One negative experience can cost a customer.
We’ve all heard that saying, and nothing better proves the point than the 2016 Samsung Galaxy Note 7 product recall. This was an error that cost the company $5.3 million. 96 reports were filed to the U.S. Consumer Product Safety Commission, stating that the devices were plagued with overheating batteries, meaning the products were dangerous and posed a potential fire risk.
In this FAT FINGER article, we’ll be taking a look at how you can prevent costly – and unsafe – product errors using Design Failure Modes and Effects Analysis (DFMEA). We’ll explain what DFMEA is, before giving you a 10-step action plan on how to implement this business framework. Our aim is to help you consistently ensure high product quality for your business.
- What is DFMEA?
- The DFMEA process in 10 simple steps for quality control
With that said, let’s jump straight in!
What is DFMEA?
DFMEA stands for Design Failure Modes and Effects Analysis. DFMEA is an industry-wide qualitative tool that puts systems, products, or processes in the spotlight to identify errors, and devise actionable solutions to mitigate them.
DFMEA definition
To explain further, let’s break down the acronym.
DFMEA:
- Design: This is used when redesigning a product or service.
- Failure Modes: The different ways in which the failure modes occur in the design are studied and analyzed.
- Effects Analysis: The consequences of the identified failure modes across different system levels are studied.
The purpose of the DFMEA process is to improve product quality. It’s especially effective when products are produced in large batches, where the recommended modifications are made before product mass production.
In instances where the identified error cannot be removed, the system error is mitigated instead. Overall, the safety and performance of a product are extensively improved.
Why use DFMEA?
To ask why should we use DFMEA is really asking why should we ensure product quality? And we both know the answer to that.
According to author Crawford C.Merle, the product failure rate across industries averages at ~35%.
The direct costs of poor quality are easy to identify, such as labor, rework, disposal of material, and recall costs. But there are other indirect costs businesses can fail to address, such as excessive overtime, warranty costs, returns, and brand reputational damage. According to Cority, poor quality can negatively impact 20% of sales. As COPQ says, if a company makes $100 million, then poor quality can cost ~$20 million.
DFMEA is a widely acclaimed system used to identify product errors early in the design process. As such, costly quality defects are prevented later down the road.
The development of DFMEA from FMEA
When thinking about the DFMEA process, you may have come across another acronym, FMEA. FMEA stands for Failure Mode Effects Analysis. Think of FMEA as a framework that’s applied to a broader scope of operations. FMEA uses an analytical approach, one that’s designed to be used by cross-functional teams. The aim is to identify failure in a myriad of applications.
It’s from the FMEA approach that DFMEA came about. That is, DFMEA is a type of FMEA (as highlighted below). The primary types of FMEA are:
- System/Functional FMEAs
- Design FMEAs
- Process FMEAs
- Service FMEAs
- Software FMEAs
- Manufacturing FMEAs
Of course, in this FAT FINGER article, we’ll be focusing on the Design FMEA.
Organizations will perform DFMEA in the following scenarios:
- Before launching a new product, process, or service,
- Before deploying existing strategies or products in new capacities,
- Before developing a system control plan for both existing and new processes,
- To make improvements to an existing process, product, or service,
- To address a routine complaint about a product.
With that said, let’s take a look at just how you can apply the DFMEA process to ensure product quality control.
The DFMEA process in 10 simple steps for quality control
Below we’ve split the DFMEA process into 10 steps, with each step having a distinct purpose and focus.
Note that each stage will be completed at a different time within the product’s design timeline, and not all at once.
Step #1: Review the current product design and define the product’s function
You want to be able to identify each component and interface of a given product. This means you need to review the product’s blueprint or schematic. Identify each product part, how they relate, their intended function, and the product’s function as a whole. (Consider using the lessons learned from previous design reviews).
Once you’ve done this, you’ll need a way of measuring that identified product function, so write down requirements – or measurements – that will describe how well the product is delivering the intended output.
Step #1 will assure your team members are familiar with the product and its design.
Step #2: Determine the potential failure modes
Once you’ve identified a product’s key parts, the next step is to figure out how those parts may fail to perform their intended function (including failures related to warranty and field errors). Refer to existing documentation and data for clues.
Note that a given product part will have more than one failure mode. It’s important not to take shortcuts and to be as thorough as possible.
Step #2 uses a multi-disciplinary team approach. Gather your team and conduct a brainstorming session while reviewing documentation for clues about the failure mode. Review customer complaints, plus warranty, health, safety, scrap, and damage reports.
On top of all this, have your team consider what could happen to the product – or product part – under adverse conditions, and when interacting with other products.
Step #3: List the failure effects
For each identified failure, there will be at least one effect. An effect is the impact of the failure, e.g. some errors will lower customer satisfaction, delay operating processes, and others could negatively impact workers. Make sure to define the product effects in a way that’s meaningful to performance.
Step #4: Assign severity ratings to the failure effects
The severity rating describes the consequence of failure, that is, how serious an effect it would be if the failure occurred. When thinking about severity ratings, think about impact. A score from 1-10 is used to measure failure mode severity, as follows:
- 9-10: The failure mode effect is dangerous and will have damaging consequences. The failure mode must be avoided at all times, as not doing so will have significant safety and regulatory implications.
- 7-8: The primary use of a given product is lost due to the failure mode effect. It’s vital to remove this failure mode effect for good business.
- 5-6: The secondary function of a product is lost or degraded. It’s vital to remove this failure mode effect for good business.
- 2-4: The failure mode effect results in an annoyance – such as a rattle – or impedes the visual appearance of the product, but doesn’t affect the product’s function. It would be wise to remove this failure mode effect for good business.
Failure mode effects should be prioritized from the highest severity and down. Actions should be identified to change the product design process and remove the failure mode effects of higher severity.
Step #5: Assign an occurrence rating for each failure effect identified
An occurrence rating measures how frequently the failure defect will occur. To determine the occurrence rating we first need to know the potential cause. By knowing the cause we can identify how frequently a specific failure mode will happen. Ranking failure modes according to their level of occurrence uses a degree of estimation based on known, past data.
Like with the severity rating, the occurrence rating is assigned on a scale from 1-to-10. An occurrence ranking of 10 defines a failure mode as highly likely to occur, and conversely, an occurrence rating of 1 means the error is unlikely to occur. Use the list below to guide you when assigning occurrence ranking to your identified failure modes.
- 1: The failure mode is prevented by using a known design standard.
- 2: An identical or similar design is used that has no history of failure.
- 3-4: Failure modes are isolated (temporary).
- 5-6: Occasional failures have been expressed in the field or during the development/testing process.
- 7-9: The product has a new design but one that uses well-known technology.
- 10: The product has a new design and is using new technology.
Sometimes the complexity of the identified failure mode may be so great that a single measure for occurrence is ineffective. In these cases, you can measure occurrence from a time-based, event-based, and piece-based perspective. Decide how you will measure occurrence for the greatest accuracy.
Step #6: Assign ranking to reflect the chances of detection
Categorizing potential defects is pointless without considering how you will detect them. This is where the detection rankings come in. A detection ranking determines the chances that the failure mode will be detected prior to the customer finding it. Again a scale running from 1-to-10 is used here.
As you might have guessed, 10 means the chance of detecting the defect is absolutely uncertain, and 1 means the chance of failure detection is certain – usually through product design solutions, or by following design or material standards.
There are two ways to identify failure modes, and that is by using either prevention or detection controls. As the name suggests, prevention controls prevent the cause or mechanism of failure or the failure mode itself from occurring. Detection controls detect the cause or mechanism of failure, or the failure mode itself once the error has happened (before the product is released from the design stage).
Step #7: Calculate the Risk Priority Number (RPN)
Once you have measures for severity, occurrence, and detection, the next step is to calculate the Risk Priority Number. This is done by multiplying the latter values together, that is:
RPN = Severity x Occurrence x Detection
The RPN number is a means of communicating the level of risk. The higher the RPN number, the more concerning the failure defect is. Since the three relative rating values range from 1-to-10, the RPN number will range from 1-to-1000.
Step #8: Develop a plan of action
Once you’ve defined your RPN value, you need a plan of action to bring that value down. Call this your action plan.
As you’ve probably inferred, you can reduce the RPN value by lowering the three rankings, severity, occurrence, or detection:
- Reducing the severity rating of the failure mode is the most difficult task, as this will usually require a design change.
- Reducing the occurrence rating is accomplished by removing or taking control over the potential causes of the failure.
- Reducing the detection rating is accomplished through the addition of detection controls or improving the means of preventing the error in the first place.
It’s up to a specific organization to decide what RPN value is acceptable for them.
Step #9: Take action
Once you’ve developed your action plan, it’s time to institute it. Sometimes a fairly easy solution is all that’s needed to remove or mitigate the failure mode. For more complex solutions, consider using a PERT or a Gantt Chart to keep your plan of action on the track. Be sure to state who is responsible for solution tasks, what their task is, and when the solution tasks must be executed.
Step #10: Calculate the final RPN value
Once your action plan has been put into effect, you want to determine the success rate of this. Re-evaluate each of the potential failures and determine their new RPN value. This will mean reassessing the severity, occurrence, and detection ratings. If the final RPN value has been reduced, then you know your action plan achieved the desired aim.
Use FAT FINGER as a product failure prevention control
FAT FINGER is a process documentation software. You can use this software to detail product design steps as an actionable checklist. This way, you can introduce prevention failure steps in the documented product design process. Then, every time the product design process is run, by whoever, the failure prevention steps are executed as part of the overall procedure. Process documentation captures best practices and acts as a fail-safe tool.
Use FAT FINGER features such as Conditional Logic and Integrations to add the complexity you need to your documented operations. You can also use FAT FINGER’s Analytical Insights feature to measure how well the design process runs, and any other product-related operation, once the failure prevention controls have been introduced (or any other means of reducing the RPN value).
To learn more about FAT FINGER and how to get started, watch the below video.
You can sign up for FAT FINGER here, to start documenting your product design operations (and in fact any business operation) to consistently deliver best practices and high-quality, error-free outputs.