|
Basic Steps of Applying Reliability
Centered Maintenance (RCM) – Part I
The
previous issue of
Reliability HotWire reflected on
the philosophy and perspective that Reliability Centered Maintenance (RCM)
brings to the field of maintenance. This article begins to present the system
analysis process that is used to implement an RCM program.
Although there is a great deal of variation
in the application of RCM, most procedures include some or all of the seven
steps shown below:
- Prepare for the Analysis
- Select the Equipment to Be Analyzed
- Identify Functions
- Identify Functional Failures
- Identify and Evaluate (Categorize) the
Effects of Failure
- Identify the Causes of Failure
- Select Maintenance Tasks
This article will discuss the first three
items, which are the preliminary steps that need to take place when embarking
on an RCM analysis project.
1- Prepare for the Analysis
Gathering the team
As with almost any project, some preliminary work and meticulous planning
will be required before beginning an RCM analysis. One of the first
steps in the analysis procedure is to assemble the proper team of
knowledgeable individuals to perform the analysis. The most efficient
and effective analysis teams are cross-functional, with different areas of
expertise represented. The size of the team should be adequate (typically 4
or 5 people) but not too large-- "too many cooks spoil the soup." At least one person from maintenance should be part
of the group. RCM is a multifaceted process that requires a thorough
understanding of the assets being considered for RCM, the purpose of the
assets and the impact of their malfunction. The goal is to gather sufficient
knowledge/expertise for an effective analysis without wasting valuable
resources and/or making meetings unmanageable.
A facilitator is recommended to ensure that the RCM analysis is carried out at
the right level, that no important items are overlooked and that the results
of the analysis are properly recorded. A facilitator also manages issues among
the team members and helps in reaching consensus in an orderly fashion as
well as
retaining the commitment of the members and keeping them engaged.
Establishing ground rules and
discussing a plan
Identifying and documenting the ground rules and assumptions that will be
followed during the analysis can facilitate the analysis process by making
sure that all members of the analysis team understand and accept the
conditions of the analysis. Issues to be discussed when preparing for an RCM
project may include setting the goals of the RCM project, addressing project
management issues and resources required to carry out the project (such as
schedule and budget, meeting procedures, reports, how to make
recommendations, manpower, tools, consultants, software and meeting rooms) and foreseeing, as
much as possible but without getting swamped, the
obstacles that lie ahead of the project (company resistance and lack of
buy-in, lack of data, bureaucracy, lack of leadership and
commitment, etc.). Develop a plan with a viable vision for the future and
run with it!
2- Select the Equipment to Be Analyzed
Scope of analysis
The RCM team has to reach a decision about the level of the asset at which the analysis should be
conducted
(e.g.
part, component, subsystem, system or plant)
and whether the entire plant/facility should be considered for
RCM. Consider starting your RCM analysis at the system level, as it is a good,
safe and manageable place to start then expand your analysis upward and downward. Typically, systems are a logical starting
point for the analysis since they constitute the building blocks for
plants/facilities. Because RCM is focused on preserving the function of
equipment, performing the analysis at the system level, where functions are
usually derived, makes good sense. Focusing on levels below the system
level (e.g. components) limits visibility for the analysts and makes them
detached from the broad significance of failure, especially when components
support multiple functions. Also, comparing failure modes and prioritizing
resources become more useful and feasible if the analysis starts at the
system level rather than the component level, which might have only a few
failure modes. On the other hand, analyzing entire
plants in one bite could be overwhelming and might run the whole RCM program
to a stall.
The suggestion of
starting the analysis at the system level may not work for everyone, of
course.
Depending on system complexity, constraints and other factors that might be
unique to your application and asset, other levels might be more appropriate
as a starting point.
System boundaries
Selecting the equipment to be analyzed also involves defining system
boundaries. Defining system boundaries helps in specifying precisely what is
included and not included in the system so that an accurate and complete
list of components can be identified and no overlap with component
lists of other systems (especially adjacent systems or systems affected by
components in other systems) can happen. More importantly, the boundaries
help in determining the inputs, outputs and functions of the system.
System description
Once the equipment to be analyzed has been selected, it is time to describe it.
Identifying and documenting the essential details of the system is necessary
in order to
perform the remaining steps in a thorough and technically sound manner.
Describing the system helps the analysts gather a comprehensive
understanding of the system. A well-documented system description will help
record an accurate baseline definition of the system as it was at the time
of the analysis (this is also useful because systems can be upgraded or
modified with time). A system description can also assure that the analysts
have identified critical design and operational parameters that play a key
role in delineating the degradation or loss of required system functions.
System descriptions may include: Functional
block diagrams, component breakdowns and hierarchies, input/output interfaces,
electrical schematics, environmental conditions, design specifications,
equipment histories (especially information pertaining to failures), the
definitions of "failure" that will be followed during the analysis,
operation manuals, previous maintenance plans, specifications of the
operational environment for the equipment and any assumptions that may
affect the analysis.
Select the equipment
Once the level of analysis has been established, the candidate systems that would
benefit the most from a new maintenance program should be identified and
prioritized. Various criteria, such as safety, legal and economic
considerations, can be
used in determining the benefit obtained from maintenance.
Various equipment selection methods are
available.
One
approach would be to evaluate maintenance records (number of failures,
outages hours, loss of productions cost, safety problems, etc.) for a given
period (e.g. 1 year or 2 years).
The 80/20 rule states that most (80%) of the
problems in a plant or system can be attributed to a few (20%) vital
players, so it is useful to prioritize the issues in your plant before deciding
on a plan of attack.
Another approach would be to apply a pre-defined set of selection questions,
For example, the MSG-3 guideline (used in the aircraft industry) proposes
four questions:
- Could failure be undetectable or not
likely to be detected by the operating crew during normal duties?
- Could failure affect safety (on ground
or in flight), including safety/emergency systems or equipment?
- Could failure have significant
operational impact?
- Could failure have significant economic
impact?
Answering "yes" to at least one of the above questions requires detailed
analysis for the equipment.
Another method, called the Criticality
Factors method, consists of a set of factors designed to evaluate the
criticality of the equipment in terms of safety, maintenance, operations,
environmental impact, quality control and other factors. Each factor is
rated according to a pre-defined scale (e.g. 1 to 5 or 1 to 10) where higher
ratings indicate higher criticality. The combined criticality value score can
then be used as a ranking system for different types of equipment or to be compared
with a threshold in order to decide whether the equipment will then be part
of the RCM analysis.
Whichever method (or combination of
methods) is selected, the goal of this task is to provide a systematic
approach to focus the RCM analysis resources on the equipment that will
provide the maximum benefit and to ensure highest return on investment.
3- Identify Functions
Since the ultimate goal of an RCM project is "to preserve system function,"
it is therefore incumbent upon the team of RCM analysts to define a complete
list of system functions. The system functions would then drive the required
functions of the equipment supporting the system functions. (Tip: The output of a system
typically captures
the function of the system; therefore, every output interface could be
translated into a function statement.) It is desirable to start function
statements with a verb (to pump water, to provide alarm, etc). It is also
recommended to specify the acceptable level of performance desired by the
user of the asset as opposed to the actual performance that may reflect an
operational or maintenance issue.
Keep in mind that function statements are
not about what types of equipment are within the system and therefore the use of
equipment names to describe system functions should be avoided. Making a
mistake about this leads to the common fallacy that the goal of maintenance
is protecting equipment.
For
example, "To maintain discharge flow of 500 gpm" would be a more useful
function statement than "To provide a centrifugal pump that delivers 500
gpm."
The function definition should be as
quantitative as possible. For example, a function should not be defined as "To
produce as many units as possible," but rather "To produce a target of 25 units
with a minimum of 22 units in an 8 hour shift." It becomes difficult to decide
on maintenance strategies or to hold the people involved in maintenance
accountable for not meeting goals of maintenance when the goals are not
defined precisely. Qualitative definitions are, however, needed in some
circumstances. For example, aesthetic functions are hard to define with
exact terms and therefore words such as "should look acceptable" or "to look
attractive" can be used; however, there has to be a common understanding and
consensus about
what such definitions mean.
Some function definitions are absolute
(e.g. "To contain liquid," where no leakage is acceptable) while others are
variable (e.g. "To remove unwanted particles of 100 microns from air
stream."). The RCM team should be careful about using absolute definitions
when a variable definition is more appropriate.
Conclusion
This article presented the first three basic steps of an RCM program, which are
the required steps to ensure that an RCM project is begun on the right
track. An article in next month's HotWire will discuss the remaining steps.
References
ATA MSG-3 "Operator/Manufacturer Scheduled Maintenance Development," updated
in March 2003.
Moubray, John, Reliability-centered Maintenance, Industrial Press,
Inc., New York City, NY, 1997.
Nowlan, F. Stanley and Howard F. Heap, Reliability-Centered Maintenance.
Issued in December, 1978.
SAE JA1012 “A Guide
to the Reliability-Centered Maintenance (RCM) Standard,” issued in January
2002.
Smith, Anthony,
Hinchcliffe, Glenn R., RCM - Gateway to World Class Maintenance,
Elsevier Inc, Burlington, MA, 2004. | 
