Design Failure Mode and Effects Analysis (DFMEA) is a cornerstone of product development and risk management. It is intended to be a living document: touched, updated, and referenced regularly throughout a product’s lifecycle. Yet, despite its importance and the number of engineering hours poured into it, the DFMEA has a reputation for tedium.
Performing the risk analysis is a highly collaborative and common practice, but reading a completed DFMEA, start-to-finish, is an infrequent activity. While regulatory requirements dictate the document is updated and audited, it’s rarely read as a cohesive narrative. Maybe it’s skimmed.
This shouldn’t be the case. Although a completed DFMEA can be tedious to read, it is rarely due to a lack of engineering skill, nor is it the fault of the FMEA template itself, which time has proven to be an incredibly robust and resilient tool. Rather, it’s often a large disjointed matrix, which is a byproduct of how the document is built. Populated over weeks or months, across dozens of meetings, by a rotating cast of cross-functional experts—all under the time pressure of launching a product—the final matrix naturally becomes disjointed.
Although each cell of the template holds vital data, the matrix rarely reads smoothly from left to right or row to row, and the overarching logic becomes fragmented. The raw data is all there, but the narrative of why a design decision was made, and how it mitigates a specific risk, gets lost in a choppy cadence.
Because the FMEA template is so robust, it is highly tolerant of this fragmentation. Even a disjointed DFMEA can still assist a team in developing a thorough and effective verification test plan, and it will pass an audit (where it is rarely read). I suspect this is why the template has been around for so long without significant change, and why teams often settle for a document that is “good enough for compliance.”
Why settle? Good writing requires good editing, so too does a highly effective DFMEA; it requires several dedicated editorial passes. We intuitively understand that a first draft of a technical manual needs to be edited for flow, clarity, and consistency. The DFMEA should be treated no differently.
Editing a DFMEA isn’t about changing the engineering data; it’s about smoothing out the logic so that a person without intimate, day-to-day knowledge of the design’s intricacies can follow the thought process. It is about ensuring that the progression from Function to Failure Mode to Recommended Action moves forward consistently.
A DFMEA edited for readability transforms from a static compliance record into a highly accessible, living document. It promotes project transparency and unlocks several distinct advantages.
First, it allows experts from outside the immediate project to contribute their knowledge. If the logic reading left-to-right is clear, an outside engineer can easily review the document and might recognize a critical omission. A specialist in corrosion or tribology, or a Sr level mechanical or software engineer from another project might spot an unusual failure mode that falls just outside the core team’s area of expertise. But if the document is an unreadable maze of internal shorthand, that outside expert will never be able to provide that vital catch.
Another reason is that readability ensures long-term traceability. Since it is highly likely that a product’s engineering ownership will transfer during its lifecycle, an easy-to-read DFMEA serves as the new engineer’s primary access point to the design’s history. Instead of reverse-engineering the previous team’s logic, the new engineer can immediately understand the design’s key attributes by reading the DFMEA.
Consider a post-market failure analysis report containing forensic evidence as to why a product broke in the field. These reports are often fascinating. They mimic detective stories where the logic is clear and the chain of events is laid out perfectly. Such reports are so compelling that they often form the basis of case studies, articles, and even television documentaries. Shouldn’t a DFMEA be exciting?
It takes effort to edit a DFMEA. It requires stepping back from the individual meetings and reviewing the matrix as a holistic document, but there are benefits. Taking the time to edit our DFMEAs for clarity and traceability is the best way to ensure that the most fascinating failure stories are the ones we prevented.
Glenn DiCostanzo, Founder & Principal
GD Consulting, LLC
October 2019