Statistics, Mechanisms, Facilitation
Chris and Fred discuss how statistics, mechanisms, and other technical words that come from ‘reliability engineering’ in contrast to softer terms like ‘facilitation.’ This comes from discussions Chris has had during his Statistical Process Control (SPC) course that he has run previously. So what is the big deal about these words? Well … you can know everything there is to know about statistics and failure mechanisms – but if you can’t make things better then it is all for nothing!
Join Chris and Fred as they discuss how we actually go about making things that are high reliability and high quality.
- Where did this talk come from? SPC. SPC stands for ‘Statistical Process Control.’ SPC is all about identifying statistical changes in a process. This could be manufacturing components or parts. It could be baking bread. It could be any sequence of events that are intended to create some sort of product. And these products usually need to comply with some specifications to be ‘good’ or ‘high’ quality. SPC is all about identifying when your system is ‘out of control’ as early as possible before you start making ‘low-quality products.’
- What is ‘in control?’ ‘In control’ refers to processes that are behaving as you expect. There is nothing weird going on. There is still natural variation, but this variation is acceptable and understood. When the variation or average value of a character starts to change and you don’t know why … then your process is ‘out of control.’ And the earlier you can identify when your process is ‘out of control’ the more time you have to bring it back in control before the specifications are violated. And a process can be ‘in control’ even if it varies … as long as you know why. For example, a component length might slowly increase as the blade in the tool of the production line that is manufacturing wears out. But if that blade is replaced every two weeks and the component length is always in specification … great! This is still in control.
- What is ‘good’ and ‘bad’? Depends on the scenario. The mass of a loaf of bread can vary by up to 10 grams and be ‘OK.’ This is quite a bit of variation for loaves of bread that typically weigh 900 grams. But this amount of variation is not OK for the mirror of the Hubble telescope. SPC is all about trying to address an out of control process before your product becomes bad.
- But then comes working out what causes the ‘bad’ thing. Once we work out that something might be wrong, then we do things like Root Cause Analysis (RCA) to find the potential root causes. Not our favourite root causes. Not our gut feels. But a raft of potential causes that are based on FACILITATING a discussion that really harnesses our actual corporate knowledge. Sometimes gut feeling works. But when we get used to going with gut feel, and it doesn’t find the root cause, then we have blame, embarrassment, et cetera. This is never good.
- So go with (1) Statistics, (2) Mechanics and (3) Facilitation. Statistics can give us early warning that something is wrong. Understanding the mechanisms means that we research and truly know how things fail. And finally, we facilitate. We make sure that our collective efforts are combined in a systemic way as opposed to going with the gut or shooting from the hip … which is not a foolproof approach.
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.