Continuing with The Bad Actor Report series, in this article we’ll take a deeper dive into the reasons why organizations fail to perform failure analysis.

If you missed my first post, you can check it out here  

There are a few different reasons why organizations fail to perform failure analysis:

• Many fail to leverage failure data from their CMMS due to bad data. 

• Some organizations simply do not believe the CMMS should be used for failure analysis. 

• In other instances, leadership teams do not believe the working level will consistently enter failure data due to a fear of being micro-managed by the data they enter. 

• Lastly, some administrators set up failure codes but they are too high a level to add value, e.g., “mechanical issue” which is too vague for analysis.

As it is, many organizations capture nothing. But at the other end you have root cause failure analysis, which is quite extensive. 

Which leads to my next point of what is the right level of detail for failure data? 

The recommended strategy is to find a middle ground, which helps the reliability team make more informed decisions but does not overwhelm the maintenance technician in terms of input. With this approach, the technician should be able to update a work order, including failure data, in about 5-10 minutes. The reliability team would then be able to meet monthly to run a bad actor report and manage by exception.

The question also pertains to the depth of failure coding. The technician only has control over the physics of failure. They know the component they replaced (or repaired) on that asset. They may also know the component problem.

For example, if the failed component is a bearing, they might also recognize that the bearing is seized, worn or damaged, whereby the cause code might be inadequate lubrication. However, there will be times when the cause is not known, and would ask for help. 

Once the failure analysis process moves from physical cause to human factors, more time is expended. 

Business rules should be in place that indicate when other roles need to get involved. If this failure does not require an RCFA but it is frequent enough whereby repair costs are adding up, then the reliability team might focus on recurring failures for critical equipment and drill down into the human factors.

The importance of linking failure data to the endgame: 

• Precision data is needed for precision reporting. For this design to be possible, it is important that the captured failure data support the endgame. The figure below shows the different types of failure data and how to move from subjective to objective in terms of decision making.