At Bluefield, one of our most common projects is helping clients design or improve their asset management plans. Often, this work starts with preparing or reviewing a maintenance strategy analysis, such as a Failure Modes, Effects and Criticality Analysis (FMECA).
Many people don’t enjoy preparing FMECA’s, and there is mixed opinion as to how effective they are. In Bluefield’s experience, a FMECA can be extremely valuable to a business, but there are many factors that can affect their quality and usefulness. So, we asked our team the following question:
When it comes to completing FMECA's, what are your tips for making them quicker and easier to complete, and for making them add more value to a business?
From my experience, it helps to simplify the model. In a perfect world, we would raise the functional failures at the level of the machine, and we would only be concerned with the functions and secondary functions at a machine level.
Practically though this makes it difficult to identify the failure modes, so it is necessary to analyse the equipment at a lower level in its functional structure. However, the lower level of detail we break the machine down to the more complicated the model becomes so it is best to stay at the highest level possible.
Another way to make it less complicated to implement is to only use it for parts of the plant that are critical or where there are reliability issues, rather than apply it everywhere.
In order to make this approach add real value to the business it is necessary to include two other key actions:
To ensure that management actions (or tactics) selected to manage a failure mode are selected through task decision criteria like that presented in RCM2.
To make the strategy a living system and continue to update it over time as we have failures.
Gerard has previously written about the importance of learning from failures to update your maintenance tactics. Read about it here.
Start with a base template of a similar machine where possible.
Use standardised formatting, structure, and terminology across all.
Think hard about what value it will add before starting. It is an extremely time-consuming process to build comprehensive and effective FMEA - any gaps could be the failure that costs you dearly. A far more time and cost-effective approach to provide the same result can be to review the existing maintenance strategies, identify gaps, and close them.
Treat them as a live document. If they are not continuously updated to capture changes and lessons, they become irrelevant and waste of the original effort and investment.
Read about Mark’s experience with FMEAs on shiploaders here.
I have never been part of carrying out a full-blown RCM style FMECA, have seen some examples sitting in bookshelves with attempts at following the whole RCM process, but never seen an example turned into a practical packaged set of tactics.
Even new model machines are comprised of a majority of maintainable items (major components) that have been used in similar types of machines or plants previously. For businesses or companies that have experience with a range of similar assets to the asset you want to develop maintenance plans for, the process should consist of pulling in data for similar maintainable items or components for all of the major items on the new asset. If there is a completely new piece of technology in a component, then start with OEM data and ensure there is a short-term review built in for those particular items.
The process can follow the basics of a FMECA but take on more of a PM review process looking at what might be different about the application the item is being used for in the new asset. Focus on maintainable item level e.g. gearbox, motor etc, not all the rats and mice. Ensure there are key experienced maintenance AND operations people involved in the process, and do not forget the value of operator inspections.
Then to make the tactics coming from the process practical, make sure there is a period of field trialling across all different shifts who carry out the maintenance, and update based on that feedback. The reliability engineer accountable and other members of the team should be part of going with the teams carrying out the field trials.
And as the others have said, make sure you go back and have a look at a set interval or if you are seeing failures or issues with process.
It is easy to get overwhelmed with the number of equipment experiencing reliability issues and unplanned downtime when first attempting to undertake a Failure Modes, Effects and Criticality Analysis (FMECA). Therefore, before starting the analysis, I recommend first developing a broad understanding of each of the key pieces of equipment and how they interact with one another. This will allow you to understand which pieces of equipment are critical to the operation of the system and thus where your attention should be focussed.
You can then dive deeper in the operational and maintenance manuals and any drawings available to develop your understanding of the functional descriptions, interlocks, and critical settings. During this stage, I like to take notes that contain specific pieces of information that will be relevant to the FMECA. This will save you a significant amount of time as you will not have to go hunting through the manuals for the required information when you are undertaking the analysis.
When starting the analysis, I suggest first breaking it up into the key Functional Locations and separating these areas using visual cues such as a highlighted row to break up each section. Depending on the client’s requirements, there may be many different cells to populate. In this case, I would suggest starting with the core parts of the analysis first (i.e. Functional Description, Functional Failure, Failure Mode, Effect and Recommendation). Complete these main sections for all Functional Locations before going back and filling in the remainder of the fields to complete the analysis. This will build a solid foundation for the FMECA and will also achieve a more wholistic analysis minimising deviations between the various functional locations.
Once the FMECA is complete, it is good practice to review any unplanned downtime or material usage data available and compare this with the FMECA. Actual data will allow you to identify any gaps in the analysis and ensure the recommended maintenance frequencies are adequate for the specific plant.