Chapter 6: Reliability Centered

Maintenance
6.1 RCM Goals and Principles
 RCM Process and Associated Questions
 RCM Components
 Reactive Maintenance
6.2 Preventive Maintenance
 Predictive Testing and Inspection
 Proactive Maintenance
6.3 Predictive Testing and Inspection Technologies
 RCM Program Effectiveness Measurement Indicators
 Equipment Availability
 INTRODUCTION
• Reliability centered maintenance (RCM) is a
systematic process used to determine what has to
be accomplished to ensure that any physical facility
is able to continuously meet its designed functions
in its current operating context.
• RCM leads to a maintenance program that focuses
preventive maintenance (PM) on specific failure
modes likely to occur. Any organization can benefit
from RCM if its breakdowns account for more than
20 to 25% of the total maintenance workload.
 RCM GOALS AND PRINCIPLES
Some of the important goals of RCM are as follows:
• To develop design-associated priorities that can
facilitate PM.
• To gather information useful for improving the design
of items with proven unsatisfactory, inherent reliability.
• To develop PM-related tasks that can reinstate
reliability and safety to their inherent levels in the event
of equipment or system deterioration.
• To achieve the above goals when the total cost is
minimal.
Many principles of RCM are discussed below:

• RCM is system/equipment focused. RCM is

concerned more with maintaining system function
as opposed to maintaining individual component
function.
• Safety and economics drive RCM. Safety is of
paramount importance, thus it must be ensured at
any cost and then cost effectiveness becomes the
criterion.
• RCM is function-oriented. RCM plays an
instrumental role in preserving system/equipment
function, not just operability for its own sake.
• Design limitations are acknowledged by RCM. The goal of RCM
is to maintain the inherent reliability of the equipment/system
design and at the same time recognize that changes in inherent
reliability can only be made through design rather than
maintenance. Maintenance at the best of times can only achieve
and maintain a level of designed reliability.
• RCM is reliability-centered. RCM is not overly concerned with
simple failure rate, but it places importance on the relationship
between operating age and failures experienced. RCM treats failure
statistics in an actuarial fashion.
• An unsatisfactory condition is defined as a failure by RCM. A
failure could be either a loss of acceptable quality or a loss of
function.
• RCM is a living system. RCM collects information from the results
achieved and feeds it back to improve design and future
maintenance.
• Three types of maintenance tasks along with run-to-failure are
acknowledged by RCM. These tasks are defined as failure-finding,
time-directed, and condition-directed.
• The purpose of the failure-finding tasks is to discover hidden
functions that have failed without providing any indication of
pending failure.
• Time-directed tasks are scheduled as considered necessary.
• Condition-directed tasks are conducted as the conditions indicate
for their need. Run-to-failure is a conscious decision in RCM.
• RCM tasks must be effective. The tasks must be cost-effective
and technically sound.
• RCM uses a logic tree to screen maintenance tasks. This
provides consistency in the maintenance of all types of equipment.
• RCM tasks must be applicable. Tasks must reduce the
occurrence of failures or ameliorate secondary damage resulting
from failure.
 RCM PROCESS AND ASSOCIATED
QUESTIONS
• The RCM process is applied to determine particular
maintenance tasks to be performed, as well as to
influence item reliability and maintainability during
design.
• Initially the RCM process is applied during the design and
development phase and then reapplied, as appropriate,
during the operational phase to sustain an effective
maintenance program based on experience in the field.
• Any RCM process should ensure that all of the following
questions are answered effectively as per their
sequence:
• What are the functions and associated expected
levels of the facility performance in its current
operating context?
• How might it fail to meet its assigned functions?
• What are the reasons for each functional failure
or failure mode?
• What are the effects of each failure?
• How does each failure matter?
• What remedial measures should be taken to
prevent or predict each failure?
• What measures should be taken in the event of
not finding a suitable proactive task?
The basic RCM process is composed of the following steps:

1. Identify important items with respect to maintenance. Usually,

maintenance important items are identified using techniques such as
failure, mode, effects, and criticality analysis (FMECA) and fault tree
analysis (FTA).

2. Obtain appropriate failure data. In determining occurrence

probabilities and assessing criticality, the availability of data on part
failure rate, operator error probability, and inspection efficiency is
essential. These types of data come from field experience, generic
failure databanks, etc.

3. Develop fault tree analysis data. Probabilities of occurrence of fault

events — basic, intermediate, and top events — are calculated as per
combinatorial properties of the logic elements in the fault tree.
4. Apply decision logic to critical failure modes. The decision logic
is designed to lead, by asking standard assessment questions, to
the most desirable preventive maintenance task combinations. The
same logic is applied to each crucial mode of failure of each
maintenance-important item.
5. Classify maintenance requirements. Maintenance requirements
are categorized into three classifications: on-condition maintenance
requirements, condition-monitoring maintenance requirements,
and hard-time maintenance requirements.
6. Implement RCM decisions. Task frequencies and intervals are
set/enacted as part of the overall maintenance strategy or plan.
7. Apply sustaining-engineering on the basis of field experience.
Once the system/equipment start operating, the real-life data begin
to accumulate. At that time, one of the most urgent steps is to re-
evaluate all RCM-associated default decisions.
 RCM COMPONENTS
The four major components of RCM are shown in Fig. 6.1.
These are: reactive maintenance, preventive maintenance,
predictive testing and inspection, and proactive maintenance.
Each component is described below

REACTIVE MAINTENANCE
• This type of maintenance is also known as breakdown, fix-
when-fail, run-to-failure, or repair maintenance. When using
this maintenance approach, equipment repair, maintenance,
or replacement takes place only when deterioration in the
condition of an item/equipment results in a functional failure.
FIGURE 6.1 Components of RCM.
• In this type of maintenance, it is assumed there is an equally likely
chance for the occurrence of a failure in any part, component, or
system.
• When reactive maintenance is practiced solely, a high replacement
of part inventories, poor use of maintenance effort, and high
percentage of unplanned maintenance activities are typical.
• Furthermore, an entirely reactive maintenance program overlooks
opportunities to influence equipment/item survivability.
• Reactive maintenance can be practiced effectively only if it is
carried out as a conscious decision, based on the conclusions of an
RCM analysis that compares risk and cost of failure with the cost of
maintenance needed to mitigate that risk and failure cost.
• A criteria for determining the priority of replacing or repairing the
failed item/equipment in the reactive maintenance program is
presented in Table 6.1.
 PREVENTIVE MAINTENANCE
• Preventive maintenance (PM), also called time-driven or interval-
based maintenance, is performed without regard to equipment
condition.
• It consists of periodically scheduled inspection, parts replacement,
repair of components/items, adjustments, calibration, lubrication,
and cleaning.
• PM schedules regular inspection and maintenance at set intervals to
reduce failures for susceptible equipment. It is important to note
that, depending on the predefined intervals, practicing PM can lead
to a significant increase in inspections and routine maintenance.
• On the other hand, it can help reduce the frequency and severity of
unplanned failures. Preventive maintenance can be costly and
ineffective if it is the only type of maintenance practiced.
 Item/Equipment Monitoring
• The main objectives in monitoring item/equipment condition
are to determine item/equipment condition and to establish
a trend to forecast future item/equipment condition.
The following approaches are useful for setting initial
periodicity:
• Failure anticipation from past experience:
In some cases, failure history of equipment and personal
experience can provide, to a certain degree, an intuitive feel
as when to expect a failure.
• Failure distribution statistics: The failure
distribution and the probability of failure must be
known when statistics are used to determine the
basis for selecting periodicities.
• Conservative approach: The common practice in
the industrial sector is to monitor the equipment
monthly / weekly when good monitoring methods and
adequate information are unavailable. Often, this
leads to excessive monitoring. In situations when
impending failure becomes apparent through the use
of trending or other predictive analysis techniques,
the monitoring interval can be shortened.
 PREDICTIVE
TESTING AND INSPECTION
• Predictive testing and inspections (PTI) is sometimes called
condition monitoring or predictive maintenance.
• To assess item/equipment condition, it uses performance
data, nonintrusive testing techniques, and visual inspection.
• PTI replaces arbitrarily timed maintenance tasks with
maintenance that is performed as warranted by the
item/equipment condition.
• Analysis of item/equipment condition-monitoring data on a
continuous basis is useful for planning and scheduling
maintenance/repair in advance of catastrophic or functional
failure.
• The collected PTI data are used to
determine the equipment condition and to
highlight the precursors of failure in several
ways, including pattern recognition, trend
analysis, correlation of multiple technologies,
data comparison, statistical process
analysis, and tests against limits and ranges.
• PTI should not be the only type of
maintenance practiced, because it does not
lend itself to all types of items/equipment or
possible modes of failure.
 PROACTIVE MAINTENANCE
• This type of maintenance helps improve maintenance
through actions such as better design, workmanship,
installation, scheduling, and maintenance procedures.
• The characteristics of proactive maintenance include
practicing a continuous process of improvement, using
feedback and communications to ensure that changes in
design/procedures are efficiently made available to item
designers/management, ensuring that nothing affecting
maintenance occurs in total isolation, with the ultimate
goal of correcting the concerned equipment forever,
optimizing and tailoring maintenance methods and
technologies to each application.
• It performs root-cause failure analysis and
predictive analysis to enhance maintenance
effectiveness, conducts periodic evaluation of
the technical content and performance interval
of maintenance tasks, integrates functions
with support maintenance into maintenance
program planning, and uses a life cycle view
of maintenance and supporting functions.
• Figure 6.2 presents eight basic methods
employed by proactive maintenance to extend
item/equipment life. Some of these methods
are described below.
FIGURE 6.2 Basic techniques employed by proactive
maintenance to extend equipment life.
 PREDICTIVE TESTING AND
INSPECTION TECHNOLOGIES
• Predictive testing and inspection (PTI) is an
important component of the RCM. This section
describes the PTI technologies in detail.
• These technologies may be described as a variety of
approaches used to determine item/equipment
condition for the purpose of estimating the most
effective time to schedule maintenance.
• These technologies include intrusive and
nonintrusive approaches in addition to using process
parameters to assess overall condition of equipment.
Six PTI technologies/approaches are described below.

1. Vibration monitoring and analysis: One of the most widely

used PTI approaches, it is useful in assessing the condition
of rotating equipment and structural stability in a system.
• The techniques of vibration monitoring and analysis include
spectrum analysis, torsional vibration, waveform analysis,
shock pulse analysis, and multichannel vibration analysis.
• Vibration monitoring effectiveness depends on such factors
as analyst’s ability, complexity of equipment, sensor
mounting, resolution, and data collection methods.
• The vibration monitoring and analysis approach is applicable
to items such as engines, shafts, motors, pumps, gearboxes,
bearings, turbines, and compressors.
2. Electrical condition monitoring: This includes various
technologies and approaches that provide a comprehensive system
evaluation.
• By monitoring important electrical parameters it provides useful
data to detect and rectify electrical related faults such as phase
imbalance, insulation breakdown, and high resistance
connections.
• Electrical faults are costly and present safety concerns because in
systems they are seldom visible.
• Table 6.2 lists several electrical condition monitoring methods.
These methods can monitor equipment such as electrical motors,
electrical distribution cabling, generators, electrical distribution
transformers, electrical distribution switchgear and controllers,
and distribution systems.
• The specific electrical condition monitoring methods for these six
types of equipment are presented in Table 6.3.
3. Thermography: Infrared thermography (IRT)
may be defined as the application of infrared
detection instruments for identifying pictures of
temperature differences (thermogram).
The test instruments used include noncontact,
thermal measurement, line-of-sight, and imaging
systems.
The noncontact nature of the IRT technique
makes it particularly attractive for identifying
hot/cold spots in energized electrical equipment,
large surface areas such as boilers and building
walls, and so on.
• With respect to specific electrical equipment, the IRT
approach can be used to identify degrading conditions
in items such as switchgear, motor control centers,
transformers, and substations.
• Similarly, in regard to specific mechanical equipment,
the IRT technique can help identify blocked flow
conditions in items such as condensers, pipes, heat
exchangers, and transformer cooling radiators.
• One limitation of thermography is that it is limited to
line of sight, and errors can be introduced due to
material geometry, color of material, and
environmental factors such as wind effects, solar
heating, etc.
4. Lubricant and wear particle analysis:
Three reasons for performing this type of
analysis are: to assess wear condition of
equipment, to assess the lubricant condition,
and to assess if the lubricant is
contaminated.
• The test used for the above purposes will
depend on factors such as cost, sensitivity
and accuracy of the test results, and the
equipment construction and application.
5. Passive (airborne) ultrasonics: Airborne ultrasonic
devices (AVD) function within the frequency spectrum of 20
to 100 kHz and heterodyne the high frequency signal to the
audible level so that the operator is able to hear changes in
noise associated with leaks, corona discharges, etc.
Two typical examples are bearing ring and housing
resonant frequency excitation due to inadequate lubrication
and minor defects. Some specific equipment application
examples are: heat exchangers, boilers, and bearings.
One of the main limitations of the airborne ultrasonics
(AUs) technique is that AUs are subjective and dependent
on perceived differences in noises.
6. Nondestructive testing: This technique can
determine material properties and quality of
manufacture for high-value parts/assemblies
without damaging the product or its function.
Usually, nondestructive testing (NDT) is practiced
when approaches such as destructive testing are
cost-prohibitive or ineffective.
NDT is associated with welding of large high-
stress parts such as pressure vessels and
structural supports. In addition, oil refineries and
chemical plants use NDT methods to assure
pressure boundaries’ integrity for systems
processing of volatile substances.
• NDT techniques include: ultrasonic testing (imaging),
magnetic particle testing, dye penetrant inspections,
hydrostatic testing, eddy current testing, and
radiography.
• Prior to implementation of an NDT program, it is
recommended that a formal plan be developed. It would
incorporate factors such as the technique to be used,
number and orientation of samples, frequency, location,
the failure mode each sample should address, and the
information to be gained from each sample.
• The interval between inspections and the location of
sampling points are two of the more difficult variables to
address.
In the case of time interval between inspections, in establishing
sample intervals or frequency, the factors that must be examined
include system operating cycle, type of contained substance, major
corrosion mechanisms, historical failure rate, expected corrosion rate,
proximity of existing material to minimum wall thickness, erosion
mechanisms, and expected erosion rate.
Similarly in the case of location of sampling points, some of the
guidelines for locating NDT sampling points are as follows:
• Welds, high stress fasteners, and stressed areas
• Areas susceptible to cavitation
• Dissimilar metals’ junctions
•Areas with identified accelerated corrosion/erosion mechanisms
• Abrupt changes in direction of flow (elbows) and changes in pipe
diameter
• “Dead-heads”
Table 6.5 presents application areas for specific NDT
techniques. Limitations associated with each NDT
technique are given in Table 6.6.
 RCM PROGRAM EFFECTIVENESS
MEASUREMENT INDICATORS
• Over the years many management indicators to measure the
effectiveness of an RCM program have been developed.
• The numerical indicators or metrics are considered the most useful
because they are objective, precise, quantitative, and more easily
trended than words, as well as consisting of a descriptor and a
benchmark.
• A descriptor may be defined as a word or group of words detailing the
units, the function, and the process under consideration for
measurement.
• A benchmark is a numerical expression of a set goal. Some of the
metrics for measuring the effectiveness of an RCM program are
presented below along with suggested benchmarks.
 EQUIPMENT AVAILABILITY
• This is expressed by

where
EA = equipment availability,
Hea = number of hours each unit of equipment is available
to run at capacity,
THrp = total number of hours during the reporting period.
The benchmark figure for this metric is 96%.
 EMERGENCY PERCENTAGE INDEX
• This is defined by

where
EP = emergency percentage,
Hej = total number of hours worked on emergency jobs,
THw = total number of hours works.

The benchmark figure for this indicator is 10% or less.