DISCLAIMER: This blog post solely represents my own opinion based on experience and/or interpretations. As a former Subject Matter Expert Quality Risk Management and trainer in one of the pharmaceutical big 5 companies I have extensive experience in performing and guiding failure mode and effect analyses.
EPISODE – 1 In this series I will give an in-depth view on the different risk assessment techniques, starting with FMEA. The FMEA series will contain 5 episodes.
Failure Mode and Effect Analysis (FMEA)
FMEA is used to define, identify, and eliminate known and/or potential failures, problems, errors… from the system, design, process, and/or services before they reach the customer.
FMEA’s motto ‘ DO THE BEST YOU CAN WHEN YOU CAN‘.
The FMEA technique was developed by army reliability engineers in the late 1950s (9 november 1949 to be exact) to study problems that might arise from malfunctions of military systems.
Generally, four types of FMEA can be distinguished, they all gave their own focus and objective:
- System or concept FMEA: A system FMEA focuses on potential failure modes between the functions of the system caused by system deficiencies;
- Design FMEA: A design FMEA focuses on failure modes caused by design deficiencies;
- Process FMEA: A process FMEA focuses on failure modes caused by process or assembly deficiencies;
- Service FMEA: A service FMEA focuses on failure modes (tasks, errors, mistakes) caused by system or process deficiencies.
The quality of the product and/or services must be a company’s number one priority in order to achieve customer satisfaction (focus of ISO, FDA and other standards/legislation) by reducing known or potential problems. FMEA will map the way to continual improvement.
Potential benefits are:
- Helps to select the best design;
- Helps in determining redundancy;
- Increases the likelihood that potential problems will be detected;
- Establishes a priority ranking for implementing improvements;
- Helps identify and eliminate potential safety concerns.
A FMEA is best started as early as possible, as soon as a reasonable amount of information is known about the product, process, service, design… the FMEA should be started. Typically an FMEA is started as soon as the project was approved and a preliminary design was approved. You should never wait for all the information to be available.
An FMEA process should start:
- When preliminary design of new products, processes, systems, or services are finished;
- When existing products, processes, systems, or services are changed (update existing FMEA(s) or start FMEA(s) when there is none available);
- When new technology is used/found for existing products, processes, systems, or services;
When improvements are considered.
- After the FMEA process is started, the FMEA documentation becomes living documentation meaning it’s a true dynamic tool of improvement.
An FMEA is never truly finished, the documentation will become static when the system, process, product, or service is discontinued.
Information may include:
- Flow Charts;
- Error or incident data;
- CAPA (Corrective and Preventive Actions);
- Post market surveillance data.
The primary output of FMEA is a list of failure modes, the failure mechanisms and effects for each process step. Information is also given on the causes of failure and the consequences.
Outputs may include:
- List of potential failure modes ranked by risk probability number;
- List of recommended actions to avoid the potential failure modes;
- List of recommended actions that could identify the potential failure modes on time.
FMEAs should be performed by a cross-functional and multidisciplinary team and cannot be performed on an individual basis.
A multidisciplinary team is best comprised of, but not limited to:
- Project Manager;
- Quality Representative;
A team is always defined for a specific project and there cannot be a predefined universal team that will perform all assessments in a company.
The FMEA process is a time consuming process but it’s worthwhile. There are no specific length or time limits, the length and time is defined by the experience of the team, objectives, and complexity of the process, product or system in scope.
In my experience it’s best to organize short sessions of maximum 3 hours with a short break included. People tend to lose their focus after one hour and half. I’ve been in 8 hour long sessions and I can only state that I’ve achieved more in the 3 hours sessions than I ever would in a complete workday session.
Next Episode (2/5) will focus on the FMEA process itself.
One Reply to “Failure Mode and Effect Analysis – FMEA”
Righht here is the riցht blog for ɑnybody who hopes to understand tһis topic.
Yоu knoѡ ɑ whole lot its almoѕt hard tο argue wіth уou (not that I ɑctually would ᴡant to…HaHa).
Yоu ceгtainly put a brand neԝ spin on ɑ subhject ᴡhich has Ьeеn ԁiscussed ffor
decades. Wonderful stuff, јust excellent!