Note: The top 10 FMEA mistakes covered in this article were adapted from a presentation titled “How To Get More Out Of Your FMEA” presented by Richard Harpster at the Quality Expo Detroit.
Top 10 FMEA Mistakes
by Richard Harpster, President of Harpco Systems
The top 10 FMEA mistakes were selected in the following manner: thousands of FMEAs were reviewed which were created over the past sixteen years in a wide variety of industries. The selection of the Top 10 mistakes was difficult since there were over 30 common mistakes identified to choose from. Systematic mistakes that affected all FMEA types were given the highest priority for inclusion in the list. One or more of the six Design & Process FMEA content mistakes discussed later in the presentation were found in more than nine out of the ten Design & Process FMEAs reviewed.
The first mistake that many companies make is to make the quality department the owner of the FMEAs. In many companies the quality department is given the responsibility for both creating and maintaining the FMEAs. This is wrong for two reasons. First the quality department does not own the design requirements, the design, or the process that the FMEAs are intended to assess. Second, in most cases they lack the detailed knowledge necessary to perform the FMEAs correctly. The System FMEA must be owned by the person responsible for defining the design requirements. The Design FMEA must be owned by the person responsible for creating the design. The Process FMEA must be owned by the person responsible for the processes that will be used to produce the product.
The second mistake is that the wrong people are selected to participate in the FMEAs. A complete cross functional team is not required FMEA type. There is nothing worse than to have to attend an FMEA that you cannot actively contribute to. System FMEA participants must include people who are knowledgeable about customer requirements and the design requirements required to meet them. Design FMEA participants must include people who are knowledgeable about the design requirements, the design, and any manufacturability issues that the design might create. Process FMEA participants must include people who are knowledgeable about the process, including possible received material issues, equipment operation and equipment maintenance.
The third mistake is that the FMEAs are done at the wrong time. Many companies do FMEA because they have to. And consequently the completion date is based more on a paper work submission date requirement rather than the role the individual FMEA type output plays in the product development system. The first cut of the System FMEA should be completed before the design requirements are released to the designers since its purpose is to assess the adequacy of the design requirements. The first cut of the Design FMEA should be completed before the prints are released to the manufacturing process since its purpose is to assess the adequacy of the design. The first cut of the Process FMEA should be completed before the process is approved for use since its purpose is to assess the adequacy of the Process FMEA in producing the product.
The fourth mistake the companies make is to treat the different FMEA types as individual nonintegrated forms to be filled out. The System FMEA, Design FMEA and Process FMEAs are integrated tools that involve input from and output to multiple elements of the product development system, if the effectiveness of the FMEAs is to be optimized. When properly performed and integrated, the three FMEA types are part of twelve different major import/output linkages within the product development system.
The fifth mistake companies make that are responsible for the design of the product is not performing a System FMEA. The major objective of the System FMEA is to identify and properly define design requirements. And a properly defined design requirement is the one mistake that cannot be contained using product testing either during the design phase or after the product is produced since the acceptance criteria for the testing is based on the design requirement. If you get the design requirements wrong nothing else matters.
The sixth mistake companies make is using RPN to determine what to work on. In the example provided the Severity, Occurrence and Detection ratings in the first line of the Process FMEA indicate the following: the Severity rating of five means a process will create an out of spec condition that will result in the reduction of a secondary function of the product and consequently its return. The Occurrence rating of a six means the out of spec condition may be present in up to one percent of the product due to the cause identified. The Detection rating of three means the out of spec condition will be contained in the station that creates it within the process.
The second line of the Process FMEA indicates the Severity of three means that the process will create an out of spec condition that will result in an appearance issue that the customer does not currently find to be a returnable issue. The Occurrence rating of ten means the out of spec condition will be present in ten percent or more of the product. The Detection rating of eight means the out of spec condition will be inspected for visually in the process. If you work on the line items with the highest RPN number you would be working on line number two with an RPN of 240. It is interesting to note that since the product will not be returned for the appearance issue this line cost the company no money at all. The second line however with the RPN of 90 will cost the company considerable funds because it basically identifies the fact that up to one percent of the product will have to be either reworked or rejected.
The seventh mistake a company can make is improper use of the Class column. The Class column or Special Characteristic designation was created to identify issues that expose the company to unacceptable safety and/or financial risk. Because of the unacceptable risk level any line of the Process FMEA with a class symbol, assuming it has been properly determined, MUST be worked on. The two key components used to determine whether or not the class symbol is warranted is the Severity of the effects of the issue and the expected probability of the issue occurring. The biggest mistake made is when design engineers who have limited knowledge of a process capability to meet a specification are allowed to specify which characteristics are to be given class symbols in the Process FMEA based on the Severity of the effects of the issue alone without considering the capability of the process.
Following are three of the most common content mistakes made when creating a Design FMEA. The first of these three most common mistakes is turning the first problem into a Bill of Materials. If the product provides a customer with what they want, the customer does not care about the participation of each of the components. The second most common mistake is only including functional related requirements in the first column. There are 17 different categories of requirements that must be considered when creating a design. Many of these categories of requirements have nothing to do with function. The third most common mistake is putting Failure Modes in the Failure Cause column. Component breakage is not a Failure Cause. It is a Failure Mode that must be prevented.
The ninth mistake that companies make is not placing or doing the Process FMEA on the Receiving actions. Many FMEA manuals tell you to assume that all received components are received within specification. After 16 years of being exposed to a wide variety of industries I can honestly say that I know of no companies whose suppliers always send them within specification product. If a company is an assembler of received components there is no process that will affect their ability to be successful more than Receiving. It is essential that all manufacturing processes assess the risk posed by those suppliers and have an adequate plan to deal with it. The proper performance of a Process FMEA on received materials is the first step of developing that Reaction Plan.
Following are three of the most common content mistakes found in the Process FMEA. The first of the three most common mistakes is putting Failure Causes in the Failure Mode column. This is a very common mistake made by people who use Dynamic Control Plan methodology because of a fundamental flaw in the methodology. The second most common mistake is putting non-root causes in the Failure Cause column. The three causes shown are three of the most popular entries in the Failure Cause column of a Process FMEA. It is not that these are not Failure Causes, it is that they are not root causes. And unless you get to the root cause you cannot identify the proper Prevention Controls to prevent them. The third most common mistake is putting generic entries in the Controls column. Again, these three entries here are three of the most common entries we see in Process FMEA Controls columns: Operator Instructions, Setup Instructions, and First Piece Inspection.
If you remove these ten mistakes from your FMEA implementation activities and use what the FMEAs teach you about the weaknesses of your products and processes you would see a significant improvement in the quality of your products and the productivity of your processes.
About The Author
Richard HarpsterRichard Harpster is president of Harpco Systems, which he founded in 1987. Harpco Systems specializes in providing software, training, and consulting for risked-based product lifecycle management (RBPLM®). During the past 30 years, Harpster has helped hundreds of companies implement improved risk-based design and manufacturing systems in a wide variety of industries. He is a recognized expert in the application of FMEAs and has invented several new concepts, including the linking of design FMEAs to process FMEAs in 1990, which became an automotive industry standard 18 years later. His latest inventions in the field of RBPLM® include Requirements Risk Assessment (RRA®), Usage Risk Assessment (URA™), Multiple Integrated Cause Analysis (MICA™) and Rapid Integrated Problem Solving (RIPS®). He has published several papers on the topic of RBPLM®.
“Harpco breaks down the barriers and corrects the pitfalls so companies can reap the full benefits of FMEA. Classroom training not only properly teaches FMEA, but participants actually build their business’s FMEA as they go and are often blown away by how much work got accomplished. Harpco Systems has become known as the Modern FMEA for a reason. Its structured, simplified and sustainable.”
“We used QPlus to achieve Q1 and ISO 16949 successfully at the Ford Motor Co. Monroe BAO Plant. The software promoted linked documentation that prevented issues at internal and external audits. Assured the quality documents at the operations production level had all relevant and concurrent information that was reviewed in the program files. QPlus allowed the program members to create a baseline “Hot End Exhaust” database that produced linked documentation from the DFMEA to the production visual aids the operators used to perform correctly.”
“Harpco’s training is first rate and helps develop new ways of thinking about the importance of creating proper specifications early. Separately, I’ve seen firsthand the effectiveness of using their approach in problem solving, helping to advance problems that had reached a stall using traditional methods.”
“We were very fortunate to work with Rich Harpster and his team as we improved our DFMEA process at Calsonic. There is no better teacher, coach, implementer than Rich when it comes to creating a knowledge base for engineers to use in creating part specs to assure that products meet the customer’s requirements. Rich teaches the basics, then he accelerates the FMEA process so that requirements and specs are related in a database that can be continually updated.”
“I would like to thank Harpco Systems for the help, advice and frankly the education in how to design, develop, source and manufacture new to world products.”