<<< Back to Index
This Month's Article >>>

FMEA Corner 
This Month's Theme is FMEA and Robust Design
Next month's theme will be decomposing system functions to lower levels

Every month in FMEA Corner, join Carl Carlson, a noted expert in the field of FMEAs and facilitation, as he addresses a different FMEA theme (based on his book Effective FMEAs) and also answers your questions.

Questions and answers are a great way to learn about FMEAs, for both experienced and less experienced FMEA practitioners. Please feel free to ask any question about any aspect of FMEAs. Send your questions to Carl.Carlson@EffectiveFMEAs.com, and your contact information will be kept anonymous. All questions will be answered, even if they are not featured in the FMEA Corner.



ro·bust [roh-buhst, adjective]
The Oxford English Dictionary has two connotations for the word "robust":

  1. (of an object) sturdy in construction

  2. (of a system) able to withstand or overcome adverse conditions

What is a Robust Design?

The engineering application of robust design is not merely "sturdy in construction." Focusing on making designs sturdy without paying attention to other parameters, such as weight and packaging, risks sub-optimizing the design. For example, if a design engineer focuses only on sturdy construction, the product may end up too heavy or too big.

The key to robust design is understanding the second connotation of the definition, "(of a system) able to withstand or overcome adverse conditions." Robust design is a methodology to find the appropriate control factor levels in a design to make the system less sensitive to variations in uncontrollable noise factors (i.e., to make the system robust). Note, the different types of noise factors are explained in Issue 182 of Reliability Hotwire.

What is the origin and application of Robust Design?

The following is an excerpt from a paper by the faculty at Iowa State University, titled "Robust Design":

Robust product design is a concept from the teachings of Dr. Genichi Taguchi. It is defined as reducing variation in a product without eliminating the causes of the variation. In other words, making the product or process insensitive to variation.

This variation (sometimes called noise) can come from a variety of factors and can be classified into three main types: internal variation, external variation, and unit to unit variation. Internal variation is due to deterioration such as the wear of a machine, and aging of materials. External variation is from factors relating to environmental conditions such as temperature, humidity and dust. Unit-to-unit variation is variations between parts due to variations in material, processes and equipment.

Examples of robust design include umbrella fabric that will not deteriorate when exposed to varying environments (external variation), food products that have long shelf lives (internal variation), and replacement parts that will fit properly (unit-to-unit variation). The goal of robust design is to come up with a way to make the final product consistent when the process is subject to a variety of "noise."

What is the relationship between Robust Design and FMEA?

As covered in chapter 3 of Effective FMEAs:

The primary objective of an FMEA is to improve the design. For System FMEAs, the objective is to improve the design of the system. For Design FMEAs, the objective is to improve the design of the subsystem or component. For Process FMEAs, the objective is to improve the design of the manufacturing process.

The end result of an FMEA should be design (or process) improvements. One of the most important types of design improvement is a robust design.

What specific ways can an FMEA support Robust Designs?

There are at least two ways that an FMEA can support robust designs. The first way is to perform a Parameter Diagram (P-Diagram) before beginning the System/Design FMEA. A P-Diagram takes the inputs from a system and relates those inputs to the desired outputs of a design that the engineer is creating, while considering non-controllable outside influences. The subject of using a P-Diagram to support an FMEA, including how to map P-Diagram elements to an FMEA, is discussed in Issue 182 of Reliability Hotwire.

The second way that an FMEA can support robust design is by using selected action strategies to reduce risk. Risk reduction strategies are thoroughly discussed in chapter 7 of Effective FMEAs. When the FMEA team identifies risk that can be reduced by making the product design more robust, they should enter the improvement task in the Recommended Actions column of the FMEA. The action(s) can be specific design changes to make the design more robust, or can launch specific techniques that can be performed to make the design more robust.

For example, there are methods other than FMEA that support robust design. Specifically, Design of Experiments (DOE), also called Taguchi methods, is the primary technique that achieves robust design. FMEA teams can use the Recommended Actions column of their FMEA to recommend a DOE.

Does Xfmea support Robust Design?

Yes. Xfmea supports both of the approaches described above.

In Version 11 of Xfmea, P-Diagrams will be integrated into the software. This will allow Xfmea users to develop P-Diagrams for any FMEA project directly in the software and sync to the FMEA worksheet. Version 11 is expected to be available in early 2017.

In the current version of Xfmea, users can use the Recommended Actions column to display the robust designs tasks identified by the FMEA team. Additionally, Xfmea can be configured to add a user-designed category to Recommended Actions. See the "Action Category" example below (in the Worksheet view of Xfmea). The following FMEA excerpt covers one line from a fictitious Hand Brake Design FMEA. Notice the highlighted Action "Redesign hand brake cable routing to reduce friction and make the system insensitive to lubrication degradation."

FMEA example

What is an example of using FMEA to implement Robust Design?

In this brief and fictitious example, we’ll take a further look at the hand brake subsystem of a bicycle.

A P-Diagram performed as part of the preparation for a hand brake Design FMEA reveals a "noise factor" of brake pad orientation. The friction generated by the brake pad against the wheel rim is sensitive to the orientation of the brake pad when it is assembled in the plant or by the user.

When the FMEA team is addressing a design-related cause that has to do with the orientation of the brake pad, they can consider robust design techniques, such as DOE or product error-proofing, to make the brake pad adjustment design more robust to the "noise" of adjustment.

What resources are there for Robust Design?

Many excellent articles and books have been written on the subject of Robust Design, and can be found through internet search. The following are a few resources for readers to learn more about this subject.

!FMEA Tip of the Month

Use the Recommended Actions column of the FMEA worksheet to initiate and track the completion of selected quality and reliability tools. This is one of the keys to successful FMEAs. The FMEA team may not always be able to identify the exact engineering solution to reduce risk to an acceptable level. When this is the case, the team should recommend the tool(s) needed to improve the product or process design.

?Something I’ve always wanted to know about FMEAs
The important thing is not to stop questioning. - Albert Einstein

A HotWire reader submitted the following question to Carl Carlson. To submit your own question about any aspect of FMEA theory or application, e-mail Carl at Carl.Carlson@EffectiveFMEAs.com.

The AIAG standard says OCC is estimated during the design life of the product (so I assume before the product launch, right?). During this stage there would only be a few prototype samples to be tested, and most likely no occurrence of failure, if we only perform pass/fail testing. How should we rank the occurrence, if there is no observed failure, but the design is not error proofed?

  1. When there is no similar design?

  2. When there is a similar design?

Also, at some point when we have enough samples being tested and may have discovered many failures, then during the DFMEA review, OCC might increase maybe from 3 to 5 for example, so again there is no improvement on RPN. When is the point the OCC will be improved given the fact that the time is running and moving from gate to gate?

Carl: I’ll begin my reply with the definition of "occurrence" from my book:

Occurrence is a ranking number associated with the likelihood that the failure mode and its associated cause will be present in the item being analyzed. For System and Design FMEAs, the occurrence ranking considers the likelihood of occurrence during the design life of the product. For Process FMEAs the occurrence ranking considers the likelihood of occurrence during production. It is based on the criteria from the corresponding occurrence scale. The occurrence ranking has a relative meaning rather than an absolute value and is determined without regard to the severity or likelihood of detection.

The phrase "the design life of the product" means at any time during the intended life of the product, in other words the period of time when the product is intended to be in service. For Design FMEAs, this is meant to be the best estimate of the Design FMEA team for the likelihood of the failure mode/cause occurring at any time during the design life of the product, and the estimate is made by the FMEA team at the time when the FMEA is being performed (it is called an initial assessment). The FMEA team may or may not have hard data (such as field data for similar products, or actual test data) when making this initial occurrence assessment. It can be a subjective assessment.

About the Author

Carl S. CarlsonCarl S. Carlson is a consultant and instructor in the areas of FMEA, reliability program planning and other reliability engineering disciplines. He has 35 years of experience in reliability testing, engineering and management positions, and is currently supporting clients from a wide variety of industries, including clients of HBM Prenscia. Previously, he worked at General Motors, most recently senior manager for the Advanced Reliability Group. His responsibilities included FMEAs for North American operations, developing and implementing advanced reliability methods and managing teams of reliability engineers. Previous to General Motors, he worked as a Research and Development Engineer for Litton Systems, Inertial Navigation Division. Mr. Carlson co-chaired the cross-industry team that developed the commercial FMEA standard (SAE J1739, 2002 version), participated in the development of SAE JA 1000/1 Reliability Program Standard Implementation Guide, served for five years as Vice Chair for the SAE's G-11 Reliability Division and was a four-year member of the Reliability and Maintainability Symposium (RAMS) Advisory Board. He holds a B.S. in Mechanical Engineering from the University of Michigan and completed the 2-course Reliability Engineering sequence from the University of Maryland's Masters in Reliability Engineering program. He is a Senior Member of ASQ and a Certified Reliability Engineer.

Effective FMEAsSelected material for FMEA Corner articles is excerpted from the book Effective FMEAs, published by John Wiley & Sons, ©2012. Information about the book Effective FMEAs, along with useful FMEA aids, links and checklists can be found on www.effectivefmeas.com. Carl Carlson can be reached at carl.carlson@effectivefmeas.com.