Is the model right?
As it turns out, the decision to forgo Boeing’s Starliner return of the stranded astronauts came down to whether they believed the model’s prediction of thruster degradation — will the thrusters have enough helium for the return trip?
Boeing says, “yes,” and NASA says, “we’re not so sure.”
During Mercury, Geminin, and Apollo (I have no idea what their policy is today), NASA had a policy of “prove to me it will work” rather than the subtly distinct but oh-so-different, “prove to me it won’t work.” We don’t go unless you can prove it works. The Challenger disaster happened because management violated that policy of decision-making.
As an engineer, I have built and used models my entire career. Models are amazing tools, but they all have limitations. The more complex, the more limitations.
Models are tools, not guarantees. That’s always the model dilemma.
Boeing might be right (I’d be willing to bet they are), but NASA made the right call. Because in this game, human lives don’t get a second chance.
