Data and Mechanisms
Join Chris and Fred as they discuss how reliability data analysis by itself doesn’t ‘mean’ a lot. What does this mean? And why is this?
- We are all human. Not computers. Which means if we are asking someone to challenge their own ‘status quo,’ then they need to understand what is going on as opposed to admiring a chart. If something is different to preconceived ideas, then it is natural for our judgment and corporate knowledge to start exploring why your data might be wrong. You need to instead, prove that you know why you are right.
- Do you want to improve reliability … or measure it? To improve something, you need to do something. We can’t fix data points. But you can fix the root causes that lead to failures (through failure mechanisms). So you need to know (for example) if your product is failing due to fatigue before we start preventing that failure from occurring.
- Not all data analysis is the same. Why? We can find all sorts of ways to modify data to suit our prejudice. And sometimes we don’t think it through. Think about a manufacturing team who produces products at various rates throughout the year. So if we simply look at ‘number of failures per month’ and ‘number of units produced per month’ … we get the wrong message. A simple reduction in production several months ago will reduce the apparent failure rate today. A simple increase in production today will increase the failure rate today. This is not helpful.
Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches.