Featured Video
This Week in Quality Digest Live
Operations Features
Jon Speer
Avoid jeopardizing the health of your QMS
Capture 3D
GE Appliances uses ATOS 3D scanners, TRITOP photogrammetry, and ScanBox to solve engineering challenges
Eryn Brown
They’re a common index of technological creativity, but research finds they can impede rather than encourage it
Megan Ray Nichols
Five approaches for addressing the shortage

More Features

Operations News
Planned manufacturing events for Oct. 5, 2018, on track to set record-breaking level of participation
Interactive event, Sept. 19–20, 2018, at California Polytechnic University
Preparing your organization for the new innovative culture
New system patches security holes left open by web browsers’ private-browsing functions
Design structure allows for progressive, efficient, and practical measurements
Management's role in improving work climate and culture
Customized visual dashboards by Visual Workplace help measure performance
Cricket Media and IEEE team up to launch TryEngineering Together

More News

Richard Harpster

Operations

Quickest Way to Effective AIAG-VDA-Harmonized FMEAs

Improve the ability to use DFMEAs and PFMEAs effectively and efficiently

Published: Wednesday, April 11, 2018 - 12:03

The AIAG-VDA FMEA Handbook committee and everyone who responded to the request for comment on the proposed AIAG-VDA failure mode and effects analysis (FMEA) manual must be applauded for their efforts. Harmonizing the VDA and AIAG FMEA methods is not an easy task. According to industry sources, there were 4,000 or more comments on the proposed handbook. I believe this shows two things. First, people recognize the importance of the document. Second, they believe significant changes are required.

The critical question that must be answered is how to proceed. Everyone on the committee is involved in the automotive industry. They know that before one designs a manufacturing process, they must first decide what they want to manufacture. In most cases, if the produced product is within specification, the method of manufacturing used is not important. The only exceptions would be if the manufacturing process violated a law, was unsafe, or harmed the environment.

Why should creating design and process FMEAs be any different? Instead of defining the process that is to be used to create the entries for the design and process FMEAs, and then defining what the FMEA forms must look like to hold the results, why not begin with the design and process FMEA forms and decide what they must contain to be effective risk management tools? If the forms have the proper information to manage risk, how the information gets there is irrelevant.

Seven key risk questions

To manage design and/or manufacturing process risk, the following seven key questions must be answered by the design and process FMEA forms:

1. What is the objectionable incident that can cause harm?
2. What is the harm and severity of harm that the objectionable incident can cause?
3. What are the potential root causes of the objectionable incident?
4. What risk controls are in place to prevent the root causes of the objectionable incident or reduce the severity of harm if the objectionable incident occurs?
5. What is the probability of the objectionable incident occurring due to each of the potential root causes when the current risk controls are used?
6. Which objectionable incident/root cause combinations should be worked on and in what order?
7. What is the status of risk reduction activities?

Considerable harmonization already exists

If one looks at the DFMEA and PFMEA forms in Potential Failure Modes and Effects Analysis FMEA, Fourth Edition (AIAG, 2008) and Quality Management in the Automotive Industry: Product and Process FMEA (VDA, 2012), they will see two things. First, the column headings of the DFMEA and PFMEA forms are very similar. Second, columns exist in both manuals’ DFMEA and PFMEA forms that are sufficient, if properly used, to answer the first six risk management questions.


Click here for larger image

The one major difference between the 2008 AIAG and 2012 VDA forms is the area used to manage recommended actions to reduce risk. The 2012 VDA form contains a column to track performance of the actions in the Preventive Actions and Detection Actions columns but does not contain a recommended actions area. The 2008 AIAG design failure mode and effects analysis (DFMEA) and process failure mode and effects analysis (PFMEA) forms contain a separate Recommended Actions area where the information required to answer question seven above can be captured.

Given the similarities between the DFMEA and PFMEA forms found in the 2008 AIAG and 2012 VDA FMEA manuals, as well as the current quality of the forms, creating harmonized DFMEA and PFMEA forms should be quite simple; however, defining what is supposed to go in each column will be more difficult. The major problem with both the 2008 AIAG and 2012 VDA FMEA manuals is that the instructions provided to define the required contents of each column are inadequate or incorrect for many of the columns. This leads to DFMEA and PFMEA processes that are not optimized. 

If the harmonization efforts resulted in a common DFMEA form, a common PFMEA form, and clear and correct definitions of what must go in each column of these forms, the automotive industry would benefit greatly via improved designs and manufacturing processes.

What the AIAG-VDA FMEA Handbook committee should do

The AIAG-VDA FMEA Handbook committee’s primary concentration should be on the final DFMEA and PFMEA forms and what filling them out correctly looks like. The definition of what goes into each column must be simple to understand. Rating tables must be as objective as possible with simple, short descriptions of what each rating means. 

Creating the correct definitions for what goes in each column of the proposed AIAG-VDA FMEA Handbook will not be easy because the majority of FMEAs are currently done incorrectly. This is due mainly to previous FMEA paradigms that have been accepted for more than 20 years. They are difficult to break because many of the improper FMEA paradigms are supported by industry FMEA manuals. The proposed handbook contains improper FMEA paradigms that can be found in its predecessor VDA and AIAG FMEA manuals. The paradigms must be broken if people are ever to experience the true power of FMEAs. 

One example of an improper FMEA paradigm found in the proposed handbook and all previous AIAG and VDA manuals, which severely inhibits the effectiveness of a PFMEA as a risk management tool, is the definition of “occurrence.” The proposed AIAG-VDA FMEA Handbook defines PFMEA occurrence as: “The Occurrence rating (O) describes the occurrence of the failure cause in the process, taking into account the associated current prevention controls.” A review of all previous versions of the AIAG and VDA FMEA manuals will show this incorrect definition has always existed.  

When managing process risk, one must know the probability of the objectionable incident (i.e., PFMEA failure mode) due to the failure cause to determine whether additional action should be taken to reduce the probability of that cause (see question five). When examining the failure mode/failure cause combinations of most processes, the presence of a failure cause does not always result in a failure mode.

For example, worn tooling can cause but does not always cause defects. Consequently, knowing the probability of producing product with worn tooling present is not the same as knowing the probability of producing a defect due to the presence of worn tooling. If one were to use the probability of producing product with worn tooling alone, the potential risk of a defect due to worn tooling could be exaggerated.

Given that the occurrence rating and severity ratings of the PFMEA are the two critical parameters for defining risk and determining what should be worked on, the impact this one mistake has on the effectiveness of the PFMEA as a risk management tool cannot be overstated. 

There are several other FMEA paradigm-related critical errors in the AIAG-VDA Handbook. Among them is the incorrect definition of occurrence for DFMEA. To identify the existing FMEA paradigms that must be changed, the committee must determine what must be placed in each column in order to correctly answer the seven key risk questions. They cannot use the content from past FMEAs to determine the content requirements. If they do that, they should not be surprised to find that their current FMEAs are not compliant with the new column content requirements.

The greatest risk to properly defining required FMEA content

The greatest barrier to the AIAG-VDA FMEA Handbook committee properly defining what the FMEA content must look like are the current tools, such as structure analysis, function analysis, failure analysis, block/boundary diagrams, and P-diagrams, used to develop the FMEA content. There has been tremendous time spent on, and financial investment in, these tools. Although it’s possible to use these tools properly to define FMEA content some of the time, doing so can also lead to creating improper FMEA content even when the tools are used correctly. 

As an example, if one uses the VDA structure, function, and failure analyses to create DFMEAs and PFMEAs, it’s likely that incorrect root causes will appear in both documents, thus limiting their value as risk management tools (see question three). One can find instances of incorrect root causes in the AIAG-VDA Handbook’s DFMEA and PFMEA examples that were created using the proposed VDA method. This could be prevented if the handbook required root causes to be entered in the forms.

If proper definitions are defined for the DFMEA and PFMEA columns, and the current tools used to populate them are assessed for their ability to properly populate the forms, weaknesses in the current methodologies will be exposed. This will be difficult for many to accept due to adhering to the methodologies over the years. However, if the definitions of allowable FMEA content include any and all content derived from these tools, then optimizing the value of FMEAs as risk management tools will never be achieved. When defining what must be put in each column of the FMEA handbook, the committee must never forget that placing the proper information in each column is more important than supporting the use of a tool used to create the information. In all cases, when defining required FMEA content, properly answering the seven key risk management questions must be the priority.

What should be done with existing FMEA development tools

It is acceptable and advisable for any FMEA manual to identify types of methods that can be used to arrive at the proper entries for the DFMEA and PFMEA columns. I would recommend that the explanations of the methods be placed in an appendix. If a tool can create both proper and improper entries when used as intended, this issue must be explained in the manual. 

If people want to use the VDA structure, function, and failure analyses and by doing so they can get to the correct answers to the seven risk management questions, let them. If someone else wants to use block/boundary diagrams to help answer the seven risk management questions and they can answer them correctly, let them. The same goes with P-diagrams and Ishikawa diagrams. If the answers to the seven key risk management questions are correct, how you get there is not important. 

The required content of the DFMEA and PFMEA columns must be strictly controlled. However, it is equally important not to require the use of any one method to arrive at the proper FMEA content. The handbook’s requirement that the VDA method be used is troubling. Anyone who understands the VDA method knows that using an Excel spreadsheet to implement it is most likely impossible except for the simplest products. The handbook committee understands this. The AIAG-VDA Handbook states in bold letters: “When products and processes are complex, it is recommended that specialized software be used to apply the FMEA method.” Companies will be forced to purchase software to implement a FMEA methodology that prevents the optimized implementation of DFMEAs and PFMEAs.  

Summary

AIAG and VDA harmonization between certain aspects of DFMEA and PFMEA implementation can and should be done. The goal of the harmonization effort should be to improve the ability of companies to use DFMEAs and PFMEAs to effectively and efficiently answer the seven key questions required for risk management. Any changes being considered for harmonizing the AIAG and VDA FMEA methodologies that do not support this goal should not be implemented.

Discuss

About The Author

Richard Harpster’s picture

Richard Harpster

Richard 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.