Can RCM (Reliability Centered Maintenance) and Streamlined RCM Peacefully Co-Exist
Can RCM (Reliability Centered Maintenance) and Streamlined RCM Peacefully Co-Exist
Can RCM (Reliability Centered Maintenance) and Streamlined RCM Peacefully Co-Exist
Religious or political zealots confront one another, often, not on the basis of the mores of
their respective doctrines, but rather from the superficial differences in the details
surrounding each other's cultural reference points...
Religious or political zealots confront one another, often, not on the basis of the mores
of their respective doctrines, but rather from the superficial differences in the details
surrounding each other's cultural reference points...
By Murray Wiseman
Introduction
Religious or political zealots confront one another, often, not on the basis of the mores of
their respective doctrines, but rather from superficial differences in
the details surrounding each other’s cultural reference points. Mathematicians take
pride in their ability to adopt a new set of definitions and symbols as effortlessly as they
would don a fresh suit of clothes. Thus they proceed, unfettered by prior points of view,
to build new theorems upon old. The world of maintenance has, not dissimilarly, spawned
a multitude of cultures and languages for formulating solutions to real problems.
This chapter explores one of several streamlined RCM software programs. We begin with
an examination of what is meant by “streamlining”. We illustrate the streamlined
approach by describing a popular representative RCM software package called RCM
Turbo[1]. We set up a cross-reference “dictionary” of terms describing similar sounding
but, sometimes, differently applied concepts in the two “languages”. Finally we
summarize the relative advantages and potential drawbacks of the “streamlined” RCM
and the RCM processes. Through this process, we discover how the juxtaposition of two
approaches may enlighten the proponents of both.
Why streamline RCM?
Chapter 11(page 137) cited the SAE Standard “Evaluation Criteria for Reliability-Centered
Maintenance (RCM) Processes” that defines RCM as:
“… a specific process used to identify the policies which must be implemented to manage
the failure modes which could cause the functional failure of any physical asset in a given
operating context.”
“…Any RCM process shall ensure that all the following seven questions are answered
satisfactorily and are answered in the sequence shown as follows:
a. What are the functions and associated desired standards of performance of the asset
in its present operating context (functions)?
b. In what ways can it fail to fulfill its functions (functional failures)?
c. What causes each functional failure (failure modes)?
d. What happens when each failure occurs (failure effects)?
e. In what way does each failure matter (failure consequences)?
f. What should be done to predict or prevent each failure (proactive tasks and task
intervals)
g. What should be done if a suitable proactive task cannot be found (default actions)?”
Were we to consider the process (of answering the 7 RCM questions in the sequence
stipulated) unacceptably resource intensive, then, understandably, we would seek to
replace it with a process that consumes less time and fewer resources, but by one that
provides, no less a responsible (sufficiently rigorous) analysis. We emphasize that the JA
1011 SAE standard stipulates a minimal set of criteria for a process to be called “RCM”.
Therefore, it is to be expected that most commercially packaged RCM software systems
and methodologies will add a considerable number of features that will enhance and
facilitate the experience.
The original[2] as well as the various streamlined RCM methods all demand that the
assembled team of analysts (operational, process, and maintenance specialists) possess,
collectively, the knowledge necessary to make informed decisions regarding the
maintenance characteristics of the item under scrutiny. The process chosen (either
original or streamlined) must, therefore, encourage the maximum contribution by each
participant so that RCM decisions will carry the force of all knowledge and experience
available on the team. The success of any “RCM” methodology, therefore, depends
heavily on its ability to gain true consensus, throughout every stage of the analysis. The
group, guided by a well trained facilitator, exercises its best judgment when visualizing
the typical worst case scenario (TWCS) surrounding each functional failure analyzed.
Strategy: CBM
Hidden Function: A Function whose Hidden Failure Mode: Same meaning as
failure will not be detected under normal RCM’s “hidden function”. It is structured in
circumstances. Identified by RCM during the fields: Component, Failure Mode &
functional analysis when examining each Effects, Task Description, Frequency,
component (from schematics, p&ids, Duration, Initiate Date, Job Group ID,
photographs, and physical walkaround) Service Period, No. of Units in Service, No.
and listing the functions they suggest. of failures, and MTBF of the protective
Code phrases (such as “able to”, “in the device (calculated).
presence of”, etc) are used to point out
that a function is hidden or protected by
a hidden function. Subsequent questions
address the hidden function. The
“hidden” consequence supplants the
other (three) failure consequences in the
RCM logic for determining a mitigating
task.
RCM records this information in the free MTBF: related to the Failure Mode.
text answer to question 4, “Failure
Effects”. However the JA1011 standard
does not specify an explicit data field or
structure for MTBF.
RCM records this information in the Strategy: related to Failure Mode. Takes
answer to question 6 and 7 “Tasks” when one of three possible values: 1) fixed time
following one of the four branches (H, S, maintenance, 2) condition based
O, N) in the RCM decision logic tree. maintenance, or 3) operate to failure
Same definition. RCM records this P-F Interval: related to Failure Mode.
information in the free text answer to Estimated interval (measured in working
question 4, “Failure effects”. age units) between the appearance of a
potential failure and a functional failure.
Potential failure: An indicator that a S/A (secondary action) Indicator: same
failure mode has initiated. meaning as “Potential failure” in RCM.
No equivalent concept in RCM. If a failure Category: related to Failure Mode. Takes
mode is due to design, lubrication, one of six possible values: 1) Design, 2)
overload, or maintenance practices, they Lubrication, 3) Normal Operation, 4)
would each constitute a separate failure Overload Condition, 5) Maintenance
mode, and this information would be practices, or 6) Safety
included in the failure mode description
itself. The word “Safety” or
“Environment” is not mentioned until the
consequence phase of the RCM logic
diagram.
RCM records this information in the free Characteristic: related to Failure Mode.
text answer to question 4, “Failure Takes one of three possible values: 1)
effects”. However no explicit data Definitive life/wearout, 2) General
structure is specified by the JA1011 degradation, and 3) Random
standard.
Consequences: Question 5. Takes one of Consequence: related to Failure Mode.
four possible values: 1) Hidden, 2) Safety Takes one of four possible values: 1) Total
/Environmental, 3) Operational, and stoppage, 2) Partial stoppage/quality, 3)
4)Non-operational. No immediate effect, or 4) No effect. This
information
RCM records RCM Turbo’s
“Consequence” in the free text answer to
question 4 “Failure effects”.
RCM records this information both in the Measurability: related to Failure Mode.
free text answer to Question 4 “Failure Takes one of three possible values: 1) Easy,
effects” and in the answer to Question 6 2) Moderate, or 3) Impossible
“Tasks”. Q6 asks whether there is
an applicable CBM task. Once a (CBM or
other) task is found to be applicable
(practical) RCM then asks whether it will
be effective. That is, will it sufficiently
reduce or entirely avoid the
consequences of failure at acceptable
cost?
Redesign: RCM records this information Design Notes: related to the Failure Mode.
in the free text answer to question 7, Records decision/recommendation to
“Default Tasks”. Differs from RCM Turbo “design-out” the failure mode. (strictly
only in the sequence in which this speaking it is presented out of “RCM
question appears (i.e. following a sequence”.)
determination that no proactive or failure
finding task adequately mitigate the
consequences of the failure.)
RCM provides no specific field for this Strategy Notes: related to Failure Mode. A
information, leaving its provision up to free text field used to store comments or
the implementer or commercial notes on the chosen maintenance
packager. strategy.Useful where a second or
alternative strategy has been considered
and rejected.
RCM records this information in the free Breakdown Action: related to Failure
text answer to question 4, “Failure Mode. Describes what must be done
Effects”. However, without an explicitly to repair the functional failure. Also has
specified structure. the specific fields: Work Order
No., SOP, Duration, Downtime, MI
Status, S/A Initiator, Resources (up to six
steps), Assumptions, Materials, Spares.
RCM develops this information in the Primary Action: Related to the Failure
decision algorithm of question 5 (Is there mode. Describes what should be done
an on-condition maintenance task that is to prevent the failure mode. Also has the
both applicable and effective?) The RCM specific fields: Work Order
standard does not elaborate an explicitly No., SOP, Duration, Downtime, MI
specified structure for recording this Status, S/A Initiator, Resources (up to six
information. steps), Assumptions, Materials, Spares.
RCM records this information in the free Secondary Action: related to Failure Mode.
text answer to question 6, “Tasks”. The Describes what must be done following
RCM standard does not elaborate an the detection of a potential failure. Also
explicitly specified structure for recording has the specific fields: Work Order
this information. No., SOP, Duration, Downtime,MI
Status, S/A Initiator, Resources (up to six
steps), Assumptions, Materials, Spares.
RCM records this information in the free Overhaul Action: related to Failure Mode.
text answer to question 4, “Failure Records Overhaul Maintenance
Effects”. The RCM standard does not actions. For example, where the
elaborate an explicitly specified structure Secondary Action was the change-out of a
for recording this information. rotable item which itself requires
subsequent overhaul. Also has the specific
fields: Work Order
No., SOP, Duration, Downtime, MI
Status,O/H Venue, S/A Initiator,
Resources (for up to six
steps), Assumptions, Materials, Spares.
Not called a “library”. However, the Failure Data Library: a table of “3 part”
records are accessible (structured as failure modes referenced by Machine
answers to the seven questions) in the Type. An administration process is used to
RCM worksheets comprising the global control the quality of data from multiple
RCM table. No corporate harmonizing sites and harmonize it for the purpose of
process need be applied because every providing“templates” where applicable in
record is a “one-off” development. future analyses of other MIs or PUs. The
However, tools, training, supervision and focus on “templating” justifies the
support are required to validate and appellation “Streamlined” in the case of
maintain and update the knowledge base RCM Turbo.
with day-to-day experience.
“Templating” of an entire item, is,
nonetheless, possible by copying any or
all records of an item after carefully
comparing their respective operating
context descriptions.
We may conclude from Table 14-1, that, although RCM Turbo refers to itself as
a streamlined process, and, that some of its terminology differs from that of RCM, it does
not omit any vital knowledge element specified by the SAE RCM minimum criteria
standard. RCM Turbo does deviate from the sequence stipulated in the standard. As
pointed out in Chapter 11 (page 137), in practice, however, RCM is not a sequential
process. RCM analysts anticipate the answers to subsequent questions while working the
current question. Furthermore, the RCM process is iterative. That is, the analysts often
return to a previous answer and adjust it in the light of revelations further on in the
process. The iterative and non-sequential nature of the RCM process tends to render less
important the differences between the two approaches.
The terminology comparisons of Table 14-1 show that RCM Turbo expands the
information elements of RCM into greater structural detail. Such data structuring
facilitates the post-RCM processes (included in the RCM Turbo software package) of
workload smoothing, frequency calculations, and CMMS integration as well as integration
with a spares optimization (optional) package.
Figure 14-1 of Example 1 shows how the RCM Worksheet of Chapter 11 (Figure 11-2 page
138) might be combined with the extended data fields of RCM Turbo.
Example 1
Q5, Q6, Q7
Function FailureFailureEffects
Statement mode
Q1
Q2 Q4
Q3
To feed Does Shaft Unit Output Reduction: Total
material not fails stoppage,
24 feed
hours/day at all PU Downtime Cost: $11,390
/ hour,
F/mode&Effects: Shaft
failure-Chemical corrosion,
overtorque, indicated by
cracks, increase in vibration
leading to shutdown of
Brownstock washer
Measurability: Moderately
easy to monitor
Q5, Q6, Q7
Function FailureFailureEffects
Statement mode
Q1
Q2 Q4
Q3
Typical Warn Time: 4 Weeks
MTBF: 5 years
Consequence: Total
stoppage
Strategy: CBM
We may conclude from Table 14-1, that, although RCM Turbo refers to itself as
a streamlined process, and, that some of its terminology differs from that of RCM, it does
not omit any vital knowledge element specified by the SAE RCM minimum criteria
standard. RCM Turbo does deviate from the sequence stipulated in the standard. As
pointed out in Chapter 11 (page 137), in practice, however, RCM is not a sequential
process. RCM analysts anticipate the answers to subsequent questions while working the
current question. Furthermore, the RCM process is iterative. That is, the analysts often
return to a previous answer and adjust it in the light of revelations further on in the
process. The iterative and non-sequential nature of the RCM process tends to render less
important the differences between the two approaches.
The terminology comparisons of Table 14-1 show that RCM Turbo expands the
information elements of RCM into greater structural detail. Such data structuring
facilitates the post-RCM processes (included in the RCM Turbo software package) of
workload smoothing, frequency calculations, and CMMS integration as well as integration
with a spares optimization (optional) package.
Figure 14-1 of Example 1 shows how the RCM Worksheet of Chapter 11 (Figure 11-2 page
138) might be combined with the extended data fields of RCM Turbo.
Example 1
Figure 14-1 RCM Worksheet applied to a RCM Turbo example
In the RCM worksheet of Figure 14-1 we note that most of the RCM Turbo “failure mode”
fields (in bold) fall quite readily into the RCM Effects column, with the possible exception
of the field “Strategy”. The latter appears to pre-empt the RCM decision logic
of Questions 6 and 7. We view this, nonetheless, as an insignificant departure (from RCM),
given that RCM analysts consider the mitigating task in the normal course describing the
effects of failure. It is essential, however, that the RCM consequences (H, S, O, or M) be
determined and the meticulous decision logic of RCM (on page 171) be applied
immediately following this RCM Turbo step.
RCM Turbo facilitates data entry with a convenient Visual Basic MS Excel form illustrated
in Figure 14-2.
RCM Turbo then will perform a “primary” (i.e. a CBM) task frequency calculation (Figure
14-3) and display the results that 14 days (i.e. half the PF interval) is the recommended
task frequency. RCM Turbo calculates the annualized cost of the CBM program so that it
may be justified by comparison with the annualized economic consequences (based on
the MTBF and the average cost of a failure) avoided by the CBM program.
For scheduled overhaul, discard, and failure finding tasks RCM Turbo performs analogous
calculations by applying a recorded MTBF, a qualitatively estimated hazard function, and
the recorded average economic consequences of failure. The complete set of RCM
Turbo’s data fields is given in Appendix 12 on page 236.
Conclusions
1. Table 14-1 illustrates that streamlined RCM (as it is embodied in RCM Turbo), is not
“streamlined” (i.e. in the sense of being “abridged” or “reduced”). Rather, it encompasses
the principles of RCM, adding features that address CMMS integration, quantitative
reliability assessment and task frequency calculations, spares, workload scheduling and
balancing, and other considerations.
2. RCM Turbo does address the 7 RCM questions, however, not in the sequence
stipulated by the RCM Standard. The software expands the 7 information elements of
RCM into various database fields. For example, MTBF, P-F Interval, and Repair time are
explicit fields related to a Failure Mode.
3. A RCM Worksheet based on the SAE JA1011 standard, will provide excellent team focus
regardless of the software adopted. If populated (perhaps adapted as in Figure 14-1) with
RCM Turbo's needs in mind, the worksheet (incorporating the RCM decision algorithm)
will benefit both streamlined and original RCM users.
4. Both RCM and RCM Turbo demand that the persons (primarily maintainers and
operators), directly impacted by maintenance decisions, participate fully in the process.
Indeed they must drive it. External consultants can only teach the principles and
techniques of RCM. Regardless of the RCM software chosen, the organization must select
its analysts from among its most experienced and competent operators and maintainers.
It must chose a facilitator, from within, who will learn the RCM process fluently, elicit,
and faithfully record the technical knowledge of the analysts. The facilitator must ask the
7 RCM questions and ensure that consensus has been reached. He or she must ask and
ensure that each of the questions along the appropriate branch of the RCM decision tree
are rigorously answered by the team, and duly recorded.
5. Finally, we emphasize that reliability-centered maintenance is not a software
dominated process. Software records the results of RCM analysis in a convenient,
accessible, and auditable format that traces every maintenance task back to a failure
mode that the RCM team identified. Software enables integration with the CMMS and
implementation therein of the RCM analysis results. As importantly, software, through
regular feedback from the field, and integration with the CMMS, supports continuous
“living” enhancement of the initial RCM analysis.
References:
[3] One whose failure has hidden, safety, environmental, or serious economic
consequences.
[4] Typical worst case scenario. A collective judgement that balances the extent of detail
recorded with the gravity and likelihood of the failure consequences.