Why Failing Can Be Good and What You Can Take Away from It.
Regardless of how good a maintenance & reliability program is set up and managed, there will be failures. This is partly due to the maintenance program itself, where the focus is on the consequences of the failures, not the failures itself. This approach allows most organizations to manage large facilities will a minimum of staff and cost.
But what should happen when something does fail? Should we just carry on as usual since we avoided the consequences? Absolutely not. When a failure occurs, we need to learn from it and improve the maintenance & reliability program. Yet, many organizations address failures by implementing a PM routine. This is not the right approach. Remember only 11% of failures are age-related. Adding these PM routines to the program will cause a collapse of the program from too much work, not to mention the maintenance induced failures that result from it.
So what should happen? The failure should be analyzed and actions implemented to reduce the chance of the failure occurring again.
What to do When a Failure Occurs
When a failure does occur in your operation, it is vitally important to not rush in as a bunch of firefighters to put out the fire. There needs to be a calculated approach to dealing with the failure.
The first step in addressing and learning from a failure is to analyze the scene of the failure like a detective would a crime scene. This is often a difficult activity, as we need to get production up and running as quickly as possible. But if we don’t take the time to collect the information and data, chances are production will be down again with the same issue.
To collect the data & information from a failure, there needs to be a systematic approach. One of the best approaches I have seen is to have a kit that is stored in the maintenance shop and brought to all failures. This kit contains everything that would be required to collect the data for a Root Cause Analysis (RCA). Included in the kit is;
- A checklist outlining the approach to use
- A digital camera
- A flash light
- Zip Lock bags of various sizes (used to collect failed components and keep as is until a failure analysis can be conducted)
- Failure Data Collection form (used to ensure all the failure data is captured in a repeatable way)
- Notepads w/ Pens (for recording observations)
- Markers (for writing on the Zip Lock bags)
- Measuring tapes of various size
- Adhesive Measuring Tape (to be used when taking pictures of the failure)
- Reference Scales (used when taking pictures of the failure scene)
- Inspection mirror
- Equipment tags (to tag the failed components that are too large for the Zip Lock bags)
It is important that this kit is maintained and replenished after each failure, so it is ready to go for the next failure. Once the failure data has been collected, it is time to learn from the failure.
Once the failure scene has been analyzed and data collected, then the repair can be made to the equipment.
How to Learn from Failure
With the failure data collected, a Root Cause Analysis can take place to learn from the failure. Depending on the severity of the failure, a different approach to determining the root cause may be taken. It could be a simple 5-Why focusing on the 3 legs (Direct root cause, detection root cause, and systemic root cause). Or the failure may warrant a Fault Tree Analysis, taking into account all human factors.
Regardless, any failed components should be sent off for detailed analysis. Bearings should be sent out for analysis to determine the true cause of the failure. Was it a lubrication issue, wear, etc.? Using expertize is critical to identifying the cause of the failure.
Many suppliers and distributor will provide this engineering service either free of charge or at a deeply discounted rate. Be sure to ask your supplier and take advantage of the service.
At the outcome of the root cause analysis, you should have a detailed reported outlining the cause of the failure, and what needs to be done to eliminate the cause of the failure.
It is important to note that the RCA process is not about finding a responsible person (even if there was a responsible person). It is about learning from the failure and making sustainable changes.
Applying the Learnings
This is where many organizations fail. They often collect the data and perform some form of Root Cause Analysis, but they fail to implement some or all the recommendations coming from the Root Cause Analysis.
It is vitally important to improving the reliability of the equipment, that all recommendations are reviewed and if it makes sense to implement them. All recommendations should be reviewed against criteria to determine the impact that it will have on the failure and the ease of implementation.
When you have this criteria set up, it should eliminate many of the recommendations that call for “setting up a PM”. You will never improve reliability by adding a PM routine for all failures encountered.
The Root Cause Analysis may determine many other factors that need to be addressed. This could include an engineering change to add the ability to monitor the condition of the component. It may also include training recommendations, equipment redesign, material changes, etc.
With the recommendations are implemented, you are on your way to improving the reliability of the plant. And if by chance this failure does occur again, be sure to review your previous Root Cause Analysis for any gaps. Take those gaps and improve your RCA process in the future.
Does your site have a failure analysis kit ready to go? Do you have defined criteria for evaluated RCA recommendations? By taking a few small steps to improve your RCA process and learn from your failures, you will improve plant performance.
Remember, to find success, you must first solve the problem, then achieve the implementation of the solution, and finally sustain winning results.