RCM stands for Reliability Centered Maintenance. It is a method for determining, for each installation, which preventive maintenance is really needed, starting from what happens when something fails. RCM originated in aviation and is now used in every industry to achieve the required reliability at the lowest possible maintenance cost.
Classic RCM is thorough but time-consuming. Industrial RCM (iRCM) follows the same principles and is 50 to 100 times faster, because you analyze a type of component or installation once and reuse that result for every following installation. This article explains how both work and when to choose which.
Why maintenance based on the manual falls short
Most preventive maintenance schedules come about in one of two ways: the manufacturer's instructions are adopted as they are, or the maintenance department adjusts the schedule after a failure. Both share the same weakness. The manufacturer does not know your operating conditions and therefore does not know what a failure costs you. A manual also rarely says anything about modern condition monitoring. The manual of an electric motor, for example, contains no vibration measurements, even though that is a proven way to follow the condition of motors.
The second problem is that you reason from the component and not from the consequences. Take two identical pumps. One feeds the only cooling circuit of the plant, while the other has two more pumps running in parallel. The component is the same, the risk is not. If you maintain both in the same way, you do too much on one and too little on the other. Fine-tuning after every failure does not help either. Attention then only goes to what has already gone wrong once, while a failure that has never happened can be the most dangerous one.
The seven questions of RCM
An RCM analysis is carried out by a team from maintenance, engineering and production, led by a facilitator. For every installation, the team answers the same seven questions:
- What should the installation do, and which performance standards must it meet in its current application?
- In which ways can it fail to fulfil that function?
- What causes each of those failures?
- What exactly happens when such a failure occurs?
- What are the consequences for safety, the environment, production and costs?
- What can we do to prevent the failure or to see it coming in time?
- What do we do if no suitable preventive task exists?
For the last two questions, RCM uses decision logic. For each failure you choose between inspecting or measuring periodically, replacing or overhauling periodically, letting it fail and repairing it, or modifying the installation. The more severe the consequences, the stricter the requirement for the measure. If a failure is unacceptable and no maintenance task helps, a design change is mandatory.
Classic RCM: thorough, but heavy
In classic RCM, a team analyzes every installation separately, with all possible failure modes and causes. This gives excellent results, including for rare failures that another team overlooks. It does take a great many man-hours. Classic RCM therefore suits truly critical installations, unique installations for which no earlier experience exists, and systems or components of which there are so many that fine-tuning pays for itself.
For a complete plant, a hospital or an infrastructure project it is not feasible. These are the well-known weak points:
- It is time-consuming, so it is not realistic in a regular organization to have a team analyze every installation separately.
- The economic assessment stays limited: is the maintenance actually worth it?
- If no preventive measure is sufficient, the method says you must repair or redesign, but not how to organize that or who does it.
- Every installation is assessed separately, so two identical and equally critical installations can end up with different schedules. In addition, not every team knows all failure modes and inspection techniques.
Industrial RCM: the same logic, faster
Industrial RCM was developed to remove these weak points. It uses the same principles and scientific basis as classic RCM, with one big difference: you do not analyze every individual unit but every type once. Instead of analyzing all centrifugal pumps in a plant separately, a team of specialists maps out once how a centrifugal pump can fail and which maintenance addresses that. The result goes into a library of analyses, in the form of templates per component type. In a following project, for example a tank farm, a new team uses that library as its starting point.
A template is never adopted unchanged. The maintenance is adapted per asset to the criticality of the installation in which the component runs. The same pump gets a more extensive schedule in a critical system, for example with more condition monitoring and shorter inspection intervals, than in a system where a failure has little impact. This way every asset gets exactly the attention it needs, without having to analyze each unit again.
As a result, iRCM works 50 to 100 times faster than classic RCM and is suitable for production lines, building systems and infrastructure. The team no longer has to think up all failure modes itself. It assesses what has been prepared and answers the questions that really matter: is the schedule practically feasible, is anything missing, is it economically worthwhile, who carries it out and which spare parts are needed if we choose to repair after failure?
Classic RCM and Industrial RCM side by side
| Aspect | Classic RCM | Industrial RCM |
|---|---|---|
| What is analyzed | Every installation separately | Every type of component or installation once |
| Lead time | Very time-consuming | 50 to 100 times faster |
| Knowledge | What the team thinks up itself | Library of earlier analyses, supplemented by the team |
| Identical installations | Can end up with different schedules | Get the same schedule at equal criticality |
| Economic assessment | Limited | Part of the team session |
| Best suited to | Critical and unique installations | Complete installations in regular industry |
How Industrial RCM works in six steps
- Management requirementsWhat does an hour of downtime cost and how long may the installation be down? Which risks to safety, the environment and quality are acceptable? You record this in a risk matrix.
- Overview of the assetsThe asset register or master equipment list. For many companies this is the most time-consuming step.
- Determine criticalityEvery asset gets a score against the requirements, from people who know daily practice and not from the management that set the requirements.
- Draw up the basic conceptAn engineer takes the templates for all components in use from the library and adapts the maintenance of each asset to its criticality: more extensive for critical assets, lighter for less important ones.
- Team sessionA team from maintenance, engineering and production reviews the concept, checks whether the adaptation to criticality is right, supplements it and tests it for feasibility and economics.
- Overall schedule and action itemsIndividual tasks are clustered into work orders. Improvement actions get an owner and a priority.
Criticality determines how much attention an asset gets
Two identical assets do not need the same maintenance. In Industrial RCM, criticality determines how you adapt the template for a component. Many companies therefore divide their assets into three groups:
| Group | Characteristic | Approach |
|---|---|---|
| Critical | Failure is unacceptable according to the risk matrix, or shuts down the system | Extensive analysis in a team and an extensive maintenance schedule |
| Important | No unacceptable risk, but the system can be down for a while | Analysis in a team and a balanced schedule |
| Less important | All other assets | An engineer sets up the maintenance alone according to best practices, supplemented by inspection where needed |
Example: 720 gear motors
A logistics department had 720 gear motors, almost all small and standardized, with a stock of each type. The company wanted to introduce vibration measurements, but there was discussion about it. Some expected it would cost more and cause more downtime than replacing a defective gear motor now and then.
A criticality analysis divided them into 20 critical, 30 important and 670 less important gear motors. The team chose its own schedule per group:
| Group | Number | Maintenance |
|---|---|---|
| Critical | 20 | Measure vibration monthly, inspect weekly for contamination, heat, fan and noise, change the oil every three years |
| Important | 30 | Measure vibration quarterly and inspect visually and by ear, change the oil every three years |
| Less important | 670 | Inspect visually and by ear every six months during maintenance on the system the gear motor is part of, change the oil every three years |
Calculating per motor whether a vibration measurement pays off was not feasible. It took too much time, the price per measurement depends strongly on the number of measurements the analyst does in one day, and there was too little data. General figures per type of component did exist and were enough to set up a well-founded schedule per category.
What does Industrial RCM deliver?
The result is a standardized and well-founded maintenance schedule, plus a list of action items: technical improvements to machines, better ways of operating, proposals for spare parts and proposals for maintenance contracts. Industrial RCM has been applied at hundreds of companies in practice, and the investment in time and money is often earned back within a few weeks.
A schedule is only worth something if it is carried out. A new schedule demands more work at the start, while later it delivers fewer failures and therefore less work. Support is therefore important. Resistance usually arises because people fall back into old patterns, do not dare to drop existing tasks, or do not see the point of the new approach. This is why iRCM uses group sessions in which people from the shop floor contribute. Whoever helps build the schedule will also defend it. So choose participants not only for their expertise but also for their influence within their own department. The rest of the implementation is part of the program: a work process for preventive maintenance, software, contracts, work instructions, tools and training.
Frequently asked questions
What is the difference between RCM and Industrial RCM?
Both follow the same method. Classic RCM analyzes every installation separately, Industrial RCM analyzes every type once and reuses the result. That makes iRCM 50 to 100 times faster.
When do you choose classic RCM?
For truly critical installations, for unique installations for which no earlier experience exists, and for systems or components of which there are many. For the rest of a regular production environment, Industrial RCM is more practical.
Who is RCM intended for?
For maintenance and reliability engineers, asset managers and maintenance managers. RCM also forms the basis for well-founded decisions before you select and configure an EAM system.
Can I learn to apply RCM?
Yes. In the three-day in-company Industrial RCM Training & Course, participants learn to apply classic RCM and Industrial RCM themselves. Would you rather have guidance? Contact us and we will support your RCM analysis.