There's a nugget of truth in the dad's explanation. Empirical data, some which are critical failures, are used to predict the limits of infrastructure. They don't destroy a copy of every single bridge because it's expensive and not even that useful for getting new data, but things in general get built and destroyed to test new materials or designs.

[–] [email protected] 1 points 1 year ago* (last edited 1 year ago)

Careful analysis of the stress tensor of a simple geometric model with parameters given by empirical testing of simple materials can give you a reasonable ballpark estimate. I wouldn't be the first one to drive on a ballpark estimate though.