Moral of the story is probably to buy fewer numbers of new designs more frequently.
With you there N-a-B. It's a hobby-horse of mine that the gestation period of major defence projects is just so long that they will
inevitably be inefficiently managed. In a 25 year period a project will be threatened by political, social, economic and technical risk, and pushed, pulled, delayed and generally buffeted from all sides.
A bit more realism at the outset concerning costs and time-scales might help - but I guess that would risk not getting started at all. Whatever, if at the end of 25 to 30 years you end up with a system that is still relevant, that is as likely to be due to luck as much as judgement. That is not an ideal way to commit hundreds of billions of dosh.