It's also not surprising that an off - the - cuff prediction that the future will look like the past is not a bad way to predict the future, but the models were not given the benefit of that shortcut, and so what was at issue was their ability to yield
reasonable output on the basis of inputted
physical principles, initial and boundary values, parametrizations, and appropriate choice of simplifying
assumptions.
For example, if your company discovers a major defect in one of its products that has the potential to cause
physical injury, it's probably a
reasonable assumption that litigation will soon follow.