Use FMEA to Choose Critical Design Features

We’re heading into pilot production.

We’ve got to finish those design specs and engineering drawings!

• Manufacturing wants to start cutting steel for molds.
• Suppliers want to work out the specs on their end.
• QA is asking what they should inspect.

Before you pass it along, have you identified what features are critical? Does it align with managing risk? And are you communicating that effectively on your specs and drawings?

Creating design specs is an important part of engineering design. We review a way to choose critical design features, based on risk.

View the Episode Transcript

Episode Transcript

We’re heading into pilot production where some initial parts might be made in a manufacturing incubator, or as a trial run of real production process. We’ve got to finish those design specs and engineering drawings. Manufacturing wants to start cutting steel from molds suppliers. Wanna work out the specs on their end and quality assurance is asking what they should inspect before you pass it along. Have you identified what features are critical and does it align with managing risk? Are you communicating that effectively on your specs and drawings? Let’s talk about critical design stuff after this brief introduction.

Hello, and welcome to Quality during Design the place to use quality thinking to create products others love for less. My name is Dianna. I’m a senior level quality professional and engineer with over 20 years of experience in manufacturing and design. Listen in and then join the conversation at qualityduringdesign.com.

Setting design specifications, which is setting the materials and any critical processing parameters like drying materials for molding and setting dimensions and tolerancing them. These are all important parts of engineering design. A lot of our internal customers are going to be using this information so that they can complete their work. Test engineers are looking at this to verify and validate the product based on those critical features. Manufacturing and suppliers are looking to qualify their processes based off of measuring those critical features. And they’re also going to be doing some ongoing process monitoring like SPC, and they’ll likely be measuring what you’ve identified on the design spec as critical. Quality assurance wants to inspect and measure critical features. They’re going to be making decisions about production lots based on their findings. Is the product made to spec and is it good? Or is it not made to spec and there’s a problem? We also have future designers. Our design may need to change. Maybe the material’s no longer available. So there needs to be material change or there’s an upgrade, a version 2.0. Future designers are going to look to validate their design changes against what’s critical on the current spec.

If our design specs are wrong, if the critical features that we have are not identified, then everyone’s going to be assessing the ongoing quality of product based on the wrong thing. The whole underlying purpose of testing and manufacturing, qualifications and monitoring and quality assurance inspections is to ensure that we’re making the product right. If we make the product right then that will ensure that we make the right product for the customer. Our product’s going to meet its requirements and our customer’s expectations. Our product is able to be used properly in a safe way. It’s going to do for the customer what we say it will do. It will be able to be maintained and disposed of properly. If we make the product right. It’ll also ensure that it meets our expectations. It’s going to perform the way we think it should. It’s going to be able to last as long as we think that it needs to, it will perform as duties or functions, the way that’s necessary. And our product is going to be able to control the risks, the risks of its own use process and the risk of its own potential failures.

What is one way we can define the critical features? Is there a way we can do it that’s based on something besides our engineering intuition. Yes! We can use our risk management analyses. One of those analyses we can use is FMEA, failure mode and effects analysis. Now, if you’ve been listening to the podcast for a while, I can hear some of you. “Oh no, she’s talking about FMEA, again!” But hear me out! Let’s go over a scenario.

We’ve already gotten an FMEA on hand. We’ve been developing them during the concept development and the design process. So we have an initial framework of potential failures effects and causes and will also have a severity. Now imagine this, we’re finishing the details on an engineering spec.

So we go to the FMEA for either that component or the subsystem that it’s used within. And we scan through the potential ways that this part of the design could fail. And we focus in on where the failures could have a high severity. What feature of the part is associated with that failure, thinking about the potential failure and the feature associated with it. Is there something about that feature that should be tested, measured or monitored to ensure that the product was made right to prevent that failure?

Then we scroll over to the controls. What controls did our team list there? We’re at the point in our design process where we might have prevention controls, because we’ll have designed-in controls within our design concept, we’re just now working on the engineering spec. So there may not be any detection controls. What controls can we add?

Let’s think about our internal customers again. For our test engineers: are there critical features that we would want to define or to ask them to test, to make sure that we’re making the product right? For our manufacturing and our suppliers: what do we want to ensure that they design their process around so that the product is consistently made right? Whatever they see as a critical feature, or is dimension on our engineering spec – that’s what they’re going to focus on for their manufacturing development. Just like product design engineering, process design engineers need to make decisions based on tradeoffs. Knowing what is critical and what must be maintained and monitored and controlled will help them develop the manufacturing process that will consistently make the right product, based on risk. And for quality assurance: they want to be able to inspect and measure what’s critical. They want to be able to look at a part and test or measure a few things to just make sure that the part continues to be made right.

Now, in the beginning of this episode, we also talked about future designers. Labeling what’s critical can help those future designers revalidate a product change. When they’re making a change to the design, they can also look to the FMEA for that part: what’s the potential failure mode and is anything about what I’m changing affect this part’s failure mode? This can help guide them on the decisions they need to make with the design change.

There are lots of other things to consider when setting design specs. Two of the ones that I see most often are 1) can manufacturing make it and 2) can things be properly measured. With “can manufacturing make it”, I’ve seen lots of parts where there are injection molded parts that have tolerances that are within two or three thousandths of an inch. Even if they can do that, it might make the process in manufacturing too expensive, because now we might have more wasted parts, extra controls, and qualification of the process might be harder.

And also thinking about defining critical features that can be properly measured: This affects not only manufacturing, but also quality assurance and test. If they can’t properly measure it, it makes it difficult or impossible to monitor. And it might delay that instant feedback that they may need. Maybe they are monitoring the manufacturing of a product, and they’d be able to tweak the machine if they had that instant feedback. But the critical feature that they’re supposed to measure requires a high end vision system that’s only located in the test lab. And the test lab can only measure it at the end of a lot. Tat may not give them that instant feedback. Now they have delayed measurement, delayed reaction and more scraped product.

So, what’s, today’s insight to action? When working on those design specs, we can align the critical features with risk and risk controls. And we can think of those internal customers who will be making the product. We want to define critical features to help them to ensure that they make the product right. That will help all of us to produce the right product for our users that meets everyone’s expectations.

If you like the content in this episode, visit quality during design.com, where you can subscribe to the weekly newsletter to keep in touch. This has been a production of enterprises. Thanks for listening.