We continue from Part 1, Part 2, and Part 3 with Failure Modes and Effects Analysis Part 4. In this lesson, we go over how to use the FMEA tool to analyze the important variables carried forward from the Cause and Effect Matrix using QuikSigma.
Let’s turn our attention a little farther to the right than the FMEA and look at the RPN or Risk Priority Number. This is just calculated by multiplying severity times occurrence times detectability. Things that produce bad effects and that are very likely to get out to our customers and happen a lot, are going to have high RPN numbers. On the other hand, things that don’t meet that criteria are going to produce low RPN numbers. Some authors suggest that action items are required for all our RPN’s above 100. Well that doesn’t make sense to me at least, because different FMEAs use different anchor points so our decision point might be different, and the other thing is that for some peculiar reason, that rule seems to produce a lot of RPNs just below 100. I don’t know why that might be. In any event, if you’ve got a lower RPN number, you’re perfectly justified in writing here in your suggested action, none and believe me, that will be your most common suggested action item and that’s wonderful. What that says is of the millions and millions of things that I could be doing, I’ve looked at this one and it’s not worth my time right now.
Most of your action items will be pretty obvious. If our position descriptions are not good, then we know we’ve got an action item to fix them. You will find that very often one action will meet many many needs and when that happens you say, “Wow, that’s neat! I’m solving many problems with one action.” So we have a manual carry forward here. You can just set this little arrow and at one instance of each of your action items gets carried forward to the action plan. So one of the effects of the FMEA is to prioritize and give you visibility of the things that are most important to work on.
Now, let me show you one other thing. I’m going to click on recipe and I’ve loaded some data here and the problem that we have is that we don’t know what the best recipe is. Well, this is one where the correct action is not obvious. So what we have to do is find that out. Well, believe it or not this is a really good candidate for a designed experiment and so the FMEA will tell us to do that and then we can carry that forward to our action plan, which you’ll see in the next video.