An_Evaluation_of_Alternative_Approaches_to_Reliabi.pdf

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/330497901
An Evaluation of Alternative Approaches to Reliability Centered Maintenance
Article in International Journal of Applied Engineering Research · December 2015
CITATIONS
0
READS
786
2 authors:
Some of the authors of this publication are also working on these related projects:
Image Forensics View project
Ontology Based News Generation Framework Using Neural Network Models View project
Deepak Prabhakar P
Mangalore Refinery and Petrochemicals Ltd
7 PUBLICATIONS 20 CITATIONS
SEE PROFILE
Prof.(Dr) V.P. Jagathy Raj
Cochin University of Science and Technology
39 PUBLICATIONS 484 CITATIONS
SEE PROFILE
All content following this page was uploaded by Deepak Prabhakar P on 21 January 2019.
The user has requested enhancement of the downloaded file.

International Journal of Applied Engineering Research ISSN 0973-4562 Volume 10, Number 19 (2015) pp 40350-40359
© Research India Publications. http://www.ripublication.com
40350
An Evaluation of Alternative Approaches to Reliability Centered
Maintenance
Deepak Prabhakar P
Research Scholar, Dept. of Management Studies and Research, Karpagam University, Coimbatore &
Deputy General Manager (Mechanical), Mangalore Refinery & Petrochemicals Ltd, Mangalore (Email- deepakani@gmail.com)
Dr. Jagathy Raj V.P.
Professor, School of Management, Cochin University of Science & Technology, Kochi (Email – jagathy@cusat.ac.in)
Abstract
Reliability Centered Maintenance (RCM) is a Maintenance
Strategy that was developed in the 1950s and has been
successfully adopted in the Airline and Military sectors for the
past many decades. However, the classical approach to RCM
is seen as highly rigorous and time consuming for the general
industries, leading to its poor adoption. Many alternatives,
while attempting to maintain the core tenet of RCM, have
tried to provide a simpler implementation or an approach that
is less rigorous than the classical RCM. These approaches too,
have not found wide application due to various reasons. This
paper lists the various alternatives proposed, develops a
baseline for evaluation, and finally evaluates the approaches
on the parameters developed, so that a clear understanding of
the options are available to those who are interested in
adopting one of these alternative approaches to RCM.
Key Words: A-RCM, Maintenance Strategy, RCM
Alternatives, RCM
1. Introduction
Reliability Centered Maintenance (RCM) is a broad strategy
for managing the maintenance and reliability requirements of
complex systems. The system was developed in the 1950s and
has found its application in the airline industry and in the US
military. RCM involves the systematic evaluation of potential
failure modes and have in place actions that aim to prevent or
predict failures. Further the strategy also calls for design
changes when a situation where failures can neither be
predicted nor prevented is encountered. The approach to
implementation of RCM has remained constant and even
today the ‗standard‘ method of implementation as defined in
the SAE-JA 1011 [1] is largely the same as the approach
originally proposed by Nowlan and Heap [2]. The so called
RCM-II approach of Moubray [3] is also nearly identical.
While this rigorous approach has paid rich dividends in the
airline industry, which has seen extremely high ‗mission‘
reliability, the complexity and high resource intensiveness of
classical RCM has resulted in its limited adoption by other
industries.
However, the fact remains that the principles of RCM can
effect dramatic improvement in the reliability and the best
approach to achieving the 100% threshold in reliability [4].
Due to this understanding researchers and practitioners have
developed varied alternatives to RCM that, while keeping the
core philosophy of RCM intact attempts to address and
overcome the limitations of classical RCM.
While there have been many alternatives proposed, there has
been no real attempt to evaluate these on a common baseline.
The authors attempt to carry out such an evaluation in this
paper.
This paper is presented in three sections. In the first, after
extensive literature review the alternatives to RCM which
were evaluated are presented. In the second, based on the
expectation in published literature from maintenance
strategies, a baseline for evaluation which was formulated is
presented and in the third section, the evaluation of the
various alternatives which was carried out using the baseline
formulated is elaborated.
2. Literature Review Methodology
Many alternative approaches to RCM have been proposed.
These have largely focused on the optimisation, streamlining
and simplifying. As a first step in the analysis an extensive
literature review was undertaken to understand the published
literature on the alternatives to RCM. The search was
conducted using a two step process. In the first step search
terms ―RCM Approaches‖, ―RCM Alternatives‖, ―RCM
Methods‖, ―RCM‖ were used. From the results that these
searches yielded, further linkages were obtained and search
done on terms ―Maintenance Strategy‖, ―Maintenance
Optimisation‖, ―Streamlined RCM‖, and ―RCM
implementation‖. From the scan of the results, results that
were simply reportage of RCM implementation and
calculations based on RCM implementations were eliminated.
Papers based on statistical calculations were also eliminated.
Literature that provided a clear description of the approach as
well as those that supported the search by providing additional
references were perused and key summaries were extracted.
For the full fledged method, the originally referenced paper
needs to be perused, as the attempt here is not to provide a
primer of the alternative methods but to introduce the
alternative and then to evaluate the same based on a
methodology so created for it.

40351
3. Alternative Approaches to RCM
There have been many attempts to define and develop
alternative approaches to the 'classical' RCM process. Selvik
and Aven [5] report that ―several methodological
improvements of the (RCM) method have been suggested, e.g.
PM Optimization, RCM 2, Stream-lined RCM, Intelligent
RCM Analysis and also a so-called probabilistic approach by
Eisinger and Rakowsky‖.
Pride elaborated on RCM alternatives as ―there are several
ways to conduct and implement an RCM program. The
program can be based on rigorous Failure Modes and Effects
Analysis (FMEA), complete with mathematically-calculated
probabilities of failure based on design or historical data,
intuition or common-sense, and/or experimental data and
modeling. These approaches may be called Classical,
Rigorous, Intuitive, Streamlined, or Abbreviated. Other terms
sometimes used for these same approaches include Concise,
Preventive Maintenance (PM) Optimization, Reliability
Based, and Reliability Enhanced‖ [6].
Extending this classification by Pride, the alternative
approaches to RCM are presented here as following the five
broad categories: one – a mix of approaches, two –
simplification of analysis, three – optimization approaches,
four – broad strategies that provide complete methodologies
of implementation, and five – mathematical models that
attempt to change one part of the RCM methodology.
3.1. Mix of Approaches
A common approach followed by practitioners in the industry
is that of following a mix of different approaches. This section
highlights some of these approaches.
Bloom has put forth an alternative approach to the RCM
implementation process. Here the approach centers on the
Consequence of Failure Analysis (COFA) as the guiding
point. He describes the alternative process steps as follows:
―1) Describing the component functions (where all functions
of the equipment are defined, 2) Describe the functional
failures (against each of the functional failures) 3) Describe
dominant component failure mode for each function failure
(where only plausible and realistic failure modes are included)
4) Assess whether the occurrence of the failure mode is
evident (by this he means whether the failure of the
component can be made evident by a control or detection
system) 5) Describe the system effect for each failure mode
(wherein the effect, functional statutory, safety etc. are listed)
6) Describe consequence of the failure based on the asset
reliability criteria 7) Defining component classification
(where the final decision has to be entered into as critical or
run to failure)‖[4].
Mokashi, Wang and Vermar report that ―there are other
approaches, which thus cannot be called RCM. They are,
however, based on the same principles and have delivered
reliable positive results. One such approach is risk-centered
maintenance or Risk-CM. NASA has in its RCM guide said
that one of the primary principles of RCM is that RCM uses
logic tree to screen maintenance tasks that is, it uses broad
categories of consequences of failure to prioritize failure
modes. However, Risk-CM uses a combination of probability
and consequence, that is, risk to prioritize failure modes. This
gives a finer failure mode ranking‖ [7].
Jones put forward Risk Based Reliability Centered
Maintenance (RBCM), a new variance of basic RCM.
―Basically, RBCM can be described as RCM, but with a
strong statistical background. This tackles and eliminates the
drawback of the ad hoc FMEA of the traditional RCM
approach. Risk based inspections (RBI) are one of the core
concepts here. The RBI methodology enables the assessment
of the likelihood and potential consequences of pressure
equipment failures. RBI provides companies with the
opportunity to prioritize equipment inspections and optimize
the inspection methods, frequencies and resources.
Furthermore, RBI helps to develop specific equipment
inspection plans and enable the implementation of RCM as
such. This results in improved safety, lower failure risks,
fewer forced shutdowns, and reduced operational costs. The
risk-based approach requires a systematic and integrated use
of expertise from the different disciplines that affect plant
integrity. These include design, materials selection, operating
parameters and scenarios, and understanding of the current
and future degradation mechanisms and of the risks
involved‖[8]. RBCM is focused on risk. This is a method that
can help prioritize the maintenance interventions.
Kelly developed a Business-Centered Maintenance (BCM), a
concept for determining a detailed maintenance plan. Kelly
emphasized the importance of identifying, mapping and
auditing the maintenance function. The BCM concept also
pays attention to the necessary administrative support. Kelly
calls his approach a BUTD approach, bottom-up/top-down
approach. ―First, it is a top-down step that starting from the
business context, the exact objectives for maintenance are
outlined considering all corporate level. The second step is a
bottom-up step. It aims at establishing a life maintenance plan
for all equipments. In a third and last step, all item life plans
are fitted in a maintenance strategy‖ [9]. Applying BCM thus
results in a detailed maintenance schedule, ready for use. The
major disadvantage of this approach is that it focuses only on
developing a schedule or a PM plan.
Selvik and Aven introduce the concept of uncertainty as
opposed to probability and state ―the traditional RCM
approach can be viewed as founded on a risk perspective
where risk is equal to the expected value or the combination
of probabilities and events/losses. To take into account
uncertainties as indicated above, we need to base the RCM on
a broader risk perspective and one way to do this is to replace
probability with uncertainty in the definition of risk‖[5]. They
further introduce a new model known as RRCM, which is ―a
framework based on the existing RCM, which improves the
risk and uncertainty assessments by adding some additional
features to the existing RCM methodology. An extended
uncertainty assessment is added, to address uncertainties
‗‗hidden‘‘ in assumptions of the standard RCM analyses. The
uncertainties are then communicated to management through
an extended uncertainty evaluation, which integrates the
results from the FMECA (and the formal maintenance
optimization if optimization models are established) and the
separate uncertainty analysis. An essential feature of the
presented framework is the managerial review and judgement,
which places the decision process into a broader management
context. In this step consideration is given to the boundaries
and limitations of the tools used.

40352
Khan and Haddara reported on a methodology, called risk-
based maintenance (RBM) that is based on integrating a
reliability approach and a risk assessment strategy to obtain an
optimum maintenance schedule. First, the likely equipment
failure scenarios are formulated. Out of many likely failure
scenarios, the ones, which are most probable, are subjected to
a detailed study. Detailed consequence analysis is done for the
selected scenarios. Subsequently, these failure scenarios are
subjected to a fault tree analysis to determine their
probabilities. Finally, risk is computed by combining the
results of the consequence and the probability analyses. The
calculated risk is compared against known acceptable criteria.
The frequencies of the maintenance tasks are obtained by
minimizing the estimated risk [10].
Prabata and Wiyana presented a case where RCM and RBI
methodology was applied together on a compressor. This was
on a single equipment and they did not extend this further
[11].
Abid, Ayb, Wali and Tariq presented an alternative approach
to RCM ―in which RCM is integrated with life data analysis in
order to accurately estimate the failure mode followed by each
component of the system‖[12]. They state that ―using this
technique a better failure management policy is developed
keeping in view the health of each equipment. This RCM plan
helps to optimize reliability of the system while being cost
effective and decreasing the system downtime‖[12]. However
this was demonstrated for a few equipment and not for a large
group of equipment.
3.2. Simplification of Analysis
Another common methodology is simplification of the process
by eliminating one or more steps in the classical RCM. This
section describes these approaches.
Endrenyi et al escribe an alternative approach to RCM called
Preventive Maintenance Optimization (PREMO). They
describe this as based on ―task analysis rather than on system
analysis. This approach is claimed to have the capability of
drastically reducing the number of maintenance tasks‖ [13].
Mokashi, Wang and Vermar report about a method called
PMO2000 ―PMO2000 has tried to address the problem of
high resource demand, especially in the analysis of failure
modes. In this approach the failure modes are identified by
analyzing the maintenance tasks. For example if the
maintenance task was to ‗‗perform vibration analysis on the
gearbox‘‘, then the failure modes analyzed would be to ‗‗gear
wears or cracks, gear bearing fails due to wear, gear box
mounting bolts come loose due to vibration and gearbox
coupling fails due to wear‘‘. These failure modes are then
passed through the RCM logic tree [7].
Bevilacqua and Braglia refer to a case where the internal
methodology developed by the company to solve the
maintenance strategy selection problem for the new IGCC
plant is based on a ―criticality analysis‖ (CA), which may be
considered as an extension of the FMECA technique [14].
This analysis takes into account the following seven
parameters:
1. Safety;
2. Machine importance for the process;
3. Maintenance costs;
4. Failure frequency;
5. Downtime length;
6. Operating conditions;
7. Additional evaluation for the machine access
difficulty
Zajicek and Kamenicky proposed a methodology to improve
effectiveness of RCM. This method prescribed a) Better team
time organization b) Use of standardised Maintenance Plans
and c) Analysis of only selected components [15].
3.3. Optimization Methods
Another alternative approach is that of Maintenance
Optimisation (MO). This has been described in detail by
Dekker [16], Turner [17], Berger [18], Idhammer [19] and
Dotzlaf [20].
Maintenance optimization is a practice that uses mathematical
models to assist in the decision making process for
maintenance implementation. These models combine
reliability with economics by quantifying costs, benefits, and
various constraints, and integrating the factors into basic
economic methods. These models are particularly helpful for
comparing the cost-effectiveness of different maintenance
policies, determining efficient inspection and maintenance
frequencies, and incorporating numerous constraints into the
decision making process [16]. The traditional optimization
model provides a simple, easy to understand example of how
optimization models work [18], [19]. While the most useful
models will optimize for multiple criteria, the traditional
model only optimizes for one variable – cost [20].
The traditional model is very helpful in understanding the
concept of maintenance optimization; however, it is not as
practical in realistic applications for two reasons: it optimizes
for only one variable and failure trends are rarely accurate.
The optimal maintenance frequency can vary depending on
the variable being optimized; since the traditional model only
optimizes for one variable, it could lead to incorrect
conclusions and poor decisions for maintenance scheduling
[18]. However, due to the fact that components rarely fail after
a predictable time, it is very difficult to accurately depict
equipment failure trends [19].
The models have the advantage that these provide a
quantitative approach for identifying the most efficient
balance of resource expenditures and maintenance benefits
[16]. When analysis reveals no optimal solution, these models
help determine candidates for reactive maintenance and the
tasks to be eliminated [17]. Similarly, these models can help
identify which systems could be more efficiently managed by
simpler or more advanced technology. During development,
optimization models help users understand how to predict
equipment life more accurately, which data to collect, and
how to assess the level of risk for a given maintenance
frequency [17], [19]. While maintenance optimization models
have obvious benefits, there are a lot of difficulties in
application that can make the benefits hard to realize. These
difficulties are among the numerous disadvantages of
maintenance optimization models. Maintenance optimization
models require massive amounts of performance and failure
data that is often hard to obtain; maintenance craftsman may
have significant knowledge about these aspects of the
equipment, although it is often difficult to translate this

40353
knowledge into data [16]. When data is available,
optimization requires a lot of detailed calculations that can be
time consuming, hard to standardize, and difficult to validate.
Further yet, the results of these calculations are rarely useful
because a large amount of guesswork must be used to
compensate for missing data or lack of expert knowledge [17].
Optimization calculations require the user to quantify all
factors, to include the benefits of maintenance; however,
many of the necessary factors are very subjective in nature
and difficult to quantify [16]. Therefore, implementing an
optimization model for an entire maintenance program with
numerous pieces of equipment and systems is rarely feasible;
the common trade-off, which often leads to suboptimal
outcomes, is a simplified approach that does not consider all
factors [21]
Besnard, Fischer and Bretling report on the Quantitative
Maintenance Optimization (QMO) techniques as that they are
―are characterized by the utilization of mathematical models
which quantify both, the cost and the benefit of maintenance
and determine an optimum balance between these. The task in
QMO is often to find the minimum total cost consisting of the
direct maintenance costs, e.g. for labour, materials and
administration, which increases with the intensity of
maintenance actions, and the costs resulting from not
performing maintenance as required, i.e. due to loss of
production and due to additional labour and materials after
component breakdowns‖ [22].
3.4. Broad Strategies
In addition to these approaches, there are broad strategies that
encompass the entire maintenance umbrella and can be used
as stand-alone alternatives to RCM unlike other approaches
described in the preceding sections. This section describes a
few such alternatives developed.
Bae, Koo, Son, Park, Jung, Han and Suh [23] proposed an
alternative algorithm to RCM. The proposed RCM planning
method (RCMP) comprises two optimization steps. The first
step uses the reliability matrix to minimize the total
maintenance cost while, at the same time, maximize the
subsystem reliability. This is achieved by using a multi-
objective optimization method. From this the maintenance
cost function can reflect the current maintenance
characteristics of the components by generating essential cost
factors defined by the reliability and maintainability of each
component. This method which was more mathematical and
model building in nature, defines the reliability function of the
system by using a reliability network between appropriate
subsystems and components, which mimic an artificial neural
network. The second optimization step allocates the
maintenance reliability of each component to the maintenance
cost, reliability function, and desired subsystem reliability. In
the case of maintenance reliability allocation, the optimization
process seeks to minimize the maintenance costs whilst
meeting the desired subsystem reliability requirements. This
research applies an evolutionary algorithm to find the best
reliability allocation by searching for the global optimum in
the nonlinear domain. Finally, Bae, Koo, Son, Park, Jung, Han
and Suh presented a maintenance plan, determined by
estimating the maintenance time of the components as derived
from the allocated reliability and reliability indexes in the
inverse analysis of the fundamental reliability function [23].
Waeyenbergh and Pintelon developed a model called the CIB
model which is also a 7 step process consisting of the
following:
Step 1: Identification of the objectives and resources. Step 2:
Selection of the MISs (Most Important Systems), Step 3:
Identification of the MCCs (Most Critical Components), Step
4: Maintenance policy selection, Step 5: Optimization of the
maintenance policy parameters. Step 6: Implementation and
evaluation, Step 7: Feedback [24]
Cheng, Jia, Gao, Wu and Wang presented an alternative to
RCM called the Intelligent RCM Analysis (IRCMA). This
approach focuses more on the use of an ‗intelligent‘ system
[25]. As it provides approaches that are generic in nature, it is
being classified as a broad strategy.
Besnard, Fischer and Bretling reported the existence of a
strategy called the Reliability-Centered Asset Maintenance
approach (RCAM) which ―is a quantitative approach of RCM
relating preventive maintenance of equipment to system
reliability and total cost. It merges the concepts of RCM and
QMO and in this way overcomes the drawbacks of the two
separate approaches. The RCAM approach is a structured
method originally developed for a combined analysis of
reliability, maintenance, and life-cycle cost of power systems‖
[22].
The three main stages of the RCAM approach are the
following:
―Stage 1: System reliability analysis: defines the system and
identifies critical components
Stage 2: Component reliability modeling: analyses the
components in detail and, based on appropriate input data,
defines the quantitative relationship between reliability and
preventive maintenance measures
Stage 3: System reliability and cost/benefit analysis: places
the results of the component level analysis (Stage 2) in a
system perspective and evaluates the effect of component
maintenance on system reliability and cost‖ [22].
Barbera, Crespo, Viveros and Stegmaier [26] presented an
advanced model for the integral maintenance management
(IMM) ―in a cycle of continuous improvement, which is
aligned with the strategies, policies and key business
indicators. This model claims to use a series of real aspects
needed to convert a theoretical model in a real and useful
maintenance management model. The model claims to take
into account the real or genuine constraints that could limit the
design of preventive maintenance plans and the resources to
do so. It considers the selection of critical spare parts
(inventory cost vs. cost due to unavailability of critical
equipment) and the positive involvement of e-technologies (e-
maintenance) in modern maintenance management on a global
level. In turn, the model consists of seven arranged stages that
follow a logical sequence of action hierarchy and align local
maintenance objectives with the global business objectives; all
these in a framework of continuous improvement using the
principles of the BSC methodology applied to maintenance
management.
The stages defined in this are:
1. Analysis of current situation
2. Ranking of equipment

40354
3. Analysing weakness in equipment
4. Design of maintenance plans
5. Maintenance scheduling and optimisation
6. Control and evaluation
7. Life cycle analysis and replacement‖
The authors had proposed a methodology called Accelerated
Reliability Centered Maintenance (A-RCM) [27]. This
involves a sequential rolling out of the program by
implementing RCM in stages. This methodology approaches
RCM as a step-by-step approach of successive analyses, rather
than the comprehensive approach advocated by the classical
approach. Here the aim is to use all extant programs and then
build on them to implement RCM, unlike the classical
approach that starts afresh. This strategy is also continuously
‗learning‘ by adjusting the program on each failure.
The objective of this method is to provide immediate
improvement in reliability and this method provides for
improvement as soon as or even concurrently as the failure
modes are identified, which takes care of one of the causes of
failure of the conventional RCM process - that of excessive
delay in implementation of actions [28].
This method in effect provides an amalgamation of the
various methods and collates the key features of CBM, TPM
and RCM into one target, that of failure prevention. However
this method is not without its limitations and the most obvious
one is the fact that establishing reasonable likelihood is
dependent on a sequential process which may result in all
potential failure modes not being apparent, at the initial stages
of the implementation. Further this also relies on a continual
system of adding on failure modes and can result in missing
certain key modes, in the event of a lapse in reporting and
analyzing a failure [28].
The authors have separately assessed that the ―A-RCM is a
process that largely follows the RCM process. It differs from
RCM in the methodology of identifying potential failures,
wherein, instead of an FMEA, this process uses a history of
past failures for providing the first round of predictive,
preventive & default actions. This allows quick realization of
reliability improvement in comparison with RCM. This
process, like RCM, is benchmarked through the SAE standard
with the exception of the demand for meeting ‗reasonable
likelihood‘ where this process may not immediately meet the
requirements of the standard. Further, the system allows for
prioritization of effort based on the criticality of the
equipment in consideration. The skill required is comparable
or lower than that required for RCM. The system builds in
continual improvement as part of the system itself. The
disadvantages of the system are that, unlike RCM, this cannot
be applied plant by plant and needs to be implemented across
all the plants in one location so as to ensure that adequate
history of failures are available. This also has a limitation in
that the method is not strictly as prescribed by the standard
[28].
3.5. Mathematical Models
There have been many attempts to provide one-off models of
RCM that are predominantly mathematical in nature and rely
on probabilistic approaches to the RCM. These models focus
on a specific aspect of the RCM rather than as a
comprehensive implementable strategy. Many of these models
have been based on Markov methods.
Most of these models have remained in the realm of academic
works without industrial adoption. As Van Horenbeek et al.
stated ―currently, there is a big gap between academic models
and application in practice, for this reason, it is very difficult
for industrial companies to adapt these models to their specific
business context‖ [29]. However, these have been presented
here for the sake of ensuring that no alternative remains
hidden.
Endrenyi, Anders and daSilva presented a model that
measured impact of maintenance on reliability [30]. Theil
presented an extension of the Markov-model of this method in
application to RCM. In this model, ―to include exploitation-
time dependent outage rates, the time-behavior is approached
by a step-by-step trend function. In that way, to each wear-out
state a special outage rate is assigned. Thiel concluded that
―because of its complexity the direct implementation of the
proposed model into reliability calculation software for large
electrical networks is not applicable in practice. However, by
neglecting state transitions which are not relevant for systems
with typical component reliability levels, the complex model
can be reduced and thus be implemented into conventional
reliability calculation software without major modifications‖
[31].
Croacker and Kimar proposed an alternative to RCM – Age
Related Replacement based on Hard-life and Soft-life and
proposed a model for suggesting replacement intervals. By
their own admission, the example they showed ―took about 10
hours to produce the output, using a full grid search for just
one part‖ [32].
Adoghe [33] developed a Markovian model to assess the
effect of RCM implementation which strictly is not a new
model but a new method of assessment.
Aurich, Siener and Wagenkneckt proposed the Quality
Oriented Analysis (QOA). The analyzing procedure assesses
the cause-and-effect coherences between the condition states
of machines as well as tools and the product quality within
manufacturing process chains. Thereby, the procedure consists
of a deductive and an inductive analysis phase. During
deductive analysis, the manufacturing process chain and
inherent cause-and-effect coherences are identified and
documented. Structure models of the manufacturing process
chain and more or less established hypotheses about cause-
and-effect coherences are the provided results. Following this,
during the inductive analysis the identified hypotheses are
verified or falsified based on the empirical analysis of data
collected within manufacturing process chains [34].
Sikos proposed a new model that considers the interaction
between maintenance cost and the reliability index [35]. Here
the ‗time-dependent reliability index as proposed by Neves,
Frangopol and Cruz [36] is used.
4. Developing a Baseline for Evaluation
With so many different directions taken by the alternatives, let
alone evaluating them on a common baseline, establishing a
baseline itself will be difficult. In order to develop the
baseline for evaluation, a survey of published literature on
what constitutes the desirable characteristics of an asset

40355
management program is done, and from there a baseline that
can be used for evaluation is developed.
Mokashi, Wang and Vermar state that ―RCM is meant to be a
‗living system‘, i.e. there is a system of feedbacks which
ensures that any newly identified failure modes are
incorporated into the system, as well as the effectiveness of
the recommended maintenance actions is recorded‖ [7].
Smith and Hinchcliffe say that they ―cannot emphasize too
strongly, however, the importance we attach to the notion of
simple‖ [37]. They further state that ―all too often O&M
organizations are heading down the path of very complex
organizational experiments, overnight attempts at cultural
change and unrealistic expectations of dramatic and highly
visible payoffs for relatively small and short-term investment‖
[37].
August asks these questions which pertain to the requirements
of an effective RCM program: ―Craft workers know
maintenance performance, but do they know the right
maintenance? Do they know when to do it? Can they show
why certain maintenance is correct? Can they discover when it
is wrong? Over time, can they incorporate learning? Do they
know when they have reached maintenance limits and what
the equipment can achieve under optimum maintenance? Does
maintenance complement operations?‖ [38]
August further states that ―industrial maintenance is best
performed when planned. The challenge is to choreograph
maintenance steps, aligning them with plant operations to
minimize operating disruptions‖ [38].
Marquez and Gupta quoting Campbell and Reyes-Picknell
[39] suggest a ―formal structure for effective Maintenance
Management. The process starts with the development of a
strategy for each asset. It is fully integrated with the business
plan. At the same time, the HR related aspects required to
produce the needed cultural change are highlighted. Next, the
organization gains control to ensure functionality of each asset
throughout its life cycle. This is done through the
implementation of a CMMS, a maintenance function
measurement system, and planning and scheduling the
maintenance activities. This is accomplished according to
various tactics employed depending on the value that these
assets represent and the risks they entail for the organization‖.
Among these tactics that Campbell and Reyes-Picknell (1995)
includes are ―(a) run to failure, (b) redundancy, (c) scheduled
replacement, (d) scheduled overhauls, (e) ad-hoc maintenance,
(f) PM, (g) age or use based, (h) condition based maintenance,
and (i) redesign‖ [40].
Li, Vaahedi and Choudhury state that RCM should include the
following components at the minimum:
- ―Collecting statistical data such as operations history,
failure records, aging status tests or assessments
- Estimating failure probabilities due to repairable and
end-of-life failures of equipment
- Evaluating impact of individual failures on the
system
- Quantifying the effects of maintenance activities
improving equipment failure frequencies/ repair
timers and whole system reliability
- Applying economic or reliability criteria to
determine the best scheme‖ [41].
Woodhouse listed the requirements of an Asset Management
program.
- ―Lost Opportunity/downtime events are monitored
and costed
- Problem/opportunity identification, investigation and
solving processes all linked together and part of
normal, daily life
- Natural cross-functional team-based working style
- Full-time facilitator(s) to make innovation ideas
happen
- Education: urgently addressing the big gaps and
backlog at management, technical and workforce
levels
- Twin track corporate planning: an ambitious but
realistic goal, on a timescale (typically 3-5 years)
sufficient to achieve fundamental behavioural
change, with clearly-connected ―quick wins‖
priorities used to pay for the sustained commitment
to end goal‖ [42].
Selvik and Aven argue that ―it is crucial to the decision
process that the RCM is adjusted to reflect uncertainties, as
ignoring these may in many applications lead to ‗‗non-
optimal‘‘ maintenance strategies‖ [5].
Spitler describes certain characteristics for any process to be
implemented. These are a) Credibility b) Consistency –
treatment of one equipment or system must parallel that of
another equipment or system c) Structured Format with
standardized yet simple procedures d) Training of key
personnel [43].
Zajicek and Kamenicky found that ―Management
requirements of RCM are a) lower time of analysis →
financial savings, faster results implementation b) unlocking
of specialists for other activities and c) maintenance plans for
all equipment‖ [15].
In all these, there are some common requirements and these
can then be adopted as the basis for evaluation. The
parameters so derived and to be used for the evaluation are
presented below:
1. Structured Format with standardized & Simple
procedures
2. Coverage of All Equipment
3. Monitoring of Failures and Actions
4. Quick Wins
5. Program to be part of day-to-day activities
6. Adjusting to Uncertainties or Trigger Events
7. Retain core feature of RCM Standard
8. HR Linkages – Full time facilitators
9. Strategic Scalability
5. Grouping the Alternatives
The basis for evaluation presupposes that the alternative to
RCM is one that has to be implemented by the industry.
Accordingly, certain parameters become more relevant and
those strategies that conform to the requirements become
preferred ones. Before carrying out the evaluation, the various
strategies described in the preceding sections are further
classified on the basis of broad heads for ease of
understanding. These are highlighted in the table 1.

40356
This table gives a quick breakdown of the various strategies
being evaluated. These will further be evaluated. It may be
noted that for the purpose of brevity only the abbreviations of
the alternatives are used here and forthwith.
Table 1 - Quick Breakdown of Alternatives
Broad
Head
Parameter Alternatives that follow this
ApproachTop-Down
(FMEA Driven)
COFA, Risk-CM, RBCM,
RRCM, RBM, CA, MO, QMO,
RCMP, CIB, RCAM, QOA
Bottom-Up
(Failure Driven)
BCM, PREMO, PMO2000,
IRCMA, IMM, A-RCM
Analysis Mathematical RBCM, RBM, MO, QMO,
RCMP, RCAM
Logical-
Analytical
COFA, Risk-CM, RRCM, CIB,
IRCMA, IMM, A-RCM, QOA
6. Evaluating the alternatives
The various alternative approaches are evaluated under each
of the heads identified in section 4.
6.1. Format and Simplicity
An important requirement of any alternative is the Simplicity
and the formal structure. This makes the system easy to adopt
and easy to manage. The mathematical models with its
reliance on Markov analyses as well as complex algorithms
fail this requirement. For the very same reason, the
Optimization methods that rely on mathematical models as its
basis also fail to meet this criterion.
Among the other alternatives, the COFA, Risk CM, RBCM,
BCM, RRCM and RBM, forming the ‗mix of approaches‘ are
all complex to use, mainly due to their reliance on FMEA as a
starting point (COFA), extensive calculation (Risk CM,
RBCM, RRCM and RBM) and elaborate methodology
(BCM). RCMP, IRCMA, RCAM and IMM from among the
‗Broad strategies‘ are also complex due to dependence on
Mathematical models. CA and CIB are simpler due to its
approach of component-criticality which is intuitive and easy
to adopt for maintenance practitioners. A-RCM in the initial
stages is a simple approach relying on a sequential buildup,
but as the stages progress, the complexity increases to some
extent.
The ‗Simplification‘ alternatives, namely PREMO and
PMO2000 by the very approach – that of simplification of PM
Tasks are the simplest to use. However these have the danger
that the methodology does not follow a formal structure.
Analysis of the approaches for the format and simplicity
indicate that there is a trade-off between the simplicity and
comprehensiveness. A simple approach like PMO does not
attempt to identify Predictive approach or Design changes,
whereas the comprehensive strategies are by no means simple.
6.2. Complete Coverage of All Equipment
A maintenance program can be effective only when it covers
all the assets of the organization. By doing analysis in only a
few equipment, the real goal of reliability improvement,
which is the optimisation of costs and enhanced plant
availability is not achieved.
Among the alternatives highlighted, the mathematical models
do not meet this requirement due to the fact that these, due to
the nature of analysis, can only be applied on a few assets, and
if attempting to apply comprehensively, will lose out on the
time horizon of benefit accrual. Among the other approaches,
CIB, due to its selection of the ‗Most Important System‘,
misses out on the completeness. COFA excludes equipment
that have low consequences and Risk-CM and RBCM those
that have low identified risk. All other approaches cover (if so
desired) all the assets.
6.3. Monitoring of Failures and Actions
For any alternative to be effective there should be a
mechanism to trigger changes in the event of a failure. An
equipment failure indicates that there is a flaw in the
methodology and this needs immediate correction. Models
that are able to correct themselves without waiting for a
review cycle will be more effective that other static models. In
order to achieve this, there has to be a mechanism that tracks
and acts on failures. Typically, top-down approaches will not
be able to meet this requirement fully. Of the top-down
approaches, CA uses failure frequency as an input to analysis.
While other models do not explicitly state this step, it needs to
assumed that those alternatives that are broad based and
relying on a FMEA would have this step built in intuitively.
Hence it can be considered that models like COFA, Risk CM,
RBCM, BCM, RRCM and RBM as well as PREMO and
PMO2000 with the focus on optimisation of PM actions,
would also have this requirement built in. A-RCM, where a
failure triggers and immediate adjustment to the actions meets
this requirement. Mathematical and Maintenance
Optimisation approaches do not meet this requirement.
6.4. Quick Wins
One of the frequently cited limitations of classical RCM is the
inability to provide quick wins. Literature reports cases where
years have passed by without implementable outcomes from
classical RCM analysis. Considering that the success of any
new strategy depends on demonstrated benefits as well as the
commitment of the management, the alternative should build
in quick wins, by which it is understood that there has to be
implementable maintenance tasks from the early stages of
implementation, even if the task is for just one equipment.
Among the alternatives, the typical top-down approach that
starts with an FMEA or an analysis of probable failures
prevents quick wins. The approaches that start off with
implementation starting in parallel with the analysis would
help in achieving quick wins in terms of reliability
improvement. The strategies that are effective in this
parameter are CIB, IRCMA, IMM and A-RCM. It can be
assumed that CA would also provide quick wins for at least a
few classes of equipment.
6.5. Integration with day-to-day activities
One of the needs identified calls for any strategy to be
integrated with day-to-day activities. While RCM by its very
nature is separate from normal maintenance activities, some of
the alternatives are tightly integrated with the existing

40357
practices. These alternatives are easier to implement than
others. Of the alternatives, BCM, PREMO, PMO2000 with its
focus on PM provide a fair degree of integration with the
existing practices. IMM with its continuous analysis of current
situation and A-RCM with its in-built integration with the
existing practices fit this requirement.
6.6. Event/ Uncertainty Handling
A limitation of the classical RCM approach is the time delay
in a trigger event translating into an action. A failure in an
equipment already covered by the RCM will not immediately
see an action on related and similar equipment, unless
specifically built in. This will normally reflect only in the next
cycle of FMEA analysis. The alternative proposed should
ideally have a mechanism to incorporate this into the system
immediately. Due to this all top-down approaches will fail in
this requirement. Of the bottom-up approaches, the ones
which are focused only on PM tasks, will, again not meet this
requirement. Considering this, the alternatives that meet the
requirement are IRCMA with its intelligence based approach
to analysis, IMM with its analysis of current situation as the
basis for analysis, CA with its monitoring of failure and
downtime length and A-RCM which used the failure as the
trigger for deciding RCM tasks.
6.7. Correlation to Classical RCM/ Standard
There exists the SAE JA1011/ 1012 standards that have
codified how RCM should be implemented. While this
standard nearly mandates the use of FMEA in order to
establish the failure modes that are ―reasonably likely‖ to
occur, alternatives that do not follow the FMEA approach
may not comply to the standard. Of the alternatives, nearly all
of the top-down approached will cnform with the
requirements of the standard. The alternatives BCM, PREMO,
PMO2000, IRCMA, IMM, in addition to MO, QMO and none
of the mathematical models meet this requirement. CA and A-
RCM meet this requirement partially, if a more liberal
interpretation of reasonable likelihood is applied.
6.8. Human Resource - Facilitators
One of the requirements that were identified was the need for
organizational support in the form of a full time facilitator.
The presence of this facilitator ensures that the RCM
implementation stays on track as well as ensures that the
system remains under control. The scan of the literature
indicated that in all the alternatives analysed here, there is no
mention about the requirement or presence of full time
facilitators. However, it can be surmised, taking into account
the fact that these are all additions to existing maintenance
practices, and there will be the need for a full time facilitator
who ‗drives‘ the system forward.
6.9. Strategic Scalability
As with any strategy, the alternatives to RCM (and indeed
RCM itself) needs to be scalable in that it should allow
organizations the option of slowly ensuring complete
coverage. This was treated as one of the biggest drawbacks of
the classical RCM approach and the literature cited reports the
pitfalls of the need for organization wide implementation
upfront (eg. August, Ramey and Vasudevan [44] report on an
implementation in a nuclear industry). Hence the alternative
needs to be scalable, in that it can be adopted system by
system and slowly cover all systems.
Among the alternatives, the following systems necessarily
needs to be implemented organization-wide, and hence cannot
be deemed as meeting these criteria. QMO and RCMP with its
minimisation of Total Cost of maintenance and CIB with the
need to identify the most important system from all do not
meet the requirement of scalability. MO methods may also not
fit into scalability, since the primary concern is to ensure
optimality in costs, tasks and resources. A-RCM is scalable,
not on application but on intensity and depth.
6.10. Summary of Evaluation
The sections above evaluated the alternatives against each of
the nine parameters. While each of the alternatives have
something specific to offer as an advantage, the evaluation
showed that, of all the alternatives none met all the parameters
completely. Of the alternatives CA, A-RCM and IMM
complied with the majority of requirements, while
Mathematical Models and both the Optimisation models (MO,
QMO) complied with the least number of parameters.
The summary of the evaluation is presented in the table 2
below, so as to provide a ready reference to those wishing to
choose one of these alternatives for implementation. The
tabulation is done as O – Meeting Fully, X – Not Meeting, P-
Partially Meeting and ? – Possibly meeting.
Table 2 - Summary of Evaluation
Baseline
→
1.
Format
&
Simplicity
2.
Complete
Coverage
3.
Monitoring
failures
4.
Quick
Wins
5.
Integration
to
day-to-
day
activities
6.
Uncertainty
Handling
7.
Correlation
to
Standard
8.
Full
time
Facilitators
9.
Strategic
Scalability
Alternatives↓
RCM* X O O X X O O O O
COFA X X O X X X O O O
Risk-CM X X O X X X O O O
RBCM X O O X X X O O O
BCM X O O X P X X O O
RRCM X O O X X X O O O
RBM X O O X X X O O O
PREMO O O P X P X X P O
PMO2000 O O P X P X X P O
CA O X O X X O O O O
MO X O X X X X X O X
QMO X O X X X X X O X
RCMP X O X X X X O O X
CIB O X X O X X O O X
IRCMA X O X O X O X O O
RCAM X O X X X X O O O
IMM X O X O O O X O O
A-RCM P O O O O O X/? P O
MM X X X O X X X O O
* Classical RCM was not analysed in detail, but is presented
here so that a quick comparison can be made

40358
7. Conclusion and Future Work
This paper attempted to provide an evaluation of the various
alternatives proposed for classical RCM. A baseline for
evaluation was generated based on the generic requirements
of maintenance strategy and the alternatives were subject to a
qualitative evaluation based on the understanding of the
alternatives. Certain alternatives were presented in great detail
by the respective authors which resulted in a better
understanding and consequently a more accurate evaluation.
A few alternatives identified here could not be evaluated as
the details available did not provide enough information to
carry out a proper evaluation.
While an attempt has been made to evaluate a large number of
alternatives which were identified using varied search terms,
this paper does not claim to cover all the published
alternatives of RCM. However, the evaluation methodology
presented can easily be applied to an alternative being
developed or one that has not been evaluated in this paper.
RCM is a very complex, yet effective tool for maintenance
management. The realisation of the effectiveness of the
approach and the understanding of the complexity of the
implementation has resulted in the development of varied
alternatives. These alternatives have their own advantages and
limitations. As seen from the evaluation, no alternative could
meet all parameters fully, though a few come close. This is
also an indication that the maintenance management strategies
need greater attention so that an effective strategy can be
developed and presented to the industry.
8. References
[1] S. A. E. (1999). JA1011, Evaluation criteria for
Reliability-Centered maintenance (RCM) processes.
Society for Automotive Engineers..
[2] Nowlan, F. S., & Heap, H. F. (1978). Reliability-
Centered maintenance. United Air Lines Inc San
Francisco CA.
[3] Moubray, J. (1997). Reliability Centered
maintenance. Industrial Press.
[4] Bloom, N. (2005). Reliability-Centered Maintenance
(RCM). McGraw-Hill.
[5] Selvik, J. T., & Aven, T. (2011). A framework for
reliability and risk centered maintenance. Reliability
Engineering & System Safety, 96(2), 324-331.
[6] Pride, A. (2008). Reliability-centered maintenance
(RCM). WBDG Electronic Resource [http://www.
wbdg. org/resources/rcm. php].
[7] Mokashi, A. J., Wang, J., & Vermar, A. K. (2002). A
study of reliability-centred maintenance in maritime
operations. Marine Policy, 26(5), 325-335.
[8] Jones, R.B., (1995), Risk-Based Maintenance, Gulf
Professional Publishing (Elsevier), Oxford
[9] Kelly, A., (1997), Maintenance Organizations &
Systems: Business-Centred Maintenance,
Butterworth-Heinemann, Oxford.
[10] Khan, F. I., & Haddara, M. R. (2004). Risk-based
maintenance of ethylene oxide production facilities.
Journal of Hazardous Materials, 108(3), 147-159.
[11] Prabata, S., & Wiyana, A. F. (2012). Application of
combined Reliability Centered Maintenance and Risk
Based Inspection Method to Improve the
Effectiveness of Maintenance Management of
Compressor ZR 5 Intercooler. The Indonesian
Journal of Business Administration, 1(2).
[12] Abid, M., Ayub, S., Wali, H., & Tariq, M. N.(2014).
Reliability Centered Maintenance Plan for the Utility
Section of a Fertilizer Industry: A Case Study.
International Journal of Science and Advanced
Technology Volume 4, No 3, 9-16.
[13] Endrenyi, J., Aboresheid, S., Allan, R. N., Anders, G.
J., Asgarpoor, S., Billinton, R.,... & Singh, C. (2001).
The present status of maintenance strategies and the
impact of maintenance on reliability. Power Systems,
IEEE Transactions on, 16(4), 638-646.
[14] Bevilacqua, M., & Braglia, M. (2000). The analytic
hierarchy process applied to maintenance strategy
selection. Reliability Engineering & System
Safety,70(1), 71-83.
[15] Zajicek, J., & Kamenicky, J. (2007). Effectiveness
optimization of RCM process. In Proceedings and
Monographs in Engineering, Water and Earth
Sciences, European Safety and Reliability
Conference (ESREL 2007) (pp. 1887-1891).
[16] Dekker, R. (1996). Applications of maintenance
optimization models: a review and analysis.
Reliability Engineering & System Safety, 51(3), 229-
240.
[17] Turner, S. (2002). PMO Optimisation: A Tool for
Improving Operations and Maintenance in the 21st
Century. International Conference of Maintenance
Professionals. Melbourne.
[18] Berger, D. (2005). Maintenance Optimization and
Your Plant. Online article from
http://www.plantservices.com/articles/2005/492.html
?page=print
[19] Idhammer, C. (2008). Preventive Maintenance
Optimization. http://www.idcon.com/resource-
library/articles/preventive-maintenance/541-
preventive-maintenance-optimization.html
[20] Dotzlaf, R. E. (2009). Modernizing a Preventive
Maintenance Strategy for Facility and Infrastructure
Maintenance (No. AFIT/GEM/ENV/09-M03). Air
Force Inst of Tech Wright-Patterson AFB OH School
Of Engineering And Management.
[21] Vatn, J., Hokstad, P., & Bodsberg, L. (1996). An
overall model for maintenance optimization.
Reliability Engineering & System Safety, 51(3), 241-
257.
[22] Besnard, F., Fischer, K., & Bertling, L. (2010,
October). Reliability-Centred Asset Maintenance—A
step towards enhanced reliability, availability, and
profitability of wind power plants. In Innovative
Smart Grid Technologies Conference Europe (ISGT
Europe), 2010 IEEE PES (pp. 1-8). IEEE.
[23] Bae, C., Koo, T., Son, Y., Park, K., Jung, J., Han, S.,
& Suh, M. (2009). A study on reliability centered
maintenance planning of a standard electric motor
unit subsystem using computational techniques.

40359
Journal of mechanical science and technology, 23(4),
1157-1168.
[24] Waeyenbergh, G., & Pintelon, L. (2002). A
framework for maintenance concept development.
International journal of production economics, 77(3),
299-313.
[25] Cheng, Z., Jia, X., Gao, P., Wu, S., & Wang, J.
(2008). A framework for intelligent reliability
centered maintenance analysis. Reliability
Engineering & System Safety, 93(6), 806-814.
[26] Barberá, L., Crespo, A., Viveros, P., & Stegmaier, R.
(2012). Advanced model for maintenance
management in a continuous improvement cycle:
integration into the business strategy. International
Journal of System Assurance Engineering and
Management, 3(1), 47-63.
[27] Prabhakar, Deepak P & Raj, Jagathy (2013).
Accelerated Reliability Centered Maintenance: A
New Maintenance Strategy. Indian Journal of
Applied Research, Vol.III, Issue.XII, 261-265
[28] Prabhakar, Deepak P & Raj, Jagathy (2013). CBM,
TPM, RCM and A-RCM – A Qualitative
Comparison of Maintenance Strategies. International
Journal of Management & Business Studies, Vol.4,
Issue.3, 49-56
[29] Van Horenbeek, A., Pintelon, L., & Muchiri, P.
(2010). Maintenance optimization models and
criteria. International Journal of System Assurance
Engineering and Management, 1(3), 189-200.
[30] Endrenyi J, Anders, G.J and A.M. Leite da Silva
(1998), "Probabilistic evaluation of the effect of
maintenance on reliability – an application", IEEE
Transactions on Power Systems, Vol. 13, No.2, pp.
576 -583.
[31] Theil, G. (2005). Markov models for Reliability-
Centered Maintenance Planning. na.
[32] Crocker, J., & Kumar, U. D. (2000). Age-related
maintenance versus reliability centred maintenance: a
case study on aero-engines. Reliability Engineering
& System Safety, 67(2), 113-118.
[33] Adoghe, A. U. (2010). Reliability Centered
Maintenance (RCM) for Asset Management in
Electric Power Distribution System (Doctoral
dissertation, Covenant University).
[34] Aurich, J. C., Siener, M., & Wagenknecht, C. (2006,
May). Quality oriented productive Maintenance
within the life cycle of a manufacturing system. In
13th CIRP international conference on life cycle
engineering (p. 31).
[35] Sikos, L. (2010). New trends in reliability,
availability and maintenance optimisation of waste
thermal treatment plants (Doctoral dissertation, PhD
Thesis, Information Science & Technology PhD
School, University of Pannonia, Veszprém,
Hungary).
[36] Neves, L. C., Frangopol, D. M., & Cruz, P. S.
(2004). Cost of life extension of deteriorating
structures under reliability-based maintenance.
Computers & structures, 82(13), 1077-1089.
[37] Smith, A. M., & Hinchcliffe, G. R. (2003). RCM--
Gateway to World Class Maintenance. Butterworth-
Heinemann.
[38] August, J. (2004). RCM guidebook: Building a
reliable plant maintenance program. PennWell
Books.
[39] Campbell, J. D., & Reyes-Picknell, J. (1995).
Strategies for excellence in Maintenance
management. Productivity Prsee Portland, Oregon
pages-185.
[40] Marquez Crespo, A., & Gupta, J. N. (2006).
Contemporary maintenance management: process,
framework and supporting pillars. Omega, 34(3),
313-326.
[41] Li, W., Vaahedi, E., & Choudhury, P. (2006). Power
system equipment aging. Power and Energy
Magazine, IEEE, 4(3), 52-58.
[42] Woodhouse, J. (1997). What is Asset Management?.
Maintenance and Asset Management, 12, 26-28.
[43] Spitler, W. W. (1990). A study of Reliability
Centered aircraft maintenance and opportunities for
application by the United States Coast Guard
(Doctoral dissertation, Massachusetts Institute of
Technology).
[44] August, J. K., Ramey, B., & Vasudevan, K. (2005,
January). Baselining Strategies to Improve PM
Implementation. In ASME 2005 Power Conference
(pp. 287-297). American Society of Mechanical
Engineers.
View publication stats
View publication stats

An_Evaluation_of_Alternative_Approaches_to_Reliabi.pdf

Recommended

Recommended

More Related Content

Similar to An_Evaluation_of_Alternative_Approaches_to_Reliabi.pdf

Similar to An_Evaluation_of_Alternative_Approaches_to_Reliabi.pdf (20)

Recently uploaded

Recently uploaded (20)

An_Evaluation_of_Alternative_Approaches_to_Reliabi.pdf