>>16784685
> Public safety are safeguarded by rigorous testings, not rigorous math.
How do you think those tests are derived? Where do those tests and their metrics come from? How do we know what standards of evidence for the tests are sufficient, and against what baseline? The answer to that question is math and statistics. Mathematical modeling is used to determine the range of conditions required for testing, and the expected failure states. Without the modeling, you don't even know what to test for and what conditions would count as success or failure, because you have no insight into the relevant safety factors and occurrence of events.
> If your product are one calculation away from public safety catastrophy, you are not doing engineering at all.
Again, how do you think they determine safety factors? Yes, in practice lower level engineering and manufacturing software is used, but where do you think the relationships and functions used within this software comes from?
> "My math model/simulation checks out" convinces no one its safe to ship and buy the product and engineers are the most skeptical of math and a little less skeptical of prototypes and full scale tests.
You are quite confused about how the engineering process works. You never just "trust the model," no matter where it came from. In order to do any development at the application level at all, you have to quantify what problems need to be solved, what factors are relevant to those problems, and what the relationships are between those factors.
That quantification is a mathematical process. There are parts that have been automated to some level for lower-rank "dummy checking," but that automation required someone to actually do the math and figure out the problem themselves.
Without actually being able to do the math yourself, you don't even have a starting point to solving the problem, let alone evaluating how good your solution is relative to alternatives.
