Ever heard (or said) the following?
“We’re basically operational right now. Things aren’t perfect, but we’ve done a lot worse along the path to get here. We’ve got a system in place that, while not optimized, gets the job done. Everyone has lots of plates in the air, so I just can’t justify making a huge change to perfect one particular plate. That might bring them all crashing down.”
Yep, I can understand that. With 3 kids under 6 and a demanding (but really fun!) career, it would take a lot to get me to consider switching from gas to solar heat in my home. Maybe solar heat would eventually save me 50% on my outrageous New England oil bills… but my decision would be far more based on fear of the unknown. I already have a stable system that gets the job done.
Sure, I’d be worried that solar heat might not meet the hot shower and bath needs of my whole family… but I’d be more worried about the unexpected problems that could pop up. For instance, if there is a heat failure, are all the water pipes in my house going to freeze and explode? How much would that cost to fix? More importantly, how long would it take?
Sometimes I get overly evangelical about the “radical transforming power” upfront CFD can have on your quality and bottom line. That probably scares the crap out of some of you.
Let me step back a moment and make this key point very clear:
Your adoption of Upfront CFD will not philosophically change your Engineering process. It will also not demand that your existing design engineers gain superhero PhD skills. In the end, your process will look much the same as it does today. Instead of a bunch of monkeys beating on prototypes in the back room until something good happens, that same bunch of monkeys will do some of that hammering on 3D models from their CAD tool. (By the way, I used to be a monkey… so I’m free to discuss primates.) It’s all still about trial and error.
So, your fundamental, known, reliable process will stay intact after implementing upfront CFD. You’ll simply compress the time required for trial and error… or, conversely, allow for much more productivity in the same time period. You will also get far more feedback from each hammer clang. That feedback will simply give your monkeys a way better idea where to whack next.