Failure Reporting and Failure Analysis

The Asset Manager Has Questions for His Senior Staff

What if the Data is Bad?

Bad data can be due to poor processes, unclear update roles, lack of accountability, lack of training, inadequate CMMS design, and systemic issues, i.e. poor culture. But this is stuff that can be fixed. The first step is to define the end game. And sometimes you need to go ahead and design the reports and run them to see how bad the data is. Sounds strange, I know, but only then will management understand the issues.

Improve Your Asset Management Knowledge

Tip:

You can improve your learning rate by becoming a student of asset management and pursuing certifications in reliability. And, of course, there is higher education courseware, e.g., Maintenance & Reliability Engineering.

The Best Path Forward

Definitions Matter

There are Different Forms of Failure Analysis, but One is Better Than the Other When Pursuing Recurring Failures

Capture the Right Failure Data

Note: The assumption here is that leadership knows what the bad actor report looks like.

Failure Code Hierarchies are Obsolete

We live in an age of Google search. If you wanted a recipe for apple pie, you are not going to start with FOOD, then DESSERT, and then type APPLE PIE. No. Instead, you simply type APPLE PIE. The Google search algorithm works so fast it is mind-boggling. The CMMS should be set up similarly.

Point #2:

Every CMMS should recognize the importance of a failure mode. This is the language of RCM. Failure data should be recorded at this level.

Point #3:

The failure mode can be recorded as 3 separate fields: failed component, component problem, and cause code. These fields are standalone and do not need to be part of an equipment hierarchy – which would be monstrous.

Point #4:

The maintenance technician knows the part he worked on. In this case, it was the impeller [below picture]. He can enter this value into a validated field.

Point #5:

The reliability team wants to know the component problem - which is a separate field. Plus, if there are issues relating to human factors, then this is the place to capture that data.

Maintenance is Performed at the Component Level Which is Where Failure Analysis Needs to be

In Summary

As an advanced process, chronic failure analysis may be the most significant tool you have for optimizing O&M costs. Chronic failure analysis identifies recurring failures that by themselves may not be noteworthy, but because they are almost routine, their frequent occurrence impacts the bottom line. These are problematic assets that are underperforming, energy-inefficient, and costing too much to maintain. Thus, organizations would be wise to focus on ways to identify the worst offenders and eliminate the cause.

b.t.w. -- the sort I recommend is the SMRP #1 economic metric which is

Average Annual Maintenance Cost divided by Replacement Cost.

You may have your own favorite. It may be a good idea to run multiple sorts.