How To Properly Perform DFMEA & PFMEA [Practical Examples Included]

Thomas Edison once said, “I have not failed. I’ve just found 10,000 ways that won’t work.” He understood that there is value in learning from failure. And lucky for us, over the many years since he made that observation, DFMEA and PFMEA were developed to help us learn from failures before they even happen.

Design FMEA and Process FMEA are fairly complex technical processes, but we will explain in simple terms when and how they are used. We will finish our step-by-step explanations for each methodology with a practical example.

Part of the FMEA family

FMEA stands for Failure Mode and Effects Analysis. It is a step-by-step way to identify all possible failures (in a certain design, product, process, or service) and study their effects. Then, it helps you take steps to mitigate or reduce their negative outcomes.

Many organizations use FMEA as a tool to put their designs and processes under scrutiny. It serves to do three key things.

1) Anticipate what may go wrong

Nothing will run smoothly 100% of the time. Anticipating failures is the first step in being prepared for them or even creating an action plan to prevent them altogether.

2) Predict how it will impact production

Once you have a list of all possible failures, they may seem equally undesirable. Looking at the potential effects of each will soon prove that they are not. Some would be much worse than others — you’ll have to do the risk assessment. Comparing the impacts of each failure can help you prioritize them and decide which are most important to prevent.

3) Evaluate your options

Preventing all failures is not possible. Nor is it worth the effort. FMEA requires a balance between prevention and mitigation. The criticality of some failures is important enough to prevent them at all costs. Others simply need a plan or fix for when they do occur. FMEA helps you tell the difference.

The process of FMEA is mostly the same no matter what you are evaluating. But there are two common types of FMEA: Design FMEA, and Process FMEA. While they are alike, there are some key differences in the way you would evaluate a product design than you would a business process.

While they can each be done on their own, DFMEA and PFMEA are often used together as a part of the Advanced Product Quality Planning (APQP) process. We won’t go into APQP in detail here, but it is a five-stage process that helps you make sure a new production process will result in quality products that meet quality standards.

You can use DFMEA and PFMEA when you are introducing a new product to the market. DFMEA makes sure you have a quality product. PFMEA makes sure that the process you use to make that product is efficient and reliable.

If you are a manufacturing organization that always strives to get better, you may be using methods like six sigma or TPM. If so, DFMEA and PFMEA are excellent tools you can use to supplement that work. Read on to learn how and why.

DFMEA vs. PFMEA

Despite how it may sound, “D or P FMEA” is not just an alphabet version of the song “25 or 6 to 4” by the band Chicago — although its meaning may feel just as elusive. If it feels like it is too complex to understand, we’ll tell you right now (spoiler alert!) that it isn’t. While there are many steps involved, with the right people and tools, FMEA is easy to navigate.

Many industries can benefit from using FMEA, particularly those where risk reduction and failure prevention are crucial, including:

• Manufacturing
• Software
• Business processes
• Healthcare
• Service industries such as hospitality
• Regulated industries such as cosmetics or pharmaceuticals

The difference between the two lies in the types of failures being analyzed — design (product) versus process.

However, just because they are two processes with different names, doesn’t mean they aren’t closely linked. If you change the design of an existing product, the process also has to be reevaluated and vice versa. Some risks will not be found during DFMEA, but rather during PFMEA and need to be accounted for in both.

There are different steps in the analysis process for each. However, the overall arc of the analysis is the same:

1. Review design or process: Review the current state of the product or manufacturing process and all steps involved.
2. Assess failures: Identify failure modes and their causes, effects, and severity.
3. Assess controls: Assess current methods of prevention and detection and assign occurrence and detection rankings.
4. Calculate risk: Calculate your risk priority number (RPN).
5. Make improvements: Determine corrective actions and implement them.
6. Reevaluate: Re-analyze your RPN and continue improving over time.

We’ll break down the specific steps in detail for both Design and Process FMEA later. With this methodology, you’ll start thinking differently about your failures in no time.

What is DFMEA?

DFMEA stands for Design Failure Modes and Effects Analysis. Many industries worldwide adopt it as a qualitative tool that puts their designs under the microscope.

Key decision-makers then use this information to improve product quality by making recommended modifications before mass production. However, not all “negatives” of a design will be corrected. In some cases, the design flaws are reworked to mitigate the likelihood of a failure, but not eliminate it.

With DFMEA, material properties, geometry, regulatory requirements, tolerances, interactions, and how the item behaves in its environment are all things that need to be considered. When conducted properly, DFMEA will improve the performance and safety of the final product.

Why use DFMEA?

Quality issues with the design of your product can cause a wide range of problems. Issues that lead to safety hazards can damage your reputation beyond repair. Minor quality issues can frustrate customers. DFMEA can help you avoid problems early during the design process and avoid costly fixes down the road.

When to use Design FMEA

DFMEA is best done while the design is being developed. This allows methods to prevent and detect failures to be built into the design right from the beginning. (Imagine identifying and anticipating mechanisms of failure even before it ever happens!).

DFMEA should then be reviewed by quality engineers when issues arise, or when significant design changes are made. Repeating assessments like DFMEA are a core principle of continuous improvement.

Who performs DFMEA?

Because a lot of different information must be included in the DFMEA, you will need a lot of different team members to provide input. A cross-functional team is best where players bring a diverse set of skills and expertise to the table. But as with any team approach, there must be a leader.

A Product Design Engineer or someone in a similar role is a good leader for the project. Other areas that should be included in the process are:

• Design
• Testing Analysis Engineer
• Production
• Supplier Quality
• Product Quality
• Service and Logistics

In the end, you will need to choose the team that will bring the necessary input to the table at your organization. Make sure you have a broad representation of experts, and your DFMEA will be off to a good start.

Outline of DFMEA process

It is important to follow consistent steps in the DFMEA process. Each step will rely on the ones that came before it. Documentation of your work with a DFMEA worksheet will help you stay on course. We will use a worksheet to help demonstrate the process below.

DFMEA template example. Source: iQASystem

Step 1: Review the design

Review the design of your product to complete columns 1-3 on the worksheet. This includes items, functions, and requirements. Every item (broken down into system, sub-systems, and parts) involved in the design of your product serves a purpose — a function — and requires certain conditions to do its job. These are its design requirements.

Breakdown of system, subsystem, and component in DFMEA . Source: Burge Hughes Walsh

By breaking down each item that goes into your design, you are setting yourself up to take focused steps that will help each item do its job.

Step 2: Identify potential failures modes

Next, you will identify potential causes for failure for each item on your list (who doesn’t love a good brainstorming session?). This is documented in column 4 on the DFMEA worksheet.

There are five failure modes:

1. Full failure: The system or component is no longer working and needs to be replaced entirely.
2. Partial failure: There is still some functionality, but the system or component is not operating as it should.
3. Intermittent failure: Malfunction occurs on an irregular basis.
4. Degraded failure: Frequent usage leads to fatigue which weakens the functionality of an item.
5. Unintentional failure: Failure from one item affects another.

A component or system may have more than one failure mode. It is important to document each at this stage.

Different things can affect failure modes:

• Operating conditions: a look at the environment (hot, cold, humid, covered, or exposed, etc.)
• Usage: look at the function of the item
• Service operations: look at the likelihood of the item to be used or operated properly.

Step 3: List the potential effects of each failure mode

In this step, you will be documenting all possible effects that could result from each failure mode. No failure happens without consequences.

The effects of your failure modes will run the spectrum. A lightbulb failing after an electric surge would be minor. A property-damaging fire caused by a failed lightning protection system would be major. But here, we are listing what the effects of the failures might be.

What goes around comes around and nothing happens in a vacuum.

Step 4: Assign severity ranking

In this step, you will decide how severe the above effects are. When something goes wrong, how bad is it? This is where we evaluate how important they are to avoid. You’ll use a simple numerical rating for this.

• 9-10: Very severe consequences where regulatory and safety factors cannot be overlooked. Often results in a change of the overall design.
• 7-8: Loss or reduction in the primary function of the item.
• 5-6: Loss or reduction in the secondary function of the item.
• 2-4: Annoyance that doesn’t affect function.

Think about car recalls, for example. Remember back in 2018 when Toyota had to recall millions of Priuses because of a fire risk? That would be a 10. But a recall impacting the function of the car radio? Chalk that up to a 2.

Keep in mind that what might be considered severe on one project might not be severe for another. Work with all the members of your team to agree on severity ratings.

Step 5: Define the cause of the failure

If the numerical rating in step four seemed too vague, you are in luck. Defining the causes will take a little more detail.

For each failure, you will provide a brief description — as concise and direct as possible — of the root cause of the potential failure. (Learn how to perform root cause analysis in this post).

Rely on your team and work together to consider and list all potential root causes. If they aren’t listed here, they will not be resolved in the following steps (where the real fun begins).

Step 6: Assess current design controls

Let’s be real. The phrase “design controls” might sound fancy, but it is just another way to say “things you plan to do to avoid problems.” It’s no different than putting antifreeze in your car or having a diplomatic answer at the ready when your spouse asks if they sound good when they sing in the shower.

Design controls are the things you put in place to avoid the effects of the failures listed in earlier steps. They fall into two categories: prevention and detection (columns 9 and 11 on your DFMEA worksheet). They provide valuable information to guide corrective actions later on:

• Prevention control: measures taken to remove the causes of potential failures to prevent them from happening.
• Detection control: measures taken to detect potential failures so that they can be quickly resolved.

The aim of each of these is to avoid the effects of a failure — either by detecting it so that it can be quickly fixed or by preventing it altogether.

Step 7: Assign occurrence ranking

It often isn’t worth spending the time and energy avoiding a failure that is very unlikely to happen in the first place. Knowing which failures are likely and which ones aren’t is a key part of your Design FMEA process.

Using a numerical rating based on the criteria below in the image will help. Assigning an occurrence ranking will give you direction for the last few steps in the DFMEA process.

Occurrence rank designation. Source: Lucas Wayne Shoults

Step 8: Assign detection rating

Step 7 rates the likelihood of failures so they can be managed. This step rates how well we can detect failures when they do occur so that they can be fixed. This rating, documented in column 12 of the DFMEA worksheet, completes our assessment of design controls. Using a numerical rating based on the criteria below in the image will help.

Detection rank designation. Source: Lucas Wayne Shoults

Step 9: Calculate the Risk Priority Number (RPN)

You may have noticed that after all these ratings, you have quite a few numbers filling up your DFMEA worksheet. And where there are numbers, there is math.

The Risk Priority Number serves as a way to sum up all the detailed information gathered in this worksheet. It is calculated by multiplying the three ratings: severity, occurrence, and detection (which were covered in steps 4, 7, and 8).

Your RPN will be between 1 and 1000. The higher the RPN, the higher the risk involved. Likewise, a low RPN means low risk. The RPN gives you clear direction on prioritizing the steps you will take next.

Step 10: Determine corrective action

A high RPN means a high risk of experiencing a design failure (or a low chance but a very severe failure). How high of an RPN is too high? That is up to you and your organization:

• What is your risk tolerance?
• What resources do you have to update your designs to reduce your risk?
• What resources do you have to manage failures if they do occur?

An item with a low RPN does not mean there are no risks. And a high RPN does not automatically earn an item a top spot on your priority list. It is good practice to look at the design holistically to see where your RPNs can and should be lowered based on your organization’s resources and priorities.

Lowering your RPN can take a variety of forms such as:

• Revising the design to look for engineering guidelines that may have been missed.
• Developing better mechanisms for detection of failures.
• Outlining procedures to reduce the likelihood of failure.
• Starting from scratch with a new design.

As a guide, severity should only change if the failure has been removed – which may involve a complete design change.

To lower the occurrence rating, you’ll need to revisit the causes of potential failures. Assess each cause or mechanism carefully with a focus on improving your design controls. It is one of the ways to create more fault tolerant products.

To lower detection, you can implement more preventative measures bearing in mind that these might increase development/production costs.

Step 11: Assign, plan, and execute

Once you have decided on your corrective actions, the next steps can feel easy by comparison. But don’t get lazy now! A solid plan for executing your changes needs to be diligently followed.

Because they are closely intertwined, you will need to include changes that come about from your PFMEA that you’ll do next. Agree with your team on accountabilities and timelines to guide you.

That being said, things often won’t go according to plan. Use column 17 of the DFMEA worksheet to note the actual completion dates. This will help you keep track of where you were before the project execution and give you a frame of reference for future DFMEAs.

When implemented properly, these changes will help you create a more fault tolerant system.

Step 12: Reanalyze the RPN

So you’ve calculated your RPN and decided to take action to lower it. You have changed some aspects of the design to remove or lower the risk and effects of failures. It is time to reanalyze your RPN. Congratulations! Little by little, the actions you take will lower your RPM.

But if we have learned anything from manufacturing best practices, it is that the work of improvement is never done. You are still not satisfied. Do you continue redesigning to take your RPN ever lower?

There will come a point when your risk has become low enough that further decreases will not be worth your time and effort. You’ll need to carefully determine which risks are still worth pursuing. Your multi-disciplinary team will help.

At the end of the day, the goal is to have an optimal design with tolerable or no risks.

What is PFMEA?

PFMEA stands for Process Failure Modes and Effects Analysis. It is a way to anticipate failures caused by the process of your production. In PFMEA, you will evaluate the impact of process breakdowns so that you can ensure efficiency.

Why use PFMEA?

PFMEA helps keep your process in tip top shape, avoiding serious issues that can waste time and resources. It ensures that your process is effective and up and running as much as possible.

Many of our processes rely on the interplay of technology, equipment, staff, and environment. PFMEA anticipates potential issues and helps keep all the pieces in place.

When to use PFMEA

PFMEA is best done as a production process is being developed, after the product development phase. This allows methods to prevent and detect failures to be built into the process right from the beginning.

PFMEA can be used in many different scenarios:

• Before control plans are developed for a new or modified process.
• When a new process, step, method, requirements, or technology is introduced.
• When an existing process step is slated for improvements.
• When there’s a new way of implementing an existing process.
• During the execution of the process for quality control measures.

PFMEA should then be reviewed whenever inefficiencies or process breakdowns occur. It should also be reviewed when significant changes to resources, equipment, or the operating environment are made. Because the PFMEA and DFMEA impact one another, it is important to consider one when performing the other.

Who performs Process FMEA?

The structure of the team performing the PFMEA will be similar to DFMEA. The leader in this case will be an individual who oversees the PDN process. Other areas that should be included in the process are:

• Design
• Testing Analysis Engineer
• Production
• Supplier Quality
• Product Quality
• Service and Logistics

Each team member should be experienced in his field so they are able to understand how the analyzed process will impact their department and what are the limitations of the process from their perspective.

Outline of PFMEA process

Documentation and consistency is just as important for PFMEA as it is for DFMEA. Below is a sample worksheet that will guide you through the assessment.

PFMEA worksheet. Source: Quality One

Step 1: Review the process

Begin your FMEA by mapping out your process. You will end with a detailed flowchart that identifies what the process does and doesn’t do – a.k.a. process flow diagram.

The diagram is a visual representation of each step in your process, from input to output. Every step has a function, and a set of requirements in order to be completed well. Once the steps are laid out in your diagram, begin documenting the steps, their function, and their requirements in the worksheet.

• Function: the reason, purpose, or intended outcome of a step in the process.
• Requirement: the “what” or the inputs needed for a step in the process to be completed successfully.

Step 2: Identify potential failures modes

As in DFMEA, now we dive into assessing failures — potential process failures, in this case. For each line on your worksheet — representing each step in your process — you will identify how each item on your list could fail to fulfill its function.

We use the same failure modes as we do in DFMEA:

1. Full failure: The system or component is no longer working and needs to be replaced entirely.
2. Partial failure: There is still some functionality, but the system or component is not operating as it should.
3. Intermittent failure: Malfunction occurs on an irregular basis.
4. Degraded failure: Frequent usage leads to fatigue which weakens the functionality of an item.
5. Unintentional failure: Failure from one item affects another.

Step 3: List the potential effects of each failure mode

Failures in any step of your process can have a long tail and can impact the process several steps downstream. They can impact users, internal and external customers, timelines, and the list goes on.

System model of a failure. Source: Burge Hughes Walsh

These impacts should be consistent with any related impacts that were found in the DFMEA process. But no shortcuts here — overlapping effects should be documented in the PFMEA worksheet as well.

Step 4: Assign severity rating

The severity rating scale is applied to process failures as well. Use the experience and knowledge of your team to determine how critical a process failure may be.

• 9-10: Very severe consequences where regulatory and safety factors cannot be overlooked.
• 7-8: Loss or reduction in the primary function of that step in the process.
• 5-6: Loss or reduction in the secondary function of that step in the process.
• 2-4: Annoyance that doesn’t affect function.

In PFMEA, this classification helps prioritize failure modes. Because our process steps take materials, inputs, and regulations into account, some may require urgent attention.

Step 5: Define the cause of the failure

This is where we ask the age-old question, “how could this happen?”

For each failure, provide a brief description — as concise and direct as possible — of the root cause of the potential failure. Rely on your team and work together to consider and list all potential root causes. If they aren’t listed here, they will not be resolved in the following steps.

Step 6: Assess current design controls

If you recall, “design controls” are nothing more than the things you plan to do to avoid problems. And they take the form of prevention and detection in PFMEA too. They provide valuable information to guide corrective actions later on:

• Prevention control: measures taken to remove the causes of potential failures to prevent them from happening.
• Detection control: built-in measures taken to detect potential failures so that they can be quickly resolved.

The aim of each of these is to avoid the effects of a failure — either by detecting the failure so that it can be quickly fixed or by preventing it altogether. Work with your team to discuss hypothetical “what if” scenarios to help you determine whether prevention or detection are the appropriate approaches for each potential failure.

Step 7: Assign occurrence ranking

It isn’t worth spending the time and energy avoiding a failure that isn’t likely to happen in the first place. Assigning an occurrence ranking will give you direction for the last few steps in your PFMEA. But process failures can be tricky to estimate. Here are some guidelines to help you estimate occurrence rates.

• 1: Prevented through product/process design
• 2: 1 in 1,000, 000
• 3: 1in 100,000
• 4: 1 in 10,000
• 5: 1 in 2,000
• 6: 1 in 500
• 7: 1 in 100
• 8: 1 in 50
• 9: 1 in 20
• 10: 1 in 10

Step 8: Assign Detection Ranking

Completing the assessment of process controls is again, the detection ranking. The criteria for rating your ability to detect process failures is slightly different than for DFMEA, but it will play the same role in the next steps.

Step 9: Calculate Risk Priority Number

Your process RPN is also calculated by multiplying the three ratings: severity, occurrence, and detection.

Again, your RPN will be between 1 and 1000. The higher the RPN, the higher the risk involved. The RPN gives you clear direction on prioritizing recommended actions to address the failure.

Step 10: Determine corrective action

A high RPN means a higher risk of experiencing that process failure (or that the failure will have severe negative effects on the system). Your organization can apply the same threshold to your process RPN that you applied to your design RPN.

Bring the same nuanced approach to prioritizing this work as you did with your DFMEA corrective action.

Step 11: Assign, plan, and execute

Any changes that you make as a result of your PFMEA should be planned in tandem with those from your DFMEA (if you are performing them in tandem). Agree with your team on accountabilities and timelines to ensure changes are implemented.

Step 12: Recalculate your RPN

Your PFMEA and DFMEA changes have been made. Reanalyzing your RPN will help you evaluate the effectiveness of your changes. It will also keep you in touch with your organization’s tolerance for risk, and help you continue to improve.

As we said before, the goal is to have an optimal process with tolerable or no risks.

Real-life Design and Process FMEA

Now it’s time to apply the steps to an example: the lightning protection system (LPS). Using the diagram below, we will perform the DFMEA for a couple of items as an example: item 1 (the ESE Lightning Rod) and item 3 (the Down Conductor).

Lightning arrestor system. Source:Clouds Power Systems

The documentation of the steps in the process will look like the following in our DFMEA worksheet.

DFMEA for the capture device and the down conductor.

We can see from the table above that the RPN is 800 in all cases. If we had previously decided that our RPN threshold was 200 for example, then we would have to find ways to reduce our failure modes by looking at the severity, occurrence, and detection of each item.

The PFMEA, on the other hand, will evaluate the process of maintaining our lightning protection system. In this worksheet, we can see the different steps that go into the process, their potential failures, and effects.

PFMEA (steps 1-7)) for the process of setting up a lightning protection system.

Below is the summary of the control methods and severity, occurrence, and detection rankings, along with the associated RPNs.

PFMEA (steps 8-10) for the process of setting up a lightning protection system.

There is clear variation in the RPNs on our PFMEA. Using that information will help us prioritize, and create action plans for improvement and elimination of risk.

[limblequote]As all people in technical fields know, theory is not worth much if it doesn’t translate into practice. Your process might rely on visual inspections and other NDT methods to detect a specific failure mode. If so, maintenance teams need to have a strict preventive maintenance schedule to follow. CMMS is here to help you coordinate those activities and ensure your PFMEA and DFMEA efforts are not wasted.[/limblequote]

Common DFMEA and PFMEA mistakes

When it comes to DFMEA and PFMEA, it is important to be thorough. The more accurate your data, the better your results will be. Here are some common mistakes you’ll want to avoid in these processes.

1. Misunderstanding the scope and objective

You know what they say: “Beware of the scope creep.” Ok, maybe no one says that. But staying laser-focused on the scope of your FMEA will help you stay on track and on time. Allowing the scope to creep into other areas — other products or processes — will bog down your analysis, and muddy your results.

2. Skipping necessary steps

All steps are equally important. Skipping any of them can deprive you of valuable analysis and context, leaving you to set priorities with less than complete information.

3. Overlooking items in the design or steps in the process

In our example of PFMEA, you can see that many of the detection control methods are listed as “none.” You may find yourself in a similar position as you set out to conduct your first FMEAs. It is important to be thorough — don’t avoid documenting potential failures just because you do not yet have control methods or vetted effects to document.

Getting them on paper is the first step to filling in the gaps.

4. Ranking criteria too closely

If every effect on your list has a severity of 10 (or 1 or 3 or 8) it will be very difficult to prioritize your work. It is helpful to take a look at the big picture and use all the effects on your list as a reference point. The impact of lost time may feel severe, but compared to a failure that causes loss of property or equipment, a little lost time will probably fall lower on the priority list.

Don’t worry. You aren’t ranking your best friends or judging a cute puppy contest. No one’s feelings will be hurt by high or low ratings. They should run the spectrum. If they don’t, you’re either missing some, or you are not being forthright in your ratings.

Thoroughly completing one step at a time can also help. It will ensure you have all the information you need from prior steps so that you can compare your effects as you rank them.

5. Missing linkage between Design and Process FMEA

A process failure can impact the product design, and a potential design failure can impact the production process. Failing to take the link between process and design into consideration will undermine any analysis and corrective action you take.

6. The ranking is not correct

All team members should be involved in ratings. They will each represent their areas with a valuable and unique perspective on effects and severity. Working toward consensus on ratings will give you a thorough, well-rounded, and accurate analysis.

7. Mixing of Prevention Control Method and Detection Control Method

The line between prevention and detection is a fine one. Which control method is a prevention method and which one is a detection method?

8. PFMEA or DFMEA does not reflect recent defect or customer reject

Use your experience. Any and all failures you have overcome — and their solutions — should be included in your FMEAs. And consider your FMEA as something that evolves and grows more complete over time. Every new failure that you did not anticipate or account for should be added. After all, there is no teacher like experience.

Improve the accuracy of DFMEA and PFMEA with Limble

A primary function of CMMS software is tracking equipment failures. Limble CMMS can help you easily track breakdowns in real time, creating historical data that can bolster the accuracy of your FMEAs. With minimal setup, Limble can track useful info such as:

• Total downtime for specific machines/failures
• Asset history
• MTTR & MTBF
• Costs associated with each failure

Long story short, Limble can streamline the review and assessment steps of DFMEA and PFMEA by giving you easy access to historical failure data. Learn how by starting a free trial, request a demo, or even try out our online self-demo.

It can be your own personal Edison, silently logging all the ways that “don’t work.”

via GIPHY

Keep in mind that, despite their many advantages, Design FMEA and Process FMEA will not solve every failure. If that was the case, maintenance managers and reliability engineers would never have to concern themselves with failure metrics.

If you can handle a somewhat steep learning curve, these are invaluable risk assessment tools that can have a big impact on your bottom line.