By Mark Miner

    You can't fully engineer a real product.  There are too many variables.  Unless you're making steel cubes or aluminum spheres, or rods of something equally banal, you're going to have to go by your gut somewhere.  Otherwise you wouldn't do jack.  Take a month to do all the vibe mode testing of each component, and your competitor has the product to market.  He probably spent a few bucks more on unnecessary material, and it may have a natural frequency that some 4-sigma event could hit, but look at the cash he made selling it vs the cash you spent engineering it.
    The bar for "good enough" moves around, of course, and that's why your people had better know their stuff, but most importantly, everybody had better know their customer.  I work on missiles.  Uncle Sam's bar is pretty high.  Nevertheless, if you have good people with a good feel for their job and the right training, you can look at a brick and call it a brick.
   Only thing is, it had better be a brick.

The author is learning to deal with defense-grade documentation at his new job at Orbital Sciences.

 

What did you think of this article?

Liked
Disliked
No Opinion