How to Perform an FMEA in MATLAB - MATLAB & Simulink
Video Player is loading.
Current Time 0:00
Duration 5:13
Loaded: 3.17%
Stream Type LIVE
Remaining Time 5:13
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected
    Video length is 5:13

    How to Perform an FMEA in MATLAB

    Failure Mode and Effects Analysis (FMEA) is a common safety analysis methodology to identify and address potential failures in safety-critical systems. Let’s say your team has sketched out some initial entries in an FMEA on a whiteboard and then tabulated the data in a spreadsheet tool such as Microsoft® Excel®. You probably made some assumptions, such as a failure mode’s detectability. How can you validate these assumptions? In this demonstration, you will learn how you can use the power of MATLAB® and Simulink® to validate assumptions and automate steps in your FMEA.

    Published: 29 Feb 2024

    Hi, everyone. This is Pat. I'm a product manager at MathWorks. In this video, you will learn how you can use the power of MATLAB and Simulink to validate assumptions and automate steps in your FMEA.

    Let's say your team has sketched out some initial entries in an FMEA on a whiteboard and then tabulated the data in a spreadsheet tool like Microsoft Excel. You probably made some assumptions, such as a failure modes detectability. How can you validate these assumptions? This is where the Safety Analysis Manager, a feature in Simulink Fault Analyzer, can help.

    Here's a quick example workflow using one of our shipping examples of an aircraft elevator control system. See the link below for more information on this example.

    Here is a spreadsheet document opened in Safety Analysis Manager, containing data for a partial FMEA of the aircraft elevator control system. Each row represents a failure mode and a specific effect we want to analyze.

    There are columns for Function Name, which is the component; Failure Mode; Failure Rate; Flight Phase, which is an enumeration; Failure Effect; and Detection Method. I have also added columns for Severity, a value from 1 to 10; Occurrence, also a value from 1 to 10, which is proportional to the failure rate; Detectability, also from 1 to 10; and the Risk Priority Number, or RPN.

    The occurrence and RPN are derived columns. A derived column in a Safety Analysis Manager spreadsheet produces values derived from a formula. The RPN, for instance, is the product of the severity, occurrence, and detectability.

    You can add links to cells in a spreadsheet. This can be helpful for managing traceability between safety artifacts and automating workflows. In this example, I've added links from each failure mode to a fault I can simulate. And each detection method has been linked to a conditional. For more information on faults and conditionals, check out the Simulink Fault Analyzer documentation.

    Each row of an FMEA spreadsheet contains assumptions we make based on our engineering judgment. How do we know these assumptions are correct? For instance, how do we know if the detectability of a failure mode is correct? We can simulate the failure mode, model as a fault, and log the detection method during simulation. We can then use the simulation results to determine whether the detection method correctly worked.

    We can do this by leveraging two important features of the Safety Analysis Manager. The first is the ability to link cells in a spreadsheet using Requirements Toolbox. We can use the link information to automate many different workflows. And the second is the ability to run callbacks on the spreadsheet. This comes in handy when we want to automate a given analysis.

    Here is the workflow we want to automate. For each row in the spreadsheet, one, use the link information for the failure mode cell to automatically activate the associated fault for fault simulation. Two, use the link information for the detection method to automatically log the associated conditional.

    Three, run a fault simulation with the associated fault active and the associated conditional logged. Four, parse the simulation results to determine whether the detection method worked correctly. And five, annotate the detection method cell with a green check mark if the detection method worked correctly or a red error flag if otherwise.

    We can automatically perform this workflow by clicking on Analyze Spreadsheet. When I click on Analyze Spreadsheet, the analyze function callback runs. This callback runs a separate MATLAB script which performs the workflow I just described. Let's click on Analyze Spreadsheet.

    These green check marks on the detection method cells indicate that the detection method worked correctly. The Safety Analysis Manager provides three primary benefits-- the ability to link safety artifacts, such as an FMEA, spreadsheet to various artifacts in the MATLAB ecosystem, such as requirements, model elements, and test cases. This uses the Requirements Toolbox product.

    You can use the power of MATLAB while performing safety analysis through the use of analysis callbacks, which can automate parts of your workflow and check your safety artifacts for issues. And finally, you can validate assumptions using simulation results in Simulink, including the use of fault injection simulations to model failure modes using Simulink Fault Analyzer.

    You can import data from an FMEA spreadsheet in Microsoft Excel into Safety Analysis Manager, then export results back to Microsoft Excel. For more information, check out the Simulink Fault Analyzer Essentials video series.