Is There a Correlation Between Machines and Machines?

John Sutter: I'm going to check in on two things. One, you'd mentioned sort of they figured out this unique correlation that you had talked about. My supposition here is that we would have need to look at a lot of failure cases to be able to come up with that hypothesis because it's very unlikely that you could take even the most experienced guy that you have in the field and say, Hey, guy, what's the vibration sensor reading? The hypotheses about that are horrendous to guess because human beings don't really have that. So yeah, just having really good historical data, having much analysis had already been done.