Windchill Risk and Reliability Desktop Help > Windchill FMEA > FMEA Theory > The FMEA Process
The FMEA Process
A FMEA provides a systematic method for organizing and automating the study of a particular system or process to determine how it can be modified to avoid failures and improve overall reliability. The goal of performing a FMEA is to review the system or process to define ways in which a failure can occur and to indicate the resulting effects of the failures on your system. An inductive, bottom-up method is generally used to analyze the system design or process to evaluate the potential for failures.
Because the complete analysis of a complex system can quickly become hard to manage, FMEAs provide an organized framework that is vital to targeting potential design problem. For example, in the case of a complex system like an airplane, the overall analysis is likely to become very large. Typically, the approach for analyzing a complex system is to break the system into smaller, manageable subsystems. Oftentimes, different groups are responsible for analyzing the different subsystems. Results for these subsystem analyses are then “rolled up” into an overall system level FMEA.
The power in FMEAs is fully realized when the analysis is extended to include information relating to the risk of these potential system failures. The idea is to use the FMEA to assess which failure modes require effort to prevent, mitigate, detect, or ignore. This assessment of criticality lays the groundwork for an organized approach to risk management.
By using FMEAs to assign and categorize failure modes, the failure modes that are critical become easy to identify. Based on criticality results, you can assign defined plans of actions to the resulting risk categories. For example, you might flag high risk failure modes so that a plan to eliminate them can be formulated and implemented. Medium risk failure modes might require the design and installation of some type of detection mechanism. Low risk failure modes might require no action.
The complexity and number of steps involved in performing a FMEA are directly related to the type and level of FMEA being performed. However, all FMEAs require the steps described in the following table.
Step
Description
1. Define the system or process to analyze.
You create the system definition by breaking down the system or process into discrete elements. Complete system definition includes identification of internal and interface functions, expected performance at all hierarchical levels, system restraints, and failure definitions. You can break down the system definition to any level desired. For example, you can break down a large hardware system into a number of subsystems and assemblies. If desired, you can then break down assemblies into individual components or parts. For more information, see FMEA Types.
2. Obtain or construct diagrams to ensure that the system is fully understood.
Functional and reliability block diagrams that illustrate the operation, interrelationships, and interdependencies of functional entities involved in the use or operation of the system.
3. Identify all potential failure modes.
Determine all of the ways in which the items in the system definition can potentially fail. A failure mode is simply a definition of what can cause an item failure to occur.
4. Evaluate each failure mode to determine worst-case effects and assign a severity classification.
Beginning at the lowest level of the breakdown, evaluate each potential failure mode independently to determine the worst-case effects that this failure mode can have on:
• The immediate item or function
• The system
• The mission
5. Identify failure detection methods, corrective actions, and the effects of corrective actions.
For critical failure modes, you can identify failure detection methods and compensating provisions for if the failure mode occurs. You can also determine if design changes can eliminate a critical failure mode. If not, perhaps design changes can reduce either the likelihood of occurrence or the impact of the effects. You repeat this analysis until all potential failure modes pose an acceptable level of risk. The individual or agent authorizing the FMEA must clearly define what constitutes an acceptable level of risk. After this iterative process is complete, you would document problems that design changes could not correct in a summary.