Journal of Quality in Maintenance Engineering

Emerald Article: Maintenance strategy selection using AHP and ANP algorithms:
a case study
Selim Zaim, Ali Turkyilmaz, Mehmet F. Acar, Umar Al-Turki, Omer F. Demirel

Article information:
To cite this document: Selim Zaim, Ali Turkyilmaz, Mehmet F. Acar, Umar Al-Turki, Omer F. Demirel, (2012),"Maintenance strategy
selection using AHP and ANP algorithms: a case study", Journal of Quality in Maintenance Engineering, Vol. 18 Iss: 1 pp. 16 - 29
Permanent link to this document:
http://dx.doi.org/10.1108/13552511211226166
Downloaded on: 28-03-2012
References: This document contains references to 29 other documents
To copy this document: permissions@emeraldinsight.com

Access to this document was granted through an Emerald subscription provided by Emerald Author Access
For Authors:
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service.
Information about how to choose which publication to write for and submission guidelines are available for all. Additional help for
authors is available for Emerald subscribers. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
With over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in
business, society, public policy and education. In total, Emerald publishes over 275 journals and more than 130 book series, as well as an
extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is a partner
of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.
*Related content and download information correct at time of download.
 The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1355-2511.htm

JQME Maintenance strategy selection using

18,1
AHP and ANP algorithms:
a case study
16 Selim Zaim
Department of Mechanical Engineering, Marmara University, Istanbul, Turkey
Ali Turkyılmaz
Department of Industrial Engineering, Fatih University, Istanbul, Turkey
Mehmet F. Acar
Department of Management, Fatih University, Istanbul, Turkey
Umar Al-Turki
Systems Engineering Department,
King Fahd University of Petroleum & Minerals, Dhahran, Saudi Arabia, and
Omer F. Demirel
Department of Industrial Engineering, Fatih University, Istanbul, Turkey

Abstract
Purpose – The purpose of this paper is to demonstrate the use of two general purpose decision-
making techniques in selecting the most appropriate maintenance strategy for organizations with
critical production requirements.
Design/methodology/approach – The Analytical Hierarchical Process (AHP) and the Analytical
Network Process (ANP) are used for the selection of the most appropriate maintenance strategy in a
local newspaper printing facility in Turkey.
Findings – The two methods were shown to be effective in choosing a strategy for maintaining the
printing machines. The two methods resulted in almost the same results. Both methods take into
account the specific requirements of the organization through its own available expertise.
Practical implications – The techniques demonstrated in this paper can be used by all types of
organizations for selecting and adopting maintenance strategies that have higher impact on
maintenance performance and hence overall business productivity. The two methods are explained in
a step-by-step approach for easier adaptation by practitioners in all types of organizations.
Originality/value – The value of the paper is in applying AHP and ANP decision-making
methodologies in maintenance strategy selection. These two methods are not very common in the
area of maintenance, and hence add to the pool of techniques utilized in selecting maintenance
strategies.
Keywords Maintenance planning, AHP, ANP, Maintenance strategy, Strategy selection,
Maintenance, Turkey
Paper type Case study
Journal of Quality in Maintenance
Engineering Vol. 18 No. 1, 2012 pp.
16-29 q Emerald Group Publishing
Limited 1355-2511
DOI 10.1108/13552511211226166
The authors acknowledge the support of both Fatih University and King Fahd University. They
also acknowledge the anonymous referees for their constructive comments. They thank Sedat
 Maintenance
strategy
selection
Kızıltunc for his is becoming increasingly critical with the increasing
collaboration. They also competition in the business environment. The
wish to acknowledge the
support provided by
competition is leading to more focus on cost reduction
upper management of in operations and maintenance. Cost reduction may
Zaman newspaper. immediately be reflected on pricing and hence, gaining
1 edge over competitors. Maintenance cost constitutes a
major portion of total operations cost and hence is
.
central to most cost
I reduction programs. Such programs should be done with care so that other 17
requirements such as quality are not sacrificed.
n
t Maintenance costs can reach to 15-70 percent of
production costs according to different sectors
r
(Bevilacqua and Braglia, 2000). Moreover,
o
maintenance directly or indirectly influences product
d quality, safety and reliability. Nowadays, maintenance
u is considered as “profit contributor” and “partner” for
c worldclass competitiveness (Waeyenberg and Pintelon,
t 2002). Rausand (1998) identified the four probable
i consequences of failure:
o (1) safety of personnel;
n
(2) environmental impact;
T
h (3) production availability; and
e (4) cost of material loss.

c Maintenance is one of the most crucial issues in today’s

o competitive manufacturing environment. Machine
s failure may cause various business related problems
t such as; missing delivery dates, loss of image and direct
and indirect loss of profit and opportunity loss. As such,
o maintenance should be carefully dealt with in terms of
f planning, investment, and control. In terms of planning,
appropriate maintenance strategies should be selected
m that are in line with company’s global and operational
a objectives. However, maintenance strategies change
i rapidly with new options and practices. In fact any
n change in operations requires some adjustment or major
t change in the adopted maintenance strategy to be
e compatible with the new requirements. The selection
n process itself is becoming crucial for achieving highest
a performance. Such decisions that highly impact
n technology are usually dealt with in technically founded
c manner.
e The motivation of this work is the existing need for
some technical methodologies for optimum selection of
JQME
18,1
t esearch, two of the commonly methods for decision
h making, namely the Analytical Network Process (ANP)
e and the Analytical Hierarchy Process (AHP), are used
for the selection of the best maintenance policy. The
m two methods are simple but powerful in making
o decisions at different business and functional levels
s under high complex and uncertain conditions.
t This paper is organized as follows. First, literature
review is written about maintenance and maintenance
f selection, then AHP and ANP are overviewed and
i proposed model is introduced. Last, case study is
t introduced in this paper. The method is demonstrated
through a case study from local industry in Turkey.
m Results are discussed and benefits are identified.
a
i 2. Literature review
n
Maintenance is classified into two main categories:
t
corrective and preventive (Li et al., 2006; Waeyenberg
e
and Pintelon, 2004). Corrective maintenance is
n
performed after system failure and preventive
a
maintenance is performed before its failure (Wang,
n
2002). Corrective maintenance, also called breakdown
c
maintenance, is the oldest strategy in the industry
e
(Waeyenberg and Pintelon, 2002; Mechefske and
Wang, 2003; Wang et al., 2007). For large profit
s
t margin organizations, this policy can be seen as feasible
r strategy (Sharma et al., 2005).
a Preventive maintenance, in practice, has two forms;
t periodic and predictive. In
e 18 periodic maintenance, as the name suggests, maintenance is performed
g periodically to prevent sudden failure (Wang et al.,
i 2007). This strategy is also called “time-based maintenance” and is used by
e many firms in the industry following manufacturer’s recommendations, which
s sometimes results in unnecessary maintenance activities.
. In predictive maintenance, maintenance decisions
are made based on information collected from special
I measurement instruments like sensor systems,
n monitoring techniques, vibration monitoring,
lubrication analysis and ultrasonic testing (Wang et al.,
t 2007). This strategy is also known as condition-based
h maintenance.
i In addition to these, opportunistic maintenance is
s used by some large-scale industries such as petroleum
and petrochemical industries. Bevilacqua and Braglia
r (2000) defined the opportunistic maintenance as
 Maintenance
strategy
selection
“ perform all relevant maintenance interventions at the
m same time.”
ai Studies on maintenance systems in practice show
nt that some managers are unaware of the different types
e of maintenance policies (Shorrocks, 2000; Shorrocks
n and Labib, 2000) and selection methods. Luce (1999),
a Okumura and Okino (2003) presented the maintenance
n selection method based on production loss and
c maintenance cost. Azadivar and Shu (1999) showed the
e effective methods of selecting appropriate (optimum)
c maintenance strategies for just in time production
a systems. Bevilacqua and Braglia (2000) used
n Analytical Hierarchy Process (AHP) for maintenance
le selection in an oil refinery and they described some
a features in the selection of maintenance strategy, such
d as: economic factors, applicability, costs and safety. Al-
to Najjar and Alsyouf (2003), Sharma et al. (2005) used
th fuzzy inference theory and fuzzy multiple criteria
e decision making methodology. Moreover, Mechefske
w and Wang (2003) showed a new method for selecting
h the optimum maintenance strategy and condition
ol monitoring technique. Almeida and Bohoris (1995)
e developed a new method using decision-making theory
pl especially the multi-attribute utility theory.
a Triantaphyllou et al. (1997) presented AHP model with
nt four maintenance criteria:
b (1) cost;
ei
n (2) reparability;
g (3) reliability; and
s
(4) availability.
h
ut In addition to these, Bertolini and Bevilacqua (2006)
d proposed a combined goal programming and AHP for
o maintenance selection. Wang et al. (2007) developed a
w fuzzy AHP model for selection of optimum
n maintenance strategy.
at Labib et al. (1998) developed a model of
s maintenance decision making, which includes AHP. In
et the first stage, criteria are identified and then in the
ti second stage AHP is applied. Last, machines are ranked
m according to their importance. Arunraj and Maiti
e (2010) used AHP and goal programming for
s maintenance policy selection according to risk of
to failure and cost of maintenance in a chemical factory.
JQME
18,1
T ce is preferred policy over periodic maintenance.
h Similarly, if cost is chosen as a criterion, corrective
e maintenance is preferred. Nevertheless, if both risk and
y cost are considered, AHP-GP results show that
c predictive maintenance and corrective maintenance are
o best for high-risk equipment and low-risk equipment,
n respectively. Labib (2004) also developed a model for
cl maintenance
u policy selection using a computerized maintenance management system. In this study,
d 19
e
fuzzy logic and AHP are used. HajShirmohammadi and Wedley (2004) used an
d AHP model for maintenance management for
th
centralization and decentralization. Centralized system means that all
at maintenance systems are managed from a centrally administered location.
if
However, decentralized system implies that each production area manages its
ri own maintenance systems.
s
Shyjith et al. (2008) developed a model using AHP
k
and TOPSIS for maintenance selection in textile
is
industry and then Ilangkumaran and Kumanan (2009)
c
integrated fuzzy AHP and TOPSIS algorithm to select
h
the maintenance policy for textile industry.
o
It is clear from the literature that AHP has proven
s
success in maintenance strategy selection, as it did for
e
many other decision-making problems. As such, it was
n
selected to be the major tool in this paper.
a
s
a 3. Theoretical background
cr 3.1 The analytic hierarchy process (AHP) method
it The analytic hierarchy process (AHP) methodology,
er which was developed by Saaty (1980), is a powerful
io tool in solving complex decision problems. The AHP
n, helps the analysts organize the critical aspects of a
p problem into a hierarchical structure similar to a family
re tree. By reducing complex decisions to a series of
di simple comparisons and rankings, then synthesizing the
ct results, the AHP not only helps the analysts arrive at the
iv best decision, but also provides a clear rationale for the
e choices made (Chin et al., 1999).
m In the AHP approach, the decision problem is
ai structured hierarchically at different levels with each
nt level consisting of a finite number of decision elements.
e The upper level of the hierarchy represents the overall
n goal, while the lower level consists of all possible
a alternatives. One or more intermediate level embody
n the decision criteria and sub-criteria (Partovi, 1994).
 Maintenance
strategy
selection
3 The ANP method is an improved version of AHP
. method and it is more accurate with many complicated
2 models in which many criteria feedback and
interrelations among criteria are used.
The ANP method evaluates all the relationships
T systematically by adding all interactions,
h interdependences, and feedbacks in decision-making
e systems. The powerful side of our model is to represent
the decision-making problem that involves many
a complicated relationships easily. This technique does
n not only enable the pair-wise comparisons of the sub-
a criteria under main criteria, but also enables us to
compare independently all the interacting sub-criteria.
l
Decision making problems that occur in firms cannot
y be explained by only hierarchical structures. The
t criteria and alternatives in a problem can have
i interactions. At these circumstance, complicated
c analyzes can be necessary to find out the weights of all
a components. ANP technique is used for such as that
l kind of problems and it is based on pairwise
comparisons as it is in AHP. For pairwise comparisons
the 1-9 scale of Saaty (1980) is used in Table I. In ANP
n model all the components and relationships are defined
e and the relationships are determined as two-way
t interactions. In the model the network structure is used
w and all the relationships in a cluster (that is relationships
o among sub-criteria in a cluster) and 20 relationships
r between sub-criteria under different clusters are
k considered. Because of such
relationships, the ANP method is useful for getting
more accurate and effective results is a complex and
p
crucial decision making problems.
r In ANP method there are three matrix analyses;
o super matrix, weighted super matrix and limit matrix.
c The super matrix provides relative importance of all
e components and weighted super matrix finds out the
s value that is obtained by the super matrix values and
s the value of each cluster. In the limit matrix, the
constant values of each value are determined by taking
the necessary limit of the weighted super matrix. The
(
results of the decision-making problem, is gained from
A the limit matrix scores. It is important to value the
N criteria and alternatives by the experts in order to get
P more consistent and reliable results.
)
JQME
18,1
4. ategy is based on a hierarchical model composed of a
C set of criterion and sub-criterion as developed by Saaty.
a Both AHP and ANP methods are demonstrated by the
s following case study from the newspaper printing
e industry. The use of the two methods is reported along
st with their resulting solutions.
One of the most selling newspapers in Turkey,
u
“ZAMAN”, is the subject of the case study in this
d
research. It publishes national and international news in
y the fields of politics, business, economics, arts,
T cultures, sports, etc. It is published seven days a week
h with approximately 30 pages in addition to publishing
e “TODAY’SZAMAN”, the most circulated English
m newspaper in Turkey, and special supplements in
et weekends and special occasions. It won different
h awards in several design competitions.
o To meet its publication daily schedule, machines
d and equipments in its printing house must be kept
p continuously ready for production which puts high
r pressure on operations and maintenance. Thus,
o maintenance is highly crucial for this firm and therefore
p selected to be the focus of this paper. The objective of
o this study is to select the best maintenance strategy that
s meets the operations objectives. Three alternative
e maintenance strategies are considered, these are;
d corrective, periodic (time-based) and predictive
f (condition-based) maintenance policies. Opportunistic
o maintenance is not considered because long time shut
r down of equipments and machines is not expected.
s
el Description Nume
value(
e
ct Equally importance
i Moderately importance
n
g Strongly importance
m
aiTable I. Very strongly importance
Saaty’s 1-9 scale for AHP Extremely importance
n preference Intermediate values
te
n Four selection criteria are considered. These are:
a (1) added value;
n (2) cost;
c
e (3) safety; and (4) implementation.
st 21
r
 Maintenance
strategy
selection
Moreover, Step 1: Form a focus group composed of key managers
different sub- and engineers
criteria are The purpose is to determine and examine current
added to the problems and their impact at the business level of the
model. company. In this case, a project team is established.
According to The project team is composed of five managers from
the proposed production planning and control and maintenance in
addition to some experts from several universities.
model,
problems, Step 2: Evaluate the problems
criteria and Facilitate a focus group meeting to identify issues and
alternatives are problems related to maintenance and their possible
found and these causes mapped into categories and subcategories. The
are described in group also constructs maintenance method selection
the steps below. criteria and sub-criteria. In this case, the team identified
several issues related to maintenance in “ZAMAN”
printing house. Some of the issues found to be crucially
related to maintenance strategy. One of the most
important problems is the firm’s image. If machines
breakdown and production stops, newspaper may not
be issued and this situation negatively affects the image
of the firm. Another problem is found to be cost. In
case of shutdown, the firm may need to outsource the
printing of the newspaper and this causes an extra cost
for the firm.

Step 3: Determine the alternative strategies

Some maintenance strategies might not be suitable for a
certain organization. That strategy can be eliminated by
the focus group with more attention and analysis may
be conducted for feasible strategies. Throughout the
discussions held with the formed group members, three
possible alternative maintenance strategies are
identified, these are; corrective, periodic and predictive
maintenance policies. Opportunistic maintenance is not
considered because long time shut down of equipments
and machines is not expected.

Step 4: Construct a hierarchical model

Using criteria, sub-criteria and alternatives, a
hierarchical model is constructed to apply AHP and
ANP algorithms. Then relationship among criteria and
sub-criteria are determined and reflected in the
hierarchical model. Maintenance strategy selection
criteria were determined based on the review of prior
JQME
18,1
l rtaken with 22 managers from relevant departments
i including purchasing, manufacturing, quality assurance
t and maintenance. Figure 1 shows the hierarchical
e structure of the
r maintenance strategy
a selection problem,
t which includes four
u levels. The top level
r of the hierarchy
e represents the

a
n
d

s
e
m
i ultimate goal of the problem, while the second level of
- the hierarchy consists of four main maintenance policy
s selection criteria, which are namely: added value, cost,
t safety, and implementation.
r
u
22
c
t
u
r
e
d Figure 1.
AHP and ANP scores
i
n
t These criteria are decomposed into various sub-criteria
e that may affect the manager’s decision for a particular
r maintenance policy. Finally, the bottom level of the
v hierarchy represents the alternative maintenance
i policies. Each selection criterion in the tree diagram is
e briefly described below.
w The sub-criteria for each main
s criteria are identified as follows:
(1) Value adding is viewed in four
u possible dimensions (sub-
n criteria) as follows:
d
e
.
On time delivery: During the production process,
some machines may fail causing delays in order
 Maintenance
strategy
selection
in product quality showing as damaged paper
or unreadable text.
. .
Image: The image of the firm is largely affected
by production and maintenance performance.
Late deliveries, low quality printing, shortage
in quantities are some examples causing
image damage.
(2) The cost criterion includes the
following:
.
Hardware: To apply predictive maintenance, the firm
may need to acquire some new machines or
equipments.
.
Software: Different software may be required to
evaluate information which is obtained from
equipments used for predictive maintenance.
.
Training: Technicians or managers may be
required to go through special training for
effective use of equipments and software that
are used in predictive maintenance.
.
Inventory of spare parts: Maintenance
strategies, especially corrective maintenance,
some spare parts should be available in
inventory. The cost of holding spare parts
adds to the overall maintenance cost.
.
Cost of advising and consulting: For corrective
and periodic maintenance strategies, the firm
may need some special maintenance experts
to plan and control maintenance operations.
These costs are mostly necessary regardless of the type
of maintenance strategy adopted whether corrective,
periodic or predictive. However, the costing elements
vary in amount between strategies:
(3) The safety criterion consists of the following:

.
23
.
Internal environment: Safety policies and
procedures maintains healthy working
environment. Interruptions in operations due
to failure may form a source of hazard to
people and the whole internal environment.
.
External environment: Safety outside the
factory is another crucial element, especially
JQME
18,1
f e or chemical spills in the printing house may
o cause unrecoverable damage to the
r surrounding environment.
.
Personnel: Last, some breakdowns and/or
n maintenance activities may directly or
u indirectly harm workers. Therefore, it is
c essential to seek their opinion about the
l possible maintenance practices.
e
(4) The implementation criterion includes the
a
following:
r
.
Technology: Technology is an important for
o predictive maintenance, because there are no
r special equipments for some machines to
apply condition-based maintenance.
c .
Desire of workers: Some of workers may not
h want to predictive maintenance, because
e workers do some extra duties in condition-
m based maintenance.
i .
Desire of top management: Sometimes top
c managers do not want to apply predictive
a maintenance, because its setup cost which is
l sourced from buying of hardware and
software is high.
p
l
.
Decision of service company: There are lots
a companies which supply maintenance service
n as business for other companies, so this is a
t criterion for firms.
s Based on the identified criteria and sub-criteria, a
. hierarchical model is constructed and relations are
determined for our case study. The model is shown in
I Figure 2.
n
Step 5: Pair wise comparison among criteria and sub-
c criteria
a
Pair wise comparisons are done among related criteria
s
and sub-criteria following the scale suggested by Saaty.
e
A special questionnaire form is used to complete the
pair-wise comparison matrix. In this comparison,
o
criteria, sub criterias are used for comparing alternative
f
maintenance strategies by experts in the field.
f
i Step 6: Applying AHP and ANP algorithms
r These algorithms give weight to each alternative based
on which the best strategy is chosen.
 Maintenance
strategy
selection
I ria and the scores of the alternatives, which are called
n local priorities, are considered as decision elements in
the second step of the decision process. The decision-
t maker is required to provide his preferences by
h
e

A
H
P
24
a
p
p
r
o
a
c
h
,

t
h
e

w Figure 2.
e The hierarchical model for
i maintenance strategy
selection criteria
g
h
t
s pairwise comparisons, with respect to the weights and
scores. The values of the weights vi and scores rij are
o elicited from these comparisons and represented in a
f decision table. The last step of the AHP aggregates all
local priorities from the decision table by a weighted
t sum of the type:
h X
e Rj ¼ vi · rij
i

c
r The global priorities Rj thus obtained are finally used for
i ranking of the alternatives and selection of the best one.
t In the ANP approach, two matrices are calculated;
e the weighted super matrix and the limit matrix. The
weighted super matrix permits a resolution of the
JQME
18,1
in a partitioned matrix where each sub-matrix is
te composed of a set of relationships between and within
r the levels, as represented by the model. The entries of
d the super matrix are imported from the pair-wise
e comparison matrices of interdependencies. Since there
p are 20 such pairwise comparisons matrices, one for
e each interdependent criterion, the super matrix contains
n 20 non-zero columns. The weighted super matrix is
d obtained by multiplying all the elements in a
e component of the unweighted super matrix by the
n corresponding cluster weight. In other words, the
ci values in the cluster matrix are used to weight the
e unweighted super matrix by multiplying the value in
s the cell of the cluster matrix times the value in each cell
th in the component of the unweighted super matrix to
at produce the weighted super matrix. The resulting
e weighted super matrix is shown in Table II. The limit
xi super matrix is obtained by raising the weighted super
st matrix to the power 2k þ 1 where k is an arbitrarily
a large number, allows convergence of the
m interdependent relationships. When the column of
o numbers is the same for every
n column, the limit matrix has been reached and the matrix multiplication process
g is 25 halted. The limit super matrix for the Model is shown in Table III.
th
e
c
o Step 7: Prioritization
m It simply means listing alternatives in descending order
p of their weights according to both AHP and ANP
o algorithms.
n The AHP algorithm resulted in the following ranking
e (best to worst) of maintenance strategies; predictive,
nt periodic and corrective maintenance respectively and
s using the ANP algorithm; the resulting strategy ranking
o (best to worst) is; predictive, periodic and corrective
f maintenance, respectively as shown in Figure 1.
a
s Step 8: Compare results and make the decision
y The two solutions, AHP and ANP, are compared and
st evaluated by experts to make the best decision.
e The analysis clearly shows that predictive
m maintenance is to be the best strategy by both AHP and
. ANP methods. However, in the real situation,
It “ZAMAN” is using periodic maintenance for
is maintaining its printing house. In fact, predictive
 Maintenance
strategy
selection
m unnecessary expenditure for “ZAMAN”. This is not
a recognized by the firm since the maintenance
i effectiveness is quite high on the expense of efficiency
n in resource utilization. Furthermore, technicians and
t experts are occasionally interfering with the
e maintenance operations based on their intuitions before
n the time of periodic maintenance. This can be seen as
a predictive maintenance, hence, in fact, both predictive
n and periodic maintenance are used in ad-hoc basis.
c Currently there are plans underway in “ZAMAN” to
e adopt an ERP system for planning and controlling
maintenance operations and at the same time be used to
i estimate and report the cost of maintenance.
s
5. Conclusion
s
In this research, some criteria are determined about
h
maintenance selection and according to these criteria,
o
AHP and ANP models are constituted. Three
w
maintenance policies are considered; these are
n
corrective, periodic (time-based) and predictive
(condition-based) maintenance. Moreover, with the
b
help of experts and engineers, these AHP and ANP
y
models are used for machines in printing house of the
daily newspaper, “ZAMAN”. At the end of these
b
analyses, weights of three different maintenance
o
policies are determined. This research shows that
t
predictive maintenance is the most suitable
h
maintenance policy for this newspaper firm in both
AHP and ANP analyses. In the future, AHP and ANP
m
models can be used with fuzzy logic for maintenance
e
selection and new models can be done for firms that are
t
in other sectors. Further research on testing other
h
decision making tools including fuzzy logic, may be
o
done. This study can also be extended by adding a new
d
selection factor to the existing model.
s

t
o

c
a
u
s
e
 18,1

Table II.
JQME

Weighted super matrix

26

P1P2P3GS1S2S3V1V2V3V4C1C2C3C4C5I1I2I3

P10.0000.0000.0000.0000.7090.2870.2900.2900.4670.1760.2901.0001.0000.0001.0000.0000.0771.0000.091
P20.0000.0000.0000.0000.1790.6350.6550.6550.4670.2800.6550.0000.0000.7500.0000.0000.4620.0000.091
P30.0000.0000.0000.0000.1130.0780.0550.0550.0670.0440.0550.0000.0000.2500.0001.0000.4620.0000.818
G0.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S10.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S20.0000.0000.0000.4000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S30.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V10.0000.0000.0000.0000.0000.0000.0000.0000.0000.3330.0000.0000.0000.0000.0000.0000.0000.0000.000
V20.0000.0000.0000.0000.0000.0000.0000.0000.0000.1110.0000.0000.0000.0000.0000.0000.0000.0000.000
V30.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V40.0000.0000.0000.2780.0000.0000.0000.0000.0000.0560.0000.0000.0000.0000.0000.0000.0000.0000.000
C10.1810.0750.0420.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
C20.0850.0750.0420.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
C30.0170.6770.5890.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
C40.2020.0750.0420.1440.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
C50.0150.0970.2860.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
I10.1000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
I20.3000.0000.0000.1780.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
I30.1000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
 P1P2P3GS1S2S3V1V2V3V4C1C2C3C4C5I1I2I3

P10.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.2270.227
P20.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.1460.146
P30.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.1270.127
G0.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S10.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S20.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
S30.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V10.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V20.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V30.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
V40.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.000
C10.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.0570.057
C20.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.0360.036
C30.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.1780.178

Maintenance strategy selection

C40.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.0620.062
C50.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.0540.054
I10.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.023
I20.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.0680.068
I30.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.0230.023

27
 Table III. Limit matrix
JQME In addition, other well known multi-criteria methods such as TOPSIS, ELECTRE can 18,1 be used to
compare the results of this work.

References
Almeida, A.T. and Bohoris, G.A. (1995), “Decision theory in maintenance decision making”,
28 Journal of Quality in Maintenance Engineering, Vol. 1 No. 1, pp. 39-45.
Al-Najjar, B. and Alsyouf, I. (2003), “Selecting the most efficient maintenance approach using
fuzzy multiple criteria decision making”, International Journal of Production Economics,
Vol. 84, pp. 85-100.
Arunraj, N.S. and Maiti, J. (2010), “Risk-based maintenance policy selection using AHP and
goal programming”, Safety Science, Vol. 48, pp. 238-47.
Azadivar, F. and Shu, V. (1999), “Maintenance policy selection for JIT production systems”,
International Journal of Production Research, Vol. 37 No. 16, pp. 3725-38.
Bertolini, M. and Bevilacqua, M. (2006), “A combined goal programming-AHP approach to
maintenance selection problem”, Reliability Engineering and System Safety, Vol. 91, pp.
839-48.
Bevilacqua, M. and Braglia, M. (2000), “The analytic hierarchy process applied to maintenance
strategy selection”, Reliability Engineering and System Safety, Vol. 70, pp. 71-83.
Chin, K-S., Chiu, S. and Tummalo, R.V.M. (1999), “An evaluation of success factors using the
AHP to implement ISO 14001-based EMS”, International Journal of Quality & Reliability
Management, Vol. 16 No. 4, pp. 341-62.
HajShirmohammadi, A. and Wedley, W.C. (2004), “Maintenance management – an AHP
application for centralization/decentralization”, Journal of Quality in Maintenance
Engineering, Vol. 10 No. 1, pp. 16-25.
Ilangkumaran, M. and Kumanan, S. (2009), “Selection of maintenance policy for textile
industry using hybrid multi-criteria decision making approach”, Journal of
Manufacturing Technology Management, Vol. 20 No. 7, pp. 1009-22.
Labib, A.W. (2004), “A decision analysis model for maintenance policy selection using a
CMMS”, Journal of Quality in Maintenance Engineering, Vol. 10 No. 3, pp. 191-202.
Labib, A.W., O’Connor, R.F. and Williams, G.B. (1998), “An effective maintenance system
using the analytic hierarchy process”, Integrated Manufacturing Systems, Vol. 2 No. 9,
pp. 87-98.
Li, J.R., Khoo, L.P. and Tor, S.B. (2006), “Generation of possible multiple components
disassembly sequence for maintenance using a disassembly constraint graph”,
International Journal of Production Economics, Vol. 102, pp. 51-65.
Luce, S. (1999), “Choice criteria in conditional preventive maintenance”, Mechanical Systems
and Signal Processing, Vol. 13 No. 1, pp. 163-8.
Mechefske, C.K. and Wang, Z. (2003), “Using fuzzy linguistics to select optimum maintenance
and condition monitoring strategies”, Mechanical Systems and Signal Processing, Vol. 17
No. 2, pp. 305-16.
Okumura, S. and Okino, N. (2003), “A maintenance policy selection method for a critical
single-unit item in each workstation composing a FMS with CBM optimization”,
International Journal of COMADEM, Vol. 6 No. 2, pp. 3-9.
Partovi, Y.F. (1994), “Determining what to benchmark: an analytic hierarchy process
approach”, International Journal of Operations & Production Management, Vol. 14 No.
6, pp. 25-39.
Rausand, M. (1998), “Reliability centered maintenance”, Reliability Engineering System Safety,
Vol. 60 No. 2, pp. 121-32.
Saaty, T.L. (1980), The Analytic Hierarchy Process, McGraw-Hill, New York, NY. Maintenance
Sharma, R.K., Kumar, D. and Kumar, P. (2005), “FLM to select suitable maintenance strategy strategy
in process industries using MISO model”, Journal of Quality in Maintenance
 Engineering, Vol. 11 No. 4, pp. 359-74. selection
Shorrocks, P. (2000), “Selection of the most appropriate maintenance model using a decision support
framework”, unpublished report, UMIST, Manchester.
Shorrocks, P. and Labib, A.W. (2000), “Towards a multimedia based decision support system for
29
word class maintenance”, Proceedings of the 14th ARTS (Advances in Reliability
Technology Symposium), IMechE, University of Manchester, Manchester.
Shyjith, K., Ilangkumaran, M. and Kumanan, S. (2008), “Multi-criteria decision-making
approach to evaluate optimum maintenance strategy in textile industry”, Journal of
Quality in Maintenance Engineering, Vol. 14 No. 4, pp. 375-86.
Triantaphyllou, E., Kovalerchuk, B., Mann, L. and Knapp, G.M. (1997), “Determining the most
important criteria in maintenance decision making”, Journal of Quality in Maintenance
Engineering, Vol. 3 No. 1, pp. 16-28.
Waeyenbergh, G. and Pintelon, L. (2002), “A framework for maintenance concept development”,
International Journal of Production Economics, Vol. 77, pp. 299-313.
Wang, H. (2002), “A survey of maintenance policies of deteriorating systems”, European Journal
of Operational Research, Vol. 139, pp. 469-89.
Wang, L., Chu, J. and Wu, J. (2007), “Selection of optimum maintenance strategies based on a
fuzzy analytical hierarchy process”, International Journal of Production Economics, Vol.
107, pp. 151-63.

Further reading
Bhushan, N. and Rai, K. (2004), Strategic Decision Making: Applying the Analytical Hierarchy
Process, Springer-Verlag, London.
Saaty, T.L. (2001), Decision Making with Dependence and Feedback: Analytic Network Process,
RWS Publications, Pittsburgh, PA.
Waeyenbergh, G. and Pintelon, L. (2004), “Maintenance concept development: a case study”, International
Journal of Production Economics, Vol. 89, pp. 395-405.

Corresponding author
Omer F. Demirel can be contacted at: odemirel@fatih.edu.tr

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit

our web site for further details: www.emeraldinsight.com/reprints

View publication stats