I'm very interested in 
>that kind of
>> data. My source says 56% of defects are in requirements and 27% 
>are in
>> design. Only 7% are coding. The remaining 10% are defects logged 
>against
>> other things: test cases, test plans, project plans, etc.
>
>There is no reason to think that any of these percentages should be
>confined to a narrow range of values.
>
>Percentages can also be misleading.  X might have 70% in 
>requirements,
>30% in coding, while Y has 30% in requirements and 70% in coding.
>Subsequent investigation finding that X and Y had the same number
>of defects in requirements but very different processes going on
>in coding.

As I understood it, the percentages related to the origination of the errors 
that made it into the final product. It is understood that the initial writing of
the requirements specification is likely to be quite error prone. How well
the development process copes with filtering out the specification errors
is a measure that is quite interesting.

Consider that if your development  process was able to find and filter out 
the majority of the specification errors the later stage percentages would 
be higher and the specification percentage would be a lot lower. I aim to
minimise the specification error by doing enough up-front work to test the
voracity of the specification before I will begin a design phase. I may
even use early rough prototypes (these never reach production) to explore
specification space propositions. Starting design on a more solid basis 
does save time later as there is usually less effort required in the design 
and testing phases to produce the final more correct product.

Regards

Paul E. Bennett IEng MIET
Systems Engineer

-- 
********************************************************************
Paul E. Bennett IEng MIET.....<email://Paul_E.Bennett@topmail.co.uk>
Forth based HIDECS Consultancy.............<http://www.hidecs.co.uk>
Mob: +44 (0)7811-639972
Tel: +44 (0)1392-426688
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************