|
Readiness for RCM: Tuning Your CMMS to Support Reliability Based
Maintenance
By: Ken Bass, CMRP,
Management Resources Group, Inc.
Originally presented at the
Enterprise Asset
Management Summit - collocated with the
Reliability
Centered Maintenance Managers' Forum
Most providers of Reliability Services emphasize the importance of
utilizing your Computerized Maintenance Management System (CMMS) or
Enterprise Asset Management system as a key element of an overall
reliability strategy. This is because a robust and revitalized CMMS
or EAM provides both short-term and long-term benefits to the
organization. Some of the long-term benefits impact the
organizations ability to perform more robust Reliability Centered
Maintenance (RCM) Analysis and Root Cause Failure Analysis (RCFA).
An additional long-term benefit of a revitalized CMMS/EAM is to
provide better metrics for evaluating all aspects of the health of
your reliability improvement initiative.
Where we are
today
When beginning work with a new client, we often find that the data
in the current CMMS/EAM has been entered in an uncontrolled manner.
We find that little forethought has been given to establishing
consistent form and format for data structure. Functional location
hierarchy has not been well laid out and naming conventions for
classes/sub-classes of equipment has not been standardized. If
manufacturer and model data has been entered in the system, it is
often wrong or out of date due to the lack of an adequate Management
of Change process.
Very often we see that one of the CMMS/EAM systems most powerful
tools is underutilized. The ability to review equipment failure
data is one of the most important parts of a Reliability Engineer’s
functions. Here again, we see that the use of Failure Coding and
narrative history of failures is not being used or has not been
standardized. The Reliability Engineer’s ability to trend valuable
data is all but lost.
Another powerful function of most CMMS/EAM Systems is the ability to
collect data for the business metrics or Key Performance Indicators
(KPIs). Once again, the lack of forethought in setting up the
CMMS/EAM makes the ability to capture metrics difficult or near
impossible. The confidence in whatever metrics are collected is
certainly low.
Revitalizing
Your CMMS/EAM
The power a CMMS/EAM System brings to an organization is not what it
allows you to enter into the system, but in what information you can
get back out. Any revitalization effort should begin with an
analysis of what information the organization needs to retrieve from
the CMMS/EAM. The input should then be structured to facilitate
that data recovery.
Most organizations initially focus their revitalization efforts in a
few critical areas. Focus on these few areas generally results in
sufficient short-term advantages and also yields sufficient
long-term credible data to assist in the continuous improvement of
the reliability initiative. The results from these efforts produce
sufficient improvements and efficiencies for the organization that
the cost of the revitalization is easily justified.
Since the CMMS/EAM is most commonly used as a vehicle for rolling up
cost and accounting information, one of the most important
revitalization efforts includes a logical restructuring of the
Functional Location Hierarchy. Knowing what equipment by
class/sub-class and by manufacturer and model is the next important
element. Start with the lowest element that provides a discrete
function (the Functional Location). Build up through the
hierarchy. Consider how data will be extracted by Maintenance,
Operations and Finance personnel. The elements you select above the
Functional Location depends both on the organization of the plant
and how the data extracted will be used. Some users employ a
straightforward Functional Location > System > Process > Plant
structure. Some users employ a Functional Location > Building
Location (geography) > Building > Plant system. The more advanced
CMMS/EAMs may allow both. Care must be taken to avoid creating too
many layers to the hierarchy. Too many layers become confusing and
slow down reporting functions. You must decide the granularity of
the data you will be extracting before establishing the hierarchy.
Do not assume the data in the current CMMS/EAM is accurate. This is
especially true if there is not a robust Management of Change
process in place. The facility must be walked down to accurately
collect equipment nameplate data and to accurately classify the
equipment by class/sub-class. Revitalization of this information is
extremely time consuming and can be expensive if performed by a
contractor, but it is an essential part of any CMMS/EAM
revitalization effort.
Completion of these two tasks provides the organization with the
short-term benefit of being able to accurately group common
maintenance tasks to like equipment. In addition, the improved
Functional Location Hierarchy assures a more accurate roll up of
costs to the proper cost centers.
The third area of focus for most organizations undergoing CMMS/EAM
revitalization centers on the ability to capture failure data, work
type classification data and equipment history narratives. Here
again, the effort begins with a discussion of what data will be
retrieved, how often and why. Consideration needs be given to such
elements as the granularity of cost roll up and the metrics
associated with the financial goals, work performance, and cultural
change that are to be tracked. Metrics associated with PM or PdM
program compliance must be addressed.
In addressing these three distinct areas, an organization greatly
improves their ability to move forward in a Reliability Based
Maintenance model.
The Reliability Centered Maintenance and Root Cause Failure Analysis
Advantage
Most organizations embarking on a Reliability Based Maintenance
initiative must put RCM, in one form or another, in their toolbox of
reliability tools. RCM, by its definition, is a highly structured
and quantitative analytical technique. But in the early part of
their journey, many organizations find that their CMMS/EAM does not
contain enough credible data to support a truly quantitative
approach to RCM. When addressing such elements as frequency of
failure, cost of repair and time to restore, they often find they
must use a “best estimate” approach based on input from
knowledgeable personnel rather than relying on hard data pulled from
the CMMS/EAM. That is not to say that there is anything wrong with
using the “best estimate” approach. An RCM performed with this data
is still likely to provide a maintenance strategy far better than
one without an analyzed basis.
As data is entered into the CMMS/EAM using predefined formats,
established failure coding, and assigned to the proper assets; the
organization is better able to extract statistically valid data on
Mean Time Between Failure (MTBF). This data can be used in a number
of ways within the RCM Process. If the organization believed that
the “best estimate” of failure frequency was reasonably accurate,
then the analysis of the new data can be used to validate the
original “best estimate” or measure the improvement in performance
of the equipment based on the improved maintenance strategy.
Another way that the credible data can be used to improve the RCM
and overall maintenance program is to adjust task selection. As an
example, it may have been appropriate to select a Failure Finding
task to identify the onset of a failure mode. This action was
probably taken because there was insufficient data to accurately
(and with confidence) describe the failure interval within a narrow
range. Because of the lack of credible data, a shorter task
interval may be selected to assure that the majority of potential
failures are found prior to functional or hard failure of the
component. The selection of this type of intrusive task may in and
of itself, result in infant mortality type failures of the
component. With the availability of better, more credible data, it
may be possible to more accurately define the MTBF and also to
accurately calculate the variance of the data. If it proves that
the variance is small, it may be possible to replace several Failure
Finding tasks with a single Repair/Replace task thereby reducing
total equipment downtime and eliminating some of the potential for
the introduction of infant mortality type failures. The change to
Repair/Replace type tasking requires highly accurate, statistically
valid data. Only a well-structured CMMS/EAM that is being
efficiently utilized can provide this type of data.
A well-structured CMMS/EAM also allows the Reliability Engineer to
“slice the pie” in a number of different ways to facilitate
analysis. Are our maintenance costs higher in one part of the
facility versus another? Is our failure rate for pumps different
for Model X versus Model Y? Can we apply common tasking across all
centrifugal pumps or must we be manufacturer/model specific? At an
enterprise level we can ask similar questions. Does Plant A
experience the same problems as Plant B with a specific
manufacturer/model? Is the MTBF for all Rotary Valves similar at
all plants? Only with a common and well-defined CMMS/EAM set up are
these comparisons possible and credible. This type of high
quality data also supports trending of performance in multiple ways
that assist the Reliability Engineer in understanding how the
reliability initiative is progressing.
In the same way as RCM, Root Cause Failure Analysis often requires
the analysis of failure data to provide the best results. If that
failure data cannot be obtained from the CMMS/EAM system, once again
we are forced to rely on “best estimate” data from our knowledgeable
staff. This data is not only used in analyzing for root cause but
is also important in the formulation of the corrective actions
resulting from the analysis. When “best estimate” is the only data
available, it must be used. However, it may on occasion bring into
question the validity of the analysis. A revitalized CMMS/EAM
starts to build the credible data necessary to achieve continuous
improvement of an organization’s RCFA Process. Is the failure in
question isolated to a single part of the plant? Is it limited to a
particular manufacturer/model? If it is limited to a specific
condition, what is different about other components in similar
applications? Once again, the ability to sort data is the key to
successfully completing a robust RCFA.
Metrics and
Key Performance Indicators
It is often said that what gets measured, gets done. This is
clearly the case with reliability improvement initiatives. There
are generally high front-end costs with reliability improvement
processes. Management at various levels of the organization wants
to know and needs to know when and if that investment is paying
off. In the past, the principle metrics were those associated with
total maintenance spend or perhaps some of the elements around
overall plant throughput. The problem with indicators like these is
that they are lagging indicators. By the time the quarterly or
annual results were tabulated, middle management could do little but
read the numbers and hope for an improvement. Senior management
waited for the results also and hoped for a positive trend. Often
these were the only metrics used because they were the only metrics
for which reliable data could be easily obtained. Not that these
indicators are not important, but if only lagging indicators are
used, the organization may find itself “trying to close the barn
door after the horse is out”.
Through the use of a revitalized CMMS/EAM, the organization can
identify and use several key leading indicators (in addition to the
lagging indicators) to take better control of the reliability
improvement process. Metrics can be developed in three principal
areas: cost reporting, machinery performance, and cultural change
management.
As stated earlier, cost reporting has always been a principal metric
for most organizations. Generally, this has been a total cost roll
up (total maintenance spend) and is regarded as a lagging
indicator. With a revitalized CMMS/EAM, the cost reporting can be
much more granular and sorted in a number of different ways.
Segregating costs by work type, plant area or process, craft or
skill type are all possible. This depth of granularity allows the
Maintenance Manager to better see where the money is truly going.
It also allows the Reliability Engineer to evaluate the cost
benefits of specific reliability initiatives. Reports from the
CMMS/EAM can be automated and run frequently so that these
indicators become leading indicators. Because the reports are
automated from a “credible” data source, the results can be relied
upon and the trends used with confidence. The granularity of the
cost reporting from a revitalized CMMS/EAM is also invaluable in
budget development. In addition, the higher quality data now
available makes it easier to defend budget numbers.
The revitalized CMMS/EAM also allows for a more granular look at
metrics around machinery performance. The Reliability Engineer has
the ability to analyze machinery performance on many different
levels. Not only can the MTBF be accurately calculated but also the
statistical variation in the data can be evaluated with confidence.
Machinery performance by class/sub-class of component, manufacturer
and model, plant or process area can all be measured. At an
enterprise level, these types of indicators can be used to compare
from plant to plant or to compare process performance from site to
site. Based on known machinery performance, the Reliability
Engineer can elect to run some reports more frequently to better
understand the performance of machinery of interest. The reports
from the CMMS/EAM can be easily adjusted as the conditions warrant.
An often overlooked but important element of a reliability
improvement program is that of cultural change management. If the
organization has changed expectations around such issues as who will
enter work requests, planned work backlog, definition of “emergency”
maintenance or other any other element that pertains to how the
total organization behaves in the new reliability model, a
revitalized CMMS/EAM can be used to track performance in these
areas. These metrics need not be permanent but only utilized until
the desired behavior is obtained. These metrics are also considered
leading indicators of an organization’s performance. In the
beginning of a reliability initiative, these metrics need to be
monitored quite frequently to assure that the organization is
behaving in accordance with the new expectations. Allowing negative
performance to occur over too long a period of time sends the
message that “we didn’t mean what we said” and it makes the cultural
change management initiative all the more difficult. Using the
CMMS/EAM to generate these metrics makes it possible to perform the
review more frequently without a large burden on the organization.
Using these metrics, an organization can determine if problems exist
in training for the whole organization, accountability of parts of
the organization, or individual performance issues. Because the
organization will be addressing the problems through the use of
credible data, it is easier to deal with “what” the problem is as
opposed to “who” the problem is.
Conclusion
As stated earlier in this paper, the strength of any CMMS/EAM system
is not its ability to collect data, but the ability to extract
useful information that can be used to make continuous improvement
in all of the organization’s reliability initiatives.
A revitalization effort for any CMMS/EAM should begin with an
analysis of what information the organization will need to extract
from the system.
After it has been determined what information will be required, a
strategy session should be conducted to configure the CMMS/EAM, to
establish form and format for data entry, and define appropriate
status codes for the various functions that will be monitored.
These actions will help assure that the revitalized CMMS/EAM will
support the RCM and RCFA processes, provide reliable input to Key
Performance Indicators and support other important elements of the
reliability improvement initiative.
For further information, please contact Ken Bass of MRG at 203
264-0500 or by e-mail at
bassk@mrginc.net |
™
