Reliability Centered Maintenance (RCM2) is a maintenance process optimization methodology that identifies the best maintenance strategy.
The RCM, for reliability centered maintenance, appeared in the years 50 in the aerospace industry. The first descriptions of the RCM were made in 1978, in a report by Nolan and Heap1 commissioned by the U.S. Department of Defense. Then, in 1984, it will be the Electric Power Research Institute (EPRI) which introduces it into the nuclear sector2. And finally, this methodology will be popularized in its version 2 by John Moubray3. At the time, Nolan and Heap described the RCM as ” RCM is a logical discipline for developing maintenance programs. The objective of these programs is to increase the reliability of the equipment while reducing the costs ». It makes it possible to implement the cheapest programmes, with the least operational risks and a high reliability of the equipment.
In this, it evaluates, categorizes and prioritizes failures and types of failures for each equipment according to the severity of their consequences. This prioritization and typology allows us to choose between the different maintenance strategies without going to the case, perform preventive maintenance, predictive or conditional or even redesign the equipment4.
At the end of the war, traditional maintenance becomes inappropriate for new planes. At the time, the traditional approach was based on revision programs where each Part has the ” right age ” to perform maintenance (repair, change…). Through these ” reliability program “, it was discovered that many types of failures could not be foreseen or reduced and especially that they were becoming very expensive in view of the increase in the complexity of the technologies. Studies conducted at the time showed:
- Revision programs have minor effects on the reliability of a complex system.
- Many parts did not have a valid maintenance program.
- There is no correlation between the age of the part and the breakdown.
The conclusion was that imposing age limits for parts and setting up systematic maintenances often have little or no effect on increasing the reliability of complex systems.
Similarly, in the nuclear sector of the years 80, maintenance was carried out only on the manufacturers ‘ recommendations without taking into account the actual use of the equipment and little preventive maintenance was carried out on Yet critical parts.
As a result, maintenance had to respond to new issues:
- To operate efficiently with each type of failure, an appropriate maintenance tactic must be used.
- To increase maintenance productivity, you need to be more proactive and have a more planned approach.
- Lengthen the time between 2 stops for maintenance.
- Provide active and more cooperative support between maintenance personnel, equipment, operations and technical functions.
- Reduce overall maintenance costs Nevertheless, it must be pointed out that the return on investment is only for 5-10 years. So do not expect results on the term course.
The P-F curve
The P-F curve, also called degradation curve, has been made popular to explain predictive maintenance and when to do conditional tracking.
The interval P-F is the time between when we start to see a failure (the potential point of failure p) and when we can no longer use the equipment because its performance is degraded to an unacceptable level (the functional point of failure F).
The first step is to select the system in its entirety that we will study. Generally, criteria related to the safety or criticality of the equipment in the global function are taken into account for the selection.
2. Identify operational modes and their contexts
It is a question of describing the functioning and its level of performance required in the conditions and expected operating environment. For this, you can use the 5W2H tool for example.
We must describe for example the notions of speed, external temperatures, number of people transported, life spans…
3. Partition the System
To simplify the study and prioritize actions, it is necessary to partition the system to be studied in functional subsystems. The challenge is to reduce to a level of humanly understandable subsystems. The level of partitioning of the system is to be adapted to the complexity.
Generally, the correct level of partitioning is achieved when:
- It is easy to understand the boundaries of different groups and their overall functional contributions.
- One can simply identify the different modes of failure.
- The physical subsystem is the most convenient to maintenance.
From this analysis, one chooses the subsets that are going to be the subjects of priority studies. Beyond that, the criteria of financial gain or human resources necessary are taken into account:
4. Develop the system block diagram and identify functions and failures
To fully understand the system, a SADT diagram of the subsystem must be constructed. In detail, this diagram will represent not only the different functions of the subsystem, their purpose and the link between them.
5. Drive the FMEA
Once all the functions of the subsystem are identified, we will conduct an analysis FMEA On each of them. The challenge is to understand the causes and effects of failures. We will fill in the first part of the analysis document, the columns F (Function), FF (Functionnal failure) and FM (failure Mode).
6. Select the Fault maintenance strategy
Thereafter, the challenge is to follow the decision diagram below and complete the standard working document to select the most appropriate maintenance strategy.
The standard RCM2 sheet
The standard below allows you to synthesize all the work data. Thus, the columns of the FMEA (F, FF and FM) and all the squares of the decision diagram are found. For the different columns of H to S4, it is enough to answer via Y for Yes or N for No and then describe the actions according to what is indicated in the decision diagram and following the analysis of the system.
Source: J. Moubray (1997) – RCMII reliability centered Maintenance
The different strategies
In the light of the decision diagram, we find different strategies:
- on condition task: is understood under this term as both conditional and predictive maintenance.
- Restoration Task: Systemic maintenance of restoring parts without changing them
- Discard Task: systemic maintenance of replacing parts
- Failure Finding task: a task to perform a check on the proper operation of the equipment
- redesign: Part redesign See equipment
Each of these strategies and decisions taken during the RCM analysis is to be compared with what it is doing so far. The challenge is to understand the contribution of the RCM and to understand whether or not elements have been forgotten.
7. Determine spare parts
The stock of parts and the management of it must allow the maintenance to complete its program and anticipate the failures while limiting the costs. The choice and quantity of pieces follow the following considerations:
- Is the parts list aligned with the strategy?
- An analysis of the consequence of not having the Part must be done.
- Measure the quantity according to a potential consumption analysis. This analysis should be updated regularly to avoid over-stock or lack.
The decision-making diagram is this:
Source: Department of Defense (1999) – Requirements for for application of reliability centered maintenance
8. Documenting the RCM
The set of previous steps must be accurately documented. Being able to justify conclusions and regain the data of analyses is a preponderant phase both with regard to security and from an economic point of view.
9. Maintain and evaluate the RCM program
Decisions made during the RCM analysis are in constant motion. On the one hand, the experience of maintenance is improving but also the technologies and the modernization of the equipments lead to constant evolutions of the maintenance programs. Thus, the objectives of supporting the RCM program are:
- continuously monitor and optimize maintenance programs
- Remove unnecessary expectations
- Identify breakdown trends
- Understand and integrate new failures in the program
- Increase program Performance
Also, throughout the life of the company, a permanent evaluation of key indicator must take place. We can find for example the trends on the number of failures, the costs of maintenance, the number of day/man…
1 – F.S. Nowlan, H. F. Heap (1978) – Reliability centered Maintenance
2-International Atomic Energy Agency (2007) – Application of RCM to optimize operation and maintenance in nuclear power plants
3 – J. Moubray (1997) – RCMII reliability centered Maintenance
4 – M. Sondalini (2009) – A common misunderstanding about RCM
R. Kennedy (2011) – Examining the processes of RCM and TPM
R. Overman (2012) – Principles of reliability centered maintenance
America Bureau of Shipping (2004) – Guidance notes on RCM