Failure Mode and Effects Analysis (FMEA) is a systematic, and most importantly proactive, method for evaluating a process or product to find out where, and how, it might fail. Instead of waiting for something to break and then reacting, an FMEA lets a team see the likely failures in advance, rank them by risk, and act on the biggest ones before they ever happen.
This guide covers what FMEA is, the step-by-step process with a worked example, how the Risk Priority Number (RPN) works, when to use it and where it fits in Lean Six Sigma, the difference between PFMEA and DFMEA, and the most common mistake teams make with it.
What is FMEA (Failure Mode and Effects Analysis)?
FMEA is a structured way to ask “how could this fail, and what would happen if it did?” before you commit to a product, process or change. It is proactive by design: you identify the ways something could go wrong, judge how bad each would be and how likely, and use that to focus attention on the risks that matter most. It is one of the core Lean Six Sigma tools, and it belongs to the wider Lean Six Sigma toolkit for improving quality and preventing errors and defects.
How to run an FMEA, step by step
An FMEA is done as a team exercise. To make the steps concrete, take a new product launch campaign as an example: before going live, the team holds a session and works through the following.
- Identify all the potential failure modes. Brainstorm every way the campaign could fail: lack of customer interest, poor product positioning, ineffective promotional materials, and so on. There are no wrong answers here.
- Assess the effects of each failure. For each failure mode, decide what the consequences would be: lower sales, negative customer feedback, damaged brand reputation.
- Assign a severity rating. Rate each failure mode, for example on a scale of 1 to 10, where 1 is negligible and 10 is utterly disastrous. The rating reflects the impact on the launch and the wider business objectives.
- Rate the likelihood of occurrence. Rate how probable each failure mode is, using the same 1 to 10 scale so the numbers line up.
- Assign a detection rating. Rate how easily the failure mode could be detected before it leads to significant issues, again 1 to 10.
- Calculate the Risk Priority Number (RPN). Multiply the three ratings together for each failure mode. For example, a severity of 6, an occurrence of 2 and a detection of 8 gives an RPN of 96 (6 x 2 x 8). The RPN is a single measure of the overall risk of each potential failure.
How the Risk Priority Number (RPN) works
The RPN is what turns a long list of possible failures into a priority order. Because it multiplies severity, occurrence and detection, a failure that is serious, likely, and hard to spot rises straight to the top. With the RPN in hand, the team prioritizes the failure modes with the highest numbers and develops plans to mitigate or eliminate the associated risks, rather than trying to guard against everything at once.
When to use FMEA, and where it fits in DMAIC
Use an FMEA to understand the risks of a solution before you put it in place. In a DMAIC project it sits in the Improve phase, right before you implement your solutions. By working out where a solution could fail before it fails in reality, you gain insight into the impact those failures could have and can take preventative action so they do not occur, or occur far less often. It is a Lean Six Sigma tool, not a Lean-only or Six Sigma-only one.
PFMEA vs DFMEA: process versus design
There are two main variants, and the difference is simply what you are analyzing:
- Design FMEA (DFMEA). The most traditional form, used to identify and assess potential failures in the design of a product, process or system.
- Process FMEA (PFMEA). Used to identify and assess potential failures in a manufacturing process, focused on the process steps and the interactions between them.
The most common FMEA mistake
The most common mistake with FMEA is not doing one at all, because the team is under pressure to get its solution to market. The result is that the rushed solution can create new defects nobody thought of, or produce unintended consequences that only show up later.
Sleep Number, a leading US wellness company, is a good example of doing it right. The team ran an FMEA to identify how a proposed product quality fix might fail. The analysis made clear that the solution would actually have made matters worse, so a more robust solution was chosen instead, which meant lower warranty costs and a better customer experience. The FMEA caught the failure on paper, before it could happen in the real world.
Putting FMEA to work
FMEA is one of the highest-leverage habits a team can build: a structured, proactive check that surfaces the failures worth worrying about and ranks them by real risk, before a single dollar is spent implementing a solution. It works best as part of a disciplined Lean Six Sigma approach, run by people trained to use it well.
The Lean Six Sigma Company runs in-company group training and full deployment support across the United States. If you want your team to run FMEAs that actually prevent failures, those are the right starting points.
Frequently Asked Questions
1. What are the 5 steps of the FMEA process?
Identify the potential failure modes, assess the effects of each, rate them (severity, likelihood of occurrence, and detection, each 1 to 10), calculate the Risk Priority Number by multiplying those three ratings, and prioritize the highest-RPN failures for action. The RPN calculation is often counted as its own step, which is why you also see it described as six.
2. How is the RPN calculated?
RPN = Severity x Occurrence x Detection. You score each of the three on a 1 to 10 scale and multiply them, so the highest scores flag the failures that are serious, likely, and hard to detect. For example, a severity of 6, occurrence of 2 and detection of 8 gives an RPN of 96.
3. Is FMEA Lean or Six Sigma?
FMEA is a Lean Six Sigma tool, used by both. In a DMAIC project it sits in the Improve phase, just before you implement a solution, so you can catch the ways that solution might fail before rolling it out.
4. What is the difference between PFMEA and DFMEA?
DFMEA (Design FMEA) analyzes potential failures in the design of a product, process or system. PFMEA (Process FMEA) analyzes potential failures in a manufacturing process, focusing on the process steps and how they interact. Same method, different subject.
5. When should you use an FMEA?
Use an FMEA whenever you are about to implement a solution or launch something and want to understand how it could fail first. It is proactive by design: catching failures on paper, before they happen, is far cheaper than fixing them after they reach the customer.
