A Failure Modes and Effects Analysis (FMEA) identifies potential failures within a system or process and classifies them by likelihood of occurrence and potential severity. It is then used to find solutions to reduce the chance of failures and control systems to discover failure modes before they happen or worsen.
Six Sigma teams use FMEA because it accurately predicts challenges and allows for removing problematic components from a process or system with the least amount of time and money spent. The object is to identify failure modes and failures effects. FMEA defines failure mode as a potential or actual defect or error in a system. A failure effect describes how a failure mode will impact customers or end users.
Quality and reliability engineers often use FMEA as a risk management tool to identify high-risk items based on consequences of failure and likelihood of occurring. FMEA can focus on process failures and design failures.
Steps to Take in FMEA
There are four overall steps in FMEA.
- Identify potential failures and defects
- Determine potential severity and consequences of each
- Predict likelihood of occurrence
- Create systems for failure detection
Identify Potential Failures and Defects
Teams determine this by analyzing each functional requirement for the process and identifying, based on similar processes and experiences from the past, where failure moods are likely to occur. Teams note the failure effects of each failure mode.
Determine Severity
Teams determine the severity of consequence from a failure mode. Teams typically develop a scoring system in this step. For example:
1: No effect
2: Very minor effect, noticed only by discriminating or very observant users
3: Minor effect, with only a small part of the system impacted
4-6: Moderate effect, with most users “inconvenienced and/or annoyed”
7-8: High effect, which involves loss of the system’s primary function, leaving users dissatisfied
9-10: Very high effect, meaning the process, system or product has become hazardous, leading to angry customers and safety hazards
Likelihood of Occurrence
Teams rank each failure mode by the likelihood of occurrence by assigning numeric value to the potential of each failure, with those ranked with a “1” as the least likely to occur and those ranked with a “10” as the most likely to occur. This step often includes conducting a root cause analysis to determine failure mode causes.
Create Systems for Failure Detection
Teams create or improve current process controls to detect potential failures before they happen. These include inspections, tests, and other mechanisms used to evaluate the system or process – from a team of people doing periodic checks of a system to automated processes that look for derivation from acceptable ranges.
The job of controls is to prevent a failure mode from happening or, at the very least, detecting a failure after it has happened but before it can affect the customer or end user.
Use: To assess risk, FMEA looks at three measures (frequency of occurrence, severity of consequence, and chance of detection) to create a Risk Priority Number (RPN) which functions as a numerical assessment of the risk of that failure mode. The tool is sometimes referred to as Failure Modes, Effects, and Criticality Analysis, or FMECA.
