Failure data in a CMMS system are often messy. Most current practices prevent the reliability engineer from analyzing failure records in a structured way. 🧹
We should be able to link a failure recored to a pre-defined failure mode to make some inferences:
- Did we identify the failure mode in some prior analysis?
- Can we link the failure mode to a specific component?
- How can we use our newly obtained insights to make better decisions?
In the ideal world this information can be found on the work order. In a less ideal world, which means most of our current reality, a simple analysis procedure could be something like:
1. Create a Pareto chart for all failures for a particular system type from a certain period 📊
2. Use e.g. pivot charts to drill down a bit deeper in the asset hierarchy. Which subsystems suffered most failures? And which components? 📈
3. Now collect the actual text records for the top failure you have identified
4. Codify each failure in a few categories: failure mode, cause and remedy. 🗂️
Unfortunately, there is some manual labour involved in this procedure. But it is one way to get you started. If you have other ways of assessing your CMMS data, please share them!
Search by tags
Let's talk about the needs of your organization.
.png){kind=small}
