Nah its not really.
With CNC you can't get a true curve you can only get a surface with steps in it, very small ones I might add but if you need a true curve for say a lens you have to use something else.
My field was none linear finite element analysis and pretty quickly you learnt to only work in metric SI whole units. It didn't really matter to much when doing linear models you can just post process but when you then went on to failure mode analysis or critical crack intensity you were wide open for human errors to creep in even keeping within the metric system of units.
If it was a pure none linear analysis such as contact, mooney rivilin materials or large deformation you needed to run it again to get units that would make any sense if you went away from metric SI.
If you did that you were truly international.
So if it was an oil job I would run everything in metric SI then get it all working then change the input deck base units by conversions or remove the SI conversions from the customer data and then run it again to get output for the report.
But oil was the only work which was imperial. The US engineering market must be loosing billions of dollars per year do to trying to stick to imperial. And American export products would be more viable as well in an international market.
Like it or not virtually all CNC machines are designed in metric. All the shafts are in metric, all the actuator screws are in metric. Its a metric binary device.
yes I agree that a stiffness matrix doesn't know or care what the units are. But I did notice the last time I saw a young engineer doing an analysis that he was forcing a solution through with an ill conditioned matrix and the mesh was dirty with no attempt to quad/cube areas of interest. So I agree that datums and standard boundary conditions is still a problem as its always been. But there also seems to be an attitude that the computer can solve anything. The lessons of engineering discovered 100's of years ago by the likes of Alexander Bell have been forgotten.
The marine engineering side of things are the only ones who seem to have a handle on standard datums and BC's
I think Boeing is the last none metric manufacturer out there of large airliners. And its pretty well documented that they are having problems with international suppliers and also part tolerances that are supplied due to this.
So yes metric isn't the solution, but metric SI as was adopted by US congress in 1866 which then they went further in 1875 to define the imperial system using the metric system is. You might be surprised how much in the US is metric already but just labelled in Imperial. All computers are metric, all cars since the 80's, bottle sizes are actually liters but packaged in the nearest Fl/oz.
Even ASME base units are metric SI now and the imperial versions are converted for the home market.
The US was actually metric before the UK by nearly 100 years.
I think the only reason why you don't move is public outcry. But you must be nearly the only country left on imperial. It must be costing your nation a fortune.
Personally I don't like some of the metric measurements for a consumer usage point of view. A pint of beer is a pint of beer and 500ml is not a pint.