Re: Boeing and Dreamliner

["Robert I. Eachus" <rieachus@attbi.com>]

Dennis Lee Bieber wrote:

>         Surely there was a /requirements/ document for both A-4 and A-5, and 
> somewhere in there was something specifying the performance of the 
> vehicles. /That/ should have been the red-flag that something needed to 
> be retested. That is, if the A-4 had a spec stating some performance 
> feature of "x", and the A-5 equivalent shows "X+1", then shouldn't the 
> software responsible for handling the spec be tested to ensure it can 
> handle "X+1"?

Correct.  But what happened in the Ariane 5 case was a very well 
documented case of someone early on in the project defining how that 
testing was to be done, and management for cost reasons deciding not to 
do that particular testing--because the SRI had not changed one bit from 
the Ariane 4.

Anyone with major project experience should know this.  Any changes in 
the test plan have to be referenced back through the requirements 
allocations to see whether there are some requirements that are not 
being tested.  In this case the requirements that were not tested were 
that the SRI could control an Ariane 5, and that the SRI could accept 
the Ariane 5 trajectory.  Pretty fundamental requirements to fail to test.

-- 

                                                        Robert I. Eachus

�In an ally, considerations of house, clan, planet, race are 
insignificant beside two prime questions, which are: 1. Can he shoot? 2. 
Will he aim at your enemy?� -- from the Laiden novels by Sharon Lee and 
Steve Miller.