Where Does Root Cause Failure Analysis (RCFA) Fit In A RCM Strategy?

53 Claps

1647 Words

9 min

What should be done if suitable proactive maintenance tasks cannot be found?

First let me note how RCM (Reliability-Centred Maintenance) is being performed. To those new to this concept, RCM is defined as a process used to determine the maintenance requirements of any physical asset in its present operating context. This is simply being performed by answering the 7 basic questions of RCM:

- What are the functions and associated standards of performance of the asset in its present operating context?
- In what ways does it fail to fulfill its function?
- What causes each functional failure?
- What happens when each failure occurs?
- In what ways does each failure matter?
- What can be done to prevent or predict the failure?
- What should be done if a suitable proactive task cannot be found at all?

RCM consists mainly of two parts, the first part is deriving the FMEA of the asset being analyzed through what it termed as the RCM Information Worksheet, the second part includes addressing each failure modes and effects as well as the consequences of each failure in order to define the most appropriate as well as the most feasible maintenance task to use.

Difference Between A Failure Mode & Root Cause

When equipment fails, there are a wide variety of causes as to why it fails and when we speak of this vast amount of causes that prompt the equipment to fail then we speak about its failure modes. As from the late John Moubray, a Failure Mode refers to an event which could cause a functional failure to occur. These refer to the probable causes of failures in which the failure might have occurred before or is possible to occur. Root Cause is also the understanding from the things that go wrong so that we can learn from the failure themselves, the depth of the analysis in Root Cause will always lead us that for every failure, there is always a corresponding Human Cause as well as a hidden cause or Latent Cause behind the problem. Failure Modes and Root Cause are two different terms. Its difference lies in the depth & breadth of the analysis.

Both RCFA and RCM are two powerful strategies, yet they can only complement each other when we know how and when to use them respectively and this is where we can derive the most benefit from it. Again they should not contradict each other but rather complement one another. Failure Modes is like having a shotgun and shooting in all directions in the dark while a root cause is having a rifle with a scope and aiming for one target at a time.

RCFA termed Failure modes as probable causes of failure or hypothesis and is usually used in the initial levels of the analysis; each failure mode is being verified in order to proceed on to the next level. Root Cause will always be based on the evidence found during the investigation and analysis. Any analysis which tends to end up into the part or component level is still not considered as a Root Cause but rather is simply termed as the Physical Cause of the Failure or most commonly known as its Failure Analysis.

Let me provide an example to clarify this: A pump fail to fulfill its function since it is not discharging any fluid at all. This failure to discharge is known as the Functional Failure. If we think of the different reasons or probable causes as to why this pump is not discharging any fluid at all then we are speaking about Failure Modes and not its Root Cause. Therefore the Failure Modes of the pump not discharging fluid at all will be as follows:

Failure Modes Include:

- Valve is totally closed
- Motor totally burnt out
- Bearing seizure (stuck-up)
- Strainer totally clogged
- Broken pump impeller
- Supply tank is empty
- Clogged impeller
- Driver imbalance
- Insufficient suction pressure
- And many more . . . . .

I could continue on with the list of failure modes or probable causes which could warrant a pump not to discharge fluid at all, but for now lets assume the all the Failure Mode had been listed so far. Failure Modes are the probable causes that will affect the pump from not discharging any fluid at all. You may even include a lighting to strike the pump or an earthquake that severely damage the pump.

On the other hand, sad to say that this pump must actually fail in order to perform a Root Cause Failure Analysis, this is what makes Root Cause Failure Analysis reactive in the first place since a failure must occur before performing it. On the other hand, Root Cause Failure Analysis is at the same time Proactive because by knowing the cause of the failure we can learn from it and address similar situations in the future when it arises. Therefore, by analyzing the pump’s failure, we need to verify what failure mode actually happened during the time the pump had failed, and when the analyst had verified each failure mode and found out for example that the actual cause of the pump’s failure to fulfill its function is a bearing seizure then we need to analyzed what caused the bearing to seized, the other failure modes will be disregarded. After we analyze the cause of the bearing seizure a Failure Analysis will be written as.

Failure Analysis will be:

- Bearing Seizure due to lack of lubricant in the raceway

A Failure Analysis will stop at the component level, but a Root Cause Failure Analysis will still proceed with the analysis and will conclude when the Latent Cause of the problem had been identified. Hence, a complete Root Cause Failure Analysis will be written as follows:

Root Cause Failure Analysis will be:

Level 1: Physical Cause of the failure - Bearing Seizure due to lack of lubricant in the raceway

Level 2: Human Cause of the failure - Maintenance had used the wrong lubricant for the bearing

Level 3: Latent Cause of the failure - It is not clearly specify in the PM procedure as to what type of lubricant to be used for this type of bearing which had cause the maintenance to use the wrong type of lubricant.

Now that we have differentiated "Failure Modes" from "Failure Analysis" and "Root Cause Failure Analysis", the question raised is where does RCFA fit in the RCM Strategy?

Some say that RCFA can fit into some failure modes that are already the subject of some Root Cause Failure Analysis investigation. I disagree with this, since if the failure mode is subject to investigation, then the correct term to use here is Failure Analysis and not Root Cause Failure Analysis. RCFA will go much deeper and would expose the hidden causes of failure and second if a Root Cause Analysis had already been performed in any of the failure modes, then it should not be written in the RCM Analysis lists of failure modes since a recurrence is unlikely to happen if a successful RCFA had been concluded so far. Most RCFA will default to redesign or modification.

John Moubray, RCM’s author asks that which comes first, Redesign or Maintenance. He simply stated that maintenance should come first before redesign. I agree with his statement since there is always a temptation to modify or redesign the system without reviewing the current maintenance tasks at hand on the other hand most people are tempted to redesign or modify the equipment without even performing a thorough RCFA investigation which sometimes leads to new problems. RCM is being performed on an asset so that the people involved, which are the operators and maintenance can better understand the different failure modes that can affect their asset and the consequences of each failure itself when it actually occur. Not all failure modes can be address by maintenance, and when a maintenance task is not feasible to use, then the team will default to modification or redesign. A redesign will include changing a specification of a component, adding a new item, changing a components size, shape, material or even changing a process or procedure.

There is a limitation to maintenance itself, this is seen in the last question of the RCM Seven Basic Questions. What should be done if suitable proactive tasks cannot be found? One of the options to use here is to redesign. A redesign is warranted and feasible to use if the consequences of failure is simply not acceptable to the user.

RCFA is used after performing an RCM Analysis and not before starting an RCM Analysis, This is what I believe where RCFA will it in the RCM process. Hence, after performing an RCM Analysis and there are still failure modes that keep on repeating itself in which the consequences is not acceptable, then an RCFA should be conducted. 2nd point equipment failures attributed to human error should be included in the Root Cause Failure Analysis.

I have friends who consult RCM and also those that consult RCFA/RCA, and I sometimes see them in a conflicting mode as to which strategy is best to adopt. I believe that both strategies are not perfect yet when combined will create a much more powerful strategy for the maintenance arena, but this can only be possible when both of them are use in the right perspective. Again, they should not contradict each other but rather complement one another.

Where Does Root Cause Failure Analysis (RCFA) Fit In A RCM Strategy?

By Rolly A.

An Article by

Rolly A.

A Consultant from Sta. Rosa, Laguna, Philippines

Search Expertbase

Hire Rolly

Share this Article