Although the name makes it a bit confusing, reliability centered maintenance (RCM) is not a type of maintenance. It's a way of deciding which maintenance strategies work best for each of your assets. The good news is there's an established process: start with four principles then ask yourself seven questions.
Let's look at those four and seven more closely.
For a program to be RCM, it needs to:
Here's a simple example of putting these principles into practice. Let's say you're in property management, and part of your job is making sure the communal washers and dryers are working. For the dryers, you decide the failure modes include dead motors, blocked exhaust pipes, and excessive buildup in the lint traps. Any one of those can cause the dryers to stop working. When it comes to prioritizing them, you know that they're equal in terms of inconvenience. Tenants are going to be unhappy regardless of the reason. But the failure modes are different as soon as you consider safety. A dead motor isn't going to catch fire, but hot lint certainly can. So, using your property management software, you set up a recurring preventive maintenance task to check and clean the lint traps. The motors you let run to failure.
Now that we know the principles, let's look at the questions we need to ask about the assets.
Once you've decided which assets to include, you'll need to go through these questions for each one. The good news is that if you have a CMMS, answering them will be a lot easier. That's because you'll be able to quickly and easily review all your old on-demand and preventive maintenance work orders (A good CMMS always has a built-in work order software as well). A CMMS keeps your data safe and makes it searchable. On top of that, it can crunch the numbers for you with automatically generated reports.
Quick note on computerized maintenance management software: it takes care of work order management, preventive maintenance scheduling, inventory management, tracking KPIs, and generating reports. Modern versions are backed by cloud computing, making your data accessible from anywhere, at any time. Your data is always safe; providers take care of all the IT heavy lifting for you.
Here are the questions.
1. What are the functions and desired performance standards of each asset?
In a perfect world, what and how much would the asset be doing for you? We can jump back to our example with property management software and a dryer. Let's imagine it's doing five loads a day, which gives us an average of 150 loads a month. We get those numbers by looking at the number of quarters we took out of the machine during any given month without any on-demand work orders.
2. How can each asset fail to fulfill its functions?
When things go wrong, how do they tend to go wrong? Pulling the data from our CMMS, we can see it's generally user error or a malfunction related to the coin slot mechanism or the lint trap.
3. What are the failure modes for each functional failure?
Here, we're just taking the answers from the previous question and adding in some details.
User error: lack of training and poor signage around the machine
Coin slot: worn internal mechanisms prone to jamming
Lint trap: lack of training and poor signage around the machine, no PM schedule in place to clean and repair lint traps
4. What happens when each failure occurs?
Now that we know the possible failures and what's causing them, we look at the immediate fallout.
With user error and that old coin slot, you're really only going to have unhappy tenants, but with the lint trap, you could have yourself a fire.
5. What are the consequences of each failure?
Sort of the same as the previous question, but it's helpful here to think in terms of numbers and long-term consequences. For example, a fire would cause a lot of damage. You could get lucky and only have to deal with cosmetic repairs, but they could be structural. On top of that, there's insurance to deal with (your rates are going up, guaranteed), as well as the possibility of fines. And how many tenants are going to resign their lease if they're now worried about the building catching fire?
6. What can you do to predict or prevent each failure?
Now that you know the types of failures and how they tend to happen, you can start to look at different maintenance strategies. A good CMMS lets you review the work order history for any asset, showing you trends and problem spots. If the coin slot has jammed three times near the end of the last three months, you can set up a PM to check and adjust it the 20th of each month.
7. What should be done if a suitable proactive task cannot be found?
The wording for this one is a bit tricky. It's not that you can't find a proactive task to avoid the failure. Instead, it's that there's no proactive task that makes sense in terms of time and effort. Back to our dryer example. If you have one that's close to the end of its useful life, you could proactively assign a technician to babysit it around the clock. Every rattle quickly and expertly attended to. Every wobble diagnosed and adjusted for. Complete waste of time and talent.
What makes more sense is to let the machine run right up until it dies. If you know the lead time to get a new dryer ordered and installed is four days, you can have a plan in place to bring up that extra dryer from the basement. You also know tenants on the third floor won't care about having to walk up or down a floor for a few days if you promise to let them use the dryers for free. No matter how many quarters you give out over those four days, it's going to cost you a lot less than maintaining a dryer on its last cycles.
Now that we're at the end, it's worth going back and look more closely at that sixth question. The example with the coin slot is pretty basic, and there's usually a lot more to choosing the right maintenance strategy than that. To learn more, check out How to Pick the Right Preventive Maintenance Strategy. Another good resource is Find the Best Maintenance Management Strategy.