The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/1355-2511.htm

Opportunistic
Development of opportunistic maintenance
maintenance policy towards policy

optimal maintenance system

(OPTOMS) model 129

A case study Received 2 July 2018

Revised 28 November 2018
Accepted 6 January 2019
Hasnida Ab-Samat
School of Mechanical Engineering,
Universiti Sains Malaysia, George Town, Malaysia, and
Shahrul Kamaruddin
Department of Mechanical Engineering,
University Technology Petronas, Bandar Seri Iskandar, Malaysia

Abstract
Purpose – Opportunistic maintenance (OM) policy is a prospective maintenance approach that instigates for
a more effective and optimized system. The purpose of this paper is to provide the steps and methods used in
model development processes for the application of the OM policy.
Design/methodology/approach – Dubbed as opportunistic principle toward optimal maintenance system
(OPTOMS) for OM policy toward optimal maintenance system, the model is devised as a decision support
system model and contains five phases. The motivation and focus of the model resolve around the need for a
practical framework or model of maintenance policy for the application in an industry. In this paper, the
OPTOMS model was verified and validated to ensure that the model is applicable in the industry and robust
as a support system in decision making for the optimal maintenance system.
Findings – From the verification steps conducted in a case study company, it was found that the developed
model incorporated simple but practical tools like check sheet, failure mode and effect analysis (FMEA),
control chart that has been commonly used in the industry.
Practical implications – This paper provides the general explanations of the developed model and tools
used for each phase in implementing OM to achieve an optimal maintenance system. Based on a case study
conducted in a semiconductor company, the OPTOMS model can align and prepare the company in
increasing machine reliability by reducing machine downtime.
Originality/value – The novelty of this paper is based on the in-depth discussion of all phases and steps in
the model that emphasize on how the model will become practical theories in conducting an OM policy in a
company. The proposed methods and tools for data collection and analysis are practical and commonly used
in the industry. The framework is designed for practical application in the industry. The users would be from
the Maintenance and Production Department.
Keywords Preventive maintenance, Corrective maintenance, Maintenance optimization,
Failure mode and effect analysis (FMEA)
Paper type Case study

1. Maintenance system in theory vs in the industry

The basic objective of maintenance activities is to ensure that production equipment and
support items are in good condition, serviceable and safe to operate. The key for achieving
an effective and successful maintenance system is to have optimal maintenance activities
planned and scheduled according to machine specification and production planning (Parida
et al., 2015). However, the optimization process is the major challenge in the maintenance Journal of Quality in Maintenance
system as the industrial environment is characterized by various uncertainties ( Jardine and Engineering
Vol. 26 No. 1, 2020
pp. 129-165
The authors acknowledge the RU-FRGS Grant (1001PMEKANIK8045021) provided by Universiti © Emerald Publishing Limited
1355-2511
Sains Malaysia for funding the studies that resulted in this paper. DOI 10.1108/JQME-07-2018-0057
JQME Tsang, 2006). There is a big gap between theory and practice in a maintenance optimization
26,1 as discussed by Dekker (1996). The gap is discussed in six main aspects.
The first one is about how maintenance optimization models are mostly developed
based on stochastic nature (random probability distribution of failure rate, equipment
ages or costs). Thus, it is hard to be understood, interpreted or applied by practitioners
like technicians, engineers and managers in the industry. Adoption of probability
130 functions like Weibull distribution, Poisson analysis and genetic algorithm tools like
Markov decision is the opposite of the deterministic approach used in the industry.
Second, most publications on the optimal maintenance system focus on mathematical
analysis and techniques that do not portray the real scenario in the industry. The
numerical examples used in the calculation processes are rarely according to case study
and, thus, do not provide the solutions to industry problems. Therefore, it makes
practitioners hardly to have any interest in the research works and fails to apply the idea
in real industry and further test its optimality function.
The third aspect that rifted theory and practice is little communication between
researchers and industry, which halted the growth and sharing of knowledge in the
maintenance area. The argument is that even though both parties need each other to find the
optimal maintenance system, researchers are usually focused solely on the theories being
studied, whereas companies are not interested in publications. This somehow creates the
imbalance between theories and practices. In both ways, maintenance is at loss. In the fourth
issue, the gap between theory and practice is apparent due to the fact that maintenance
covers variety of aspects making the standardization and development of general model a
difficult task. Multiple models published are usually catered for a certain situation and for a
specific process or equipment. For that reason, the models are not suited for practical
problems in the industry. The fifth point is based on how optimization might not always be
necessary. Because optimization involves complex decision-making process and is based on
various criteria, sometimes the optimization is not critical in practice.
The final issue in the optimization model is related to the wrong type of maintenance
policy used. It is understood that there are many approaches to maintenance and each one is
suitable for a different condition. The ideal maintenance plan is to replace or repair the
component right before it fails (Crocker and Kumar, 2000) where failure can be avoided and
the component can be used to its maximum lifetime effectively and efficiently. All these
issues and gaps need to be addressed properly in order to have an effective implementation
of maintenance policy in the industry. The main understanding about the maintenance
system for its application in the industry and the gaps between theories and practical is that
there is a need for a proper framework or a model that works to translate the opportunity
principles into practice in order to gain the optimal maintenance system. For that, the
subsequent sections of this paper will present the discussion of opportunistic maintenance
(OM) policy and the explanation of each phase in the opportunistic principle toward optimal
maintenance system (OPTOMS) model.

2. Focus and motivation of OPTOMS model

The focus for the development of the OPTOMS model is to effectively apply OM policy in
the real industry in the effort to achieve the optimal maintenance system. OM is a
combination of preventive maintenance (PM) and corrective maintenance (CM) in the
scheduling of maintenance activities (Ab-Samat and Kamaruddin, 2014). As mentioned in
the earliest publications on OM by Rander and Jorgenson (1963) as well as by McCall (1963),
OM policy is described to the use of the concept of “opportunity.” The concept is illustrated
in Figure 1. It can be chronicled with an occurrence of a failure of a component in a system or
on a machine during operation that will result in downtime and stoppages of the whole
system. Nevertheless, this situation can be used as an opportunity to conduct PM
simultaneously on the other component in a similar system. However, the key to be Opportunistic
observed is that the preventive task selected should satisfy a few rules in order to have an maintenance
optimal maintenance system. policy
For the development of OPTOMS model, a few fundamental ideas and principles of OM
policy were used. These ideas will be described as notable points from a concept perspective
throughout this subsection.
131
2.1 Components in a system are dependent on one another
The first point of OM policy is the concept of dependency of components on one another
when operating on the same system. Without doubt, most machines or the operating
system used in a production plant is built with numbers of component and with various
structures. Whether the components work in series or in parallel, its operating condition
can impact the overall machine operation as well as its availability and reliability
(Xia et al., 2017). In a machine, there are also multiple components assembled together and
are dependent on one another to function. Each of these components will degrade with
time or usage, and even suffer from a sudden failure due to the random variables. This
means, any failure of a component might be related to another component and this is
considered as a dependency. Ozekici (1988), Bevilacqua and Braglia (2000), Castanier et al.
(2005) and Vu et al. (2012) discussed in depth of the concept of dependency. Dependency is
not limited to operational purpose only, but also related to the terms of failure and
economic function (Scarf and Deara, 2003). In a specific application, Derigent et al. (2009)
followed the dependency principle in the terms of component proximity or closeness both
in physically and time-based.
In this research, component dependency is the notable concept used in the OPTOMS
model, which allows maintenance activities to be simultaneously conducted on failed and
failure-related yet operating component. The concept of dependency in OM policy is used to
conduct PM on one component and conduct CM on other components related to the failed
one. Therefore, the principle of component dependency on one another in a system becomes
the key point in the OPTOMS model. The aim is to optimized maintenance activities during
machine stoppages. This is more important for some systems, which are needed to be
operated continuously and most of the time cannot afford to constant failure. For example,
when a machine like a pump or a generator in industrial plant fails to work properly, it
interrupts the whole production assembly line and costs a large amount of capital and labor.
The optimal maintenance system can be achieved if component dependency traits of OM
policy are applied.

2.2 The failure of a component is an opportunity for maintenance for another component
The second notable aspect from concept perspective is the principle of OM that is any failure
of a component in a system is considered as an opportunity to conduct PM tasks on another
component in the same system. This concept is related to the first notable idea discussed in

Failure of
Component

Machine
Downtime

Figure 1.
Opportunity Concepts of OM policy
JQME the previous point. In the industry, continuous operation is much desired to maximize
26,1 resources and increase production rates. Therefore, ideal machine stoppages are only done
according to maintenance planning and scheduling that had been tallied with production
planning. As so, any machine stoppages resulting from unavoidable failure should be
effectively used as an opportunity to conduct maintenance activities required for the whole
system. The use of the stoppages in conducting maintenance of group of components can
132 save maintenance cost and machine downtime.
From the review conducted, opportunities were defined by various researchers in various
perspectives. To name a few, Dekker and Smeltink (1991), Rao and Bhadury (2000), Dekker
and van Rijn (2003), Saranga (2004) and Cui and Li (2006) are the research works that agreed
on a component failure to become an opportunity to correct the failed component and at the
same time to utilize the downtime to conduct PM activities. The concept was also the
description used by McCall (1963) and Rander and Jorgenson (1963) who were responsible to
introduce the OM policy in the maintenance system. It can be stated that the machine
downtime to repair a component is an “opportunity” to maintain other components in the
system. It means any production stoppages due to a component failure is an opportunity and
can be taken advantage of by conducting maintenance activities on other related components.
Failed component attended with CM policy is also referred to as failure-based
maintenance or breakdown maintenance policy. It is the oldest maintenance policy and the
original maintenance policy implemented in the industry among the maintenance policies
(Mechefske and Wang, 2001). According to Sharma et al. (2005), this policy is considered a
feasible policy to be adopted in the cases where profit margins are large. And, despite the
fact that this maintenance policy may cause large production losses, serious damage to the
system, person and environment due to the unexpected failure, it is still applied till today.
Due to the stochastic occurrences of failure, time to conduct CM activities is unexpected
and unpredictable. Nevertheless, it becomes the perfect opportunity to conduct PM activities
on the failed system. CM activities are conducted when a failure occurs in equipment, and
then steps are taken to rectify the situation or restore the equipment to productive capability
as quickly as possible, whereas PM is the opposite of this. PM tasks are usually pre-planned
and scheduled tasks planned according to component specifications and historical failure
analysis. Hence, the time for PM activities are fixed on weekly, monthly, quarterly or yearly
basis. By using the opportunity concept, any PM activities that take place during the failure
should be conducted with CM activities in order to reduce machine downtime, prevent
future failure and also reduce maintenance cost.

2.3 Optimal maintenance system with component proximity

The third concept used as a fundamental idea for the development of the OPTOMS model is
component proximity to achieve optimal maintenance system when applying OM policy. As
discussed in research works and publications of OM that mostly relates the policy to an
optimal maintenance system the model development would focus on component proximity
concepts. The initial concepts (dependency and opportunity) of OM policy satisfy most of the
characteristics of an optimal system. In the OPTOMS model, the dependency of component is
extended on the basis of the relation between components in a system, which becomes the
decisive factor when deciding maintenance activities in a system.
Research by Derigent et al. (2009) presented a mathematical morphology method for the
decision making in their research. The concept used is in component proximity. The
principle was used to choose component based on physical and economic proximities or
closeness whenever opportunity arises. Based on the approach used by Derigent et al. (2009),
this research ventures further on the concept and focuses more on the practicality side with
the usage of physical component dependency with the component proximity as its factor in
decision making. The points highlighted in the phase are all related to the process of finding
the way to decide maintenance-significant component among the un-failed component so Opportunistic
that maintenance is at the minimum and remaining component lifetime is not wasted by maintenance
premature maintenance activities. policy
The key point of the model is it works as a decision support system in maintenance
planning and scheduling. The model is developed in the form of a framework with phases and
steps that can be applied in the process of achieving an optimal maintenance system using OM
policy. Due to the nature of the maintenance condition that involves more than one performance 133
measure, it has deterministic and stochastic approaches, respectively, for different situation
desired by the user/company. The model will allow the user to make a selection of optimization
based on its company’s mission and vision. The model is developed to show how to apply the
OM principle in the industry that usually uses PM and CM policy. The OM concept applied
in the model is to use CM time as opportunities and to conduct PM that involves restoration
task where a component that has its accumulative deterioration is reduced or eliminated.

3. OPTOMS model
The term “model” here refers to a range of concepts, principles and methods that are
structured together to achieve certain objectives. The model is developed in five main
phases. The first phase in the OPTOMS model is developed with the emphasis on the idea
that maintenance is an important element in a company operation. Therefore, the steps in
this phase are related to identifying maintenance objective that is consistent with the
company’s mission and vision. Once identified, maintenance tasks planned will be focusing
on the performance measures that relate to the company’s aim and future plan. Then, next
phase proposes that critical or important machine in the whole operation is to be selected for
implementation of OM policy. Selection criteria can be either by the experience of
maintenance engineer or analysis of failure occurrences of machines. The one with the
highest downtime will be chosen for improvement.
In the third phase, PM planning, checklist and schedules of the selected machine is studied.
The PM tasks assigned for the machine need to be identified along with its duration and
arranged time. Also, a more detailed analysis of failure and downtime on the critical machine
is carried out at this phase. The processes aim to use the failure time as an opportunity,
whereas the failure mode is used as a determining factor for the next phase steps.
Consequently, Phase IV contains the key of OPTOMS model. Here, the step-by-step process of
deciding principle, concept, rule, assumption and some limitations of OM policy is determined.
Using information from the previous three phases, OM policy is determined so that it is
applicable in the industry. Finally, the model is developed with Phase 5: execution and control
of optimal maintenance system. In this phase, the control chart is prepared for the process of
performance measurement of the maintenance system. This last phase in the OPTOMS model
suggests that the optimization of the maintenance system can be conducted to improve the
effectiveness as well as the efficiency of the initial PM approach.
There are many mathematical models, numerical analysis and simulations conducted in
research works of OM policy, yet it lacks practical theory introduced for the application of
policy in the industry. In the industry, theories in academic publications need to incorporate
the industry’s situations and practice in order for OM to work. The issues that create gaps
between theory and practice need to be addressed. And, a practical model with optimal
maintenance policy is also needed. Therefore, the OPTOMS model developed in this
research serves to reduce the gap and introduce a practical framework to apply OM policy in
the industry.
The benefit of this model is that it is a decision support system that provides an objective
and quantitative way for decision-making process when conducting maintenance activities.
The model works as a void filler between theory and practice, which is much needed in the
industry. From an industrial point of view, OM can be very useful for a fully automated
JQME manufacturing system in order to reduce unplanned downtime to zero and to ensure smooth
26,1 and continuous production at minimal cost. From a structural perspective, the model holds a
value in the following form.

3.1 Input, process and output function for each phase

OPTOMS is developed in five main phases. Referring to the full flow of the model, the whole
134 model can be re-arranged in input, process and output function, as shown in Figure 2.
As an effort to validate the developed OPTOMS model, multiple sources of data are
collected and analyzed. Data work as evidence of case study as collected from documents,
archival records, interviews as well as direct observation in the production lines of the
studied company. Documentations and archival records provide precise and quantitative
evidence, but the weakness is it has limited accessibility due to privacy and confidentiality
reasons (Yin, 1994). Nevertheless, this is the main source of data in the company, and any
assumptions based on probability function will not lead to accurate measurements. The
data collected for the model are information available in the industry without needing to
have additional software for data collection and analysis. The information is also common
as it was used as performance measurement for systems in a company.
The information collected serves as the input for all the phases in the OPTOMS model.
Some phases share similar data source, whereas other extended the analysis of the
information collected. Then, each output from the phase in the OPTOMS model will become
the input to its subsequent phase. This provides a cycle of information that ensure the whole
model works in achieving one main objective of achieving the optimal maintenance system
using OM policy. Thus, the proposed model for optimal maintenance system with the use of
the OM technique aims to provide a structured, effective yet simple analysis process that
can aid users in achieving optimal maintenance activities.

3.2 Adoption of practical tools and methods for data analysis

The focus of analysis tools chosen form the OPTOMS model is on its simplicity function. Yet
the tools should offer a multiple functions so that data can be flexibly calculated as found suited
to a company’s situation. There are five tools suggested in the model. These are check sheet,
stacked-bar chart, failure mode and effect analysis (FMEA), FBD and finally control chart.

• Machine Downtime
• PM Planning and Schedule
INPUT • PM Checklist
(Data) • Machine Specification
• Failure Record
• Production rate and cycle time

• Phase I: Objective Identification

• Phase II: Critical Machine Selection
PROCESS
• Phase III: PM and Failure Analysis
(Phase)
• Phase IV: OM Rule Setting
• Phase V: System Optimization

• Maintenance Objective
• Critical Machine
Figure 2. • Scheduled Maintenance
Input, processes OUTPUT Activities
and output in (Result) • Failure Analysis
OPTOMS model • Performance of Maintenance
System
As suggested in the first phase, check sheet is used to tally the company’s mission and vision Opportunistic
with the characteristics of objectives commonly used in the maintenance system. This way, maintenance
improvement efforts for the maintenance system can be the direct contributing factor to the policy
company’s success in the future.
Then, in the second phase, there is stacked bar chart to graphically assess machines
downtime and failure rate and find the most critical machine for improvement. The distinctive
function of this type of bar chart is to show the machine downtime in two characteristics that 135
are a number of scheduled downtime and unscheduled downtime. Therefore, the chart offers
two different choices for users to decide on the critical machine based on their preferences.
Moving forward, the third phase of OPTOMS model consists of the failure analysis tool
commonly applied in the form of FMEA. For FMEA application, the analysis provided a clear
structure for assessing a large number of failures to give a better understanding view of the
problems faced by the production line and manufacturing facilities particularly. According to
Kmeta et al. (1999), FMEA is a logical approach that can completely capture all failure modes,
effects and causes. Besides, the suitable maintenance action could also be planned from the
information gained during the analysis. This way, maintenance actions will be more accurate
and effective to counter the failures identified. Thus, it was adapted to the OPTOMS model
where the aim of the model is to analyze the failure that occurs in the process or machine.
The final tool suggested for the model development is control chart, which was
suggested in the final phase of OPTOMS. Here, the application focuses on the tool that
enables the observation of maintenance performance from time to time. In relation to
continuous improvement made in the maintenance system, the control chart provides
excellent performance measure functions for the user (Zhou and Zhu, 2008; Yang et al.,
2018). Ben-Daya and Rahim (2000) suggested the X-bar control chart for the purpose of
controlling and determining the state of process as used in this research. Due to the
diversion of maintenance objectives, this chart suggests limit lines that are acceptable
according to the operation of the system applied to OM policy. Aside from the tools applied,
another issue addressed in the development of OPTOMS model is the big gap between
theory and practice as discussed by Dekker (1996).
From the observations of various maintenance models, the gaps can be identified in
many factors. As in optimal system, a theory is developed based on data from a real system,
but hypothesis, assumptions and development of what considered as “perfect system”
actually make the theory not wholly practical. The deeper the research is, the wider the gap
will be. Although assumptions and limitations are acceptable because research works
needed to have effective solution and optimal system, the practice sometimes becomes a
problem for implementation of the theory in the real situation/industry. The “perfect
system” developed in research works differs too much with practice in the industry, thus
hindering the process of testing the theory. The similar situation is focused on throughout
this research; therefore, Phase IV is equipped with simple yet important assumptions and
limitation when applying the OM policy.
The discussion about the proposed model, including its concepts and elements
considered, has been shown to be satisfied and credible. Hence, it can be concluded that the
factors and data set derived from the OPTOMS model to evaluate maintenance system are
practical for the purpose of better understanding of steps in achieving the optimal
maintenance system with the application of the OM policy. The detailed flow of the whole
OPTOMS model can be viewed in Figure 3.

4. Phase I: identification of maintenance objectives based on company’s

mission and vision
The first phase of the model is aimed at the fact that all maintenance tasks are
conducted with the company’s mission and vision as its guidelines for improvement.
JQME DATA
OPPORTUNISTIC PRINCIPLE TOWARD
OPTIMAL MAINTENANCE SYSTEM (OPTOMS)
26,1 METHOD
Company’s
PHASE I
Mission
Identification of Maintenance Objective Check Sheet
and Vision
Based on Company’s Mission and Vision
Statement

136 Maintenance
Objective

Process
Manual PHASE II Stacked-Bar
Prioritization of System/Machine Chart
Machine
Operating
System Critical Machine

PHASE III
Assessment of Preventive Maintenance PM Checklist
Downtime (PM) Plan and Schedule
Analysis
Mode
Causes Effects
PM Plan FMEA
Failures

PM Plan and
Opportunities

PHASE IV
Production Decision on Opportunistic Maintenance
Planning (OM) Principle and Rule
FBD

Component Proximity
Failure concept
Analysis

OM Policy
Component’s
Specification
PHASE V
Execution and Control for Optimal Control Chart
Cost Maintenance System

Figure 3. Failure Age Cost

The flow of Machine Machine Maintenance
OPTOMS model Availability Reliability Cost

To change the mentality that a maintenance system is a non-added value for an

organization, the system performance should be measured and improved according to
what company’s aims to achieve in the future. The relation between a company’s
mission and vision with the maintenance system is best shown by Al-Turki (2011) as
linked in Figure 4.
 Maintenance Strategic Middle range
Opportunistic
Plan planning, forecasting, maintenance
capacity planning, policy
Maintenance vision, goals and targets
mission and objectives

Corporate Maintenance 137

strategy, vision, Strategies Short term planning
mission, (outsourcing, and Scheduling
objective structuring,
methodologies and
support systems) Implementation Figure 4.
Relation between
Performance corporate strategy,
Measurement System Performance vision, mission
and objective
Measurement with maintenance
strategic plan
Source: Al-Turki (2011)

Figure 4 shows the direct relation of corporate strategy, vision, mission, and objective with a
strategic maintenance plan in the company that is then used for planning and scheduling
the implementation and performance measurement process. Company’s strategy, vision,
mission and objective should be used as a definite factor in maintenance planning. It will
become the very first step or the starting milestones toward an optimal maintenance system.
Phase I in this OPTOMS model is developed with two steps in translating mission and
vision into maintenance objective.

4.1 Step 1: mission and vision statement identification

When a company is first built, the pillars of operation are designed in the form of business
mission and vision. Therefore, the first step in this phase is to identify the company’s
mission and vision. The mission and vision are the summaries of goals and objectives the
organization wants to achieve in the future. The company’s mission also defines the purpose
and primary objectives related to customer needs and business principle. For the company’s
vision, it mostly works as direction and inspiration for improvement for a better future.
Therefore, before applying OM policy, a company needs to identify and write down their
mission and vision.

4.2 Step 2: maintenance objective determination using check sheet

Maintenance objectives need to be thoroughly understood in order to develop a good
maintenance system ( Jiang and Ji, 2002). The second step in the OPTOMS model is focused on
channeling aim, focus and future plan of a company to achieve the maintenance objectives for
performance measures and improvement plans. The process is illustrated in Figure 5.
However, it is not an easy task to directly interpret the company’s mission and vision
into factors to measure maintenance performance for improvement. This poses no problem
in any situation because logically maintenance optimization can be done in various ways.
Yet, it becomes an aggravation in real-world application due to the fact that any changes or
improvements need to be based on significant factors in the process. To remedy this
situation, there is the need to find common objectives when conducting maintenance. Then,
the company’s mission and vision will be used as a guide to choose optimal criteria for
maintenance optimization using OM policy.
JQME
26,1

Company’s Mission Maintenance

and Vision Objective

138

Figure 5. Aim Optimal Criteria

Transformation of Objective Improvement Factors
corporate mission Performance Measures
and vision into Achivement Target
maintenance objective Specialty/Value
Future Goal

The diversity of maintenance objective of OM policy can be simplified into three main
objectives. The first one is to reduce the number of equipment breakdown, stoppages,
downtime and failure. Second, OM should able to increase equipment availability,
operating time, and reliability and lifetime and finally to minimize maintenance cost.
These objectives can be divided into three main groups that are regarding the system
failure rate, the age of component or equipment and lastly the cost involved in
maintenance tasks. The OPTOMS model considers three main maintenance performance
measures as maintenance objectives. These objectives becomes optimal criteria for
optimization of maintenance system as depicted in Table I. The table also lists the factors
that related to each group. The factors should be used to tally the company’s mission and
vision identified in the previous step.
In this step, the relation of the company’s mission and vision and maintenance
objective can be found using the check sheet. This method is effective to convert
qualitative information into quantitative data (Schultz, 2006). Also known as Tally Chart,

Table I.
Check sheet to
identify maintenance
objective
this method utilizes a simple table to record the frequency of coloration or connection Opportunistic
between mission and vision with elements of maintenance objectives listed. The frequency maintenance
is decided based on an expert judgment by a Maintenance Engineer or any high policy
management personnel responsible for the maintenance department. The frequency of
relation can be marked by drawing strokes in the worksheet. Then, a maintenance
objective that has the most number of stroke or mark can be selected as performance
measures for an optimal maintenance system in the company. Table I shows the form to 139
collect and analyze the mission and vision statement.
Since mission and vision are rarely specific and mostly vague, each statement can have
more than one coloration with maintenance objective. The key point of this phase is to find
the maintenance objective that most likely will help companies achieve their future goal.
The maintenance objective that has the most frequency selected will be used as
performance measurement throughout the OPTOMS model as well as in the final phase of
the model.

5. Phase II: prioritization of system/machine

For manufacturing plant, it is definitely a common practice to have more than one process to
produce a product. Therefore, production plant consists of a series of machines and
equipment that make it a complex system to begin an improvement process. And, due to
cost, time and resource constraints, it is impossible for maintenance management to analyze
and conduct improvement on all machines at once. For that, any implementation should be
conducted stage by stage in order to simplify the process as well as to ease observation.
Thus, the best solution is to identify and select one critical or important system or process as
an initial study point. The second phase of OPTOMS model involves the selection of a
system or machine to apply the OM policy. For the OPTOMS model, the selection process
starts with the identification of the main issue that occurs in a manufacturing plant. In this
phase, stacked-bar chart is selected as the analysis tool and the prioritization process is
developed in three steps.

5.1 Step 1: scheduled and unscheduled downtime identification and segregation

The first step in Phase II of OPTOMS model is to identify each machine downtime and
segregate it into the scheduled and unscheduled groups. The step of downtime tracking is
essential for the maintenance system as downtime costs company lots of money each year
resulting in loss of production, a downgrade of product quality and loss of the customers.
The segregation of downtime can be done based on the guidelines given in Table II.
Data for this phase can be obtained from any daily downtime tracking system that
documents details of machine downtime. Inside the company, downtime collection is done
by the machine’s operator who simply fills in a log book, noting what happened, what was
done and how long the machine was down. Then, maintenance personnel responsible for
repair or replacement fill in other information on what type of maintenance activities

Scheduled downtime Unscheduled downtime

Characteristic Planned machine stoppages for basic maintenance Unplanned machine

stoppages due to failure
Production Expected/planned by the production department Unexpected/not planned by Table II.
the production department Difference between
Example PM tasks, calibration, change of component/ Failure/breakdown, idling, scheduled and
product, setup and adjustment, start-up, upgrades minor stoppages unscheduled
Maintenance policy used Preventive maintenance (PM) Corrective maintenance (CM) downtime
JQME conducted and time maintenance finished. Depending on the company practice, this
26,1 information is either collected daily or monthly and recorded in spreadsheets and databases
usually using computer software.
For an effective recording of machine downtime, specifically the lists of data that need to
be collected are:
(1) time of machine downtime (to the minute);
140 (2) time maintenance start and finish (to the minute);
(3) type of failure, such as jammed, stopped, tripped, leaking, etc.;
(4) component/equipment are that failed;
(5) maintenance activities such as repair, replaced, filled, cleaned, etc.;
(6) personnel reporting the machine failure; and
(7) personnel conducting the maintenance.
The duration for data collection can be a minimum of one month and maximum throughout
the year. Data from six months up to one year of operation would be sufficient in identifying
the critical machine in the system/process to ensure more accurate analysis and result.

5.2 Step 2: total machine downtime calculation

The next step is to combine both scheduled and unscheduled downtime in order to compute
the total machine downtime for each machine. This step needs to be conducted so that the
machine that has the highest downtime can be prioritized and identified as a critical
machine. These data will be used for further analysis in the next step of Phase II.

5.3 Step 3: critical machine prioritization

In this research, a critical machine refers to the machine that has the highest value
of downtime. Based on numerical data obtained from Step 2, the next step is to
transform the numbers into a graphical form. In order to select the system/process with
the highest number of machine downtime in a production line, stacked-bar chart as in
Figure 6 can be drawn. This type of chart is routinely used in maintenance engineering to
identify failures responsible for the majority of the machine’s maintenance cost and
operating downtime.
The selection of critical machine is based on total downtime occurring in the whole
system. As in the example given in Figure 6, out of five machines analyzed, Machine 4 has
the longest bar that shows that it encounters the highest number of machine downtime

Total machine downtime

5 1.7 3.4

4 4.5 2.8
Machine

3 3.5 1.8

2 2.5 4.4
Figure 6.
Stacked bar chart for
both scheduled and 1 4.3 2.4
unscheduled
Downtime
downtime
Scheduled Downtime Unscheduled Downtime
during the observed time. A total of 7.3 h of downtime with 4.5 hours is planned and another Opportunistic
4.8 h is unplanned. Therefore, Machine 4 will be chosen as a critical machine for analysis in maintenance
the next phases of the OPTOMS model. policy
6. Phase III: assessment of preventive maintenance (PM) activities and
failure analysis
Once the critical machine is identified, the OPTOMS model is directed further to analyze and 141
study the PM’s plan for the machine. This phase also includes the analysis of failures using
the FMEA method. The motivation of this phase is to have complete maintenance
information that can be referred when planning maintenance activities according to OM
principle and rule in the next phase.

6.1 Step 1: preventive maintenance plan and checklist assessment

Started in the 1960s, the OM concept has been widely and numerically studied for
maintenance scheduling for multiple components, single component, two components as
well as multi-equipment and multicomponent systems. In practical, the maintenance system
is embodied by two basic maintenance policies. The first one is PM where maintenance
activities are planned and scheduled tailored to machine specification and component
lifetime. The second policy is CM that is conducted once failures occur. Therefore, the first
step in Phase III of the OPTOMS model requires the study of PM plan, checklist, and
schedule on the previously selected machine. In the industry, the PM is a time-based policy
and some maintenance activities are conducted daily. The activities involved are basic
processes of cleaning, inspection, oiling, re-tightening of a screw, adjustment, lubrication,
measurement, testing, repair, replacement, calibration, modification and the extension of
machine lifetime.
PM activities are planned and scheduled, with the belief that a regular maintenance
attendant will keep failures at bay (Basri et al., 2017). Maintenance under the PM policy is
planned and performed over a specified period of time or the amount of the equipment used
to aim to reduce the probability of failure during its operation (Mann et al., 1995; Kimura,
1997). PM is scheduled in different duration like weekly, monthly, quarterly, yearly and
some as per manufacturer’s requirements. Major PM is performed often based on
manufacturer recommendation and has a fixed time interval of 6 months, one year or five
years (Besnard et al., 2009).
When conducting PM policy in a system, there are the requirements to:
(1) determine the type and frequency of inspections and maintenance procedures
needed;
(2) define the minimum requirements for servicing and maintaining equipment/system/
plant;
(3) develop a comprehensive maintenance checklist for every equipment;
(4) prepare and supplement specific instructions, manufacturer publications, machine
specifications and standards related to the system and process;
(5) establish proper guidelines and instructions; and
(6) state specifically the safety requirements for equipment design, construction, set-up,
operation, installation, dismantling and transportation of machine/system.
The observation suggested in the first step of this Phase III is focused on PM checklist.
Maintenance checklist is a technical information paper that contains specific step-by-step
instructions for maintenance tasks depending on machine types and operation. The
JQME checklist most commonly contains activities like checking, cleaning, lubricating or minor
26,1 maintenance. In practice, this checklist is used as a guideline for daily maintenance for every
machine in order to ensure it is fit for operation.

6.2 Step 2: failures, causes and effects analysis

Once Step 1 is completed, the next step in Phase III of OPTOMS model focuses further on
142 failure analysis by identifying cause and effect of the failures. If PM is half of OM policy,
then another half or the opportunity part comes from machine downtime caused by
component’s failure. In the industry, the failure is definitely unscheduled or unexpected, and
it is repaired based on CM policy. CM is performed immediately after there is a failure to
restore any failed equipment/system to its normal operational condition.
Like failure, CM is unpredictable and cannot be integrated into maintenance planning
and scheduling. However, to remedy the problem, its stochastic occurrences (a probabilistic
model that uses randomness to account for immeasurable factors) can be studied based on
historical data of failures on the machine. In the industry, it is never the issue of whether the
failure happens deterministically or stochastically because machine breakdown happens for
various possible reasons, making it hard to individually study each component and decide
on it. Therefore, in this research, historical data of failure rate are used instead of probability
function to predict failure and decide on opportunities to conduct maintenance.
For the failure analysis method, there are various approaches such as Ishikawa or the
Fishbone diagraming technique, the event and causal factor analysis, change analysis,
barrier analysis, Management Overnight and Risk Tree approach, human performance
evaluation and the Kepner-Tregoe problem solving and decision-making process (Smith and
Hawkins, 2004). However, for conducting the critical machine assessment, the most
applicable and practical method generally used in the maintenance analysis is as FMEA.
FMEA is a structured and a bottom-up approach that starts with knowing the potential
failure modes at one level and then investigates the effect on the next subsystem level
(Sharma et al., 2005).
The purpose of the FMEA is to take actions to eliminate or reduce failures, starting
with the highest priority ones. A good use of the FMEA report can provide management
with several benefits such as higher product reliability, less design modification, better
quality planning, continuous improvement in product and process design, and lower
manufacturing cost, in addition to fulfilling customer requirements (Gary Teng et al.,
2006). As shown in Figure 7, FMEA is a procedure for analyzing the potential failure
modes within a system through the classification of severity or determination of the effect
of failures upon the system.
From the FMEA framework in Figure 7, it can be observed that the first aspect
considered for any failure that occurs is its severity or weight of impact in the whole
process. Any process function experiencing high severity is considered potentially
hazardous failures that should be immediately improved. Besides, the number of
occurrences and also detectability of the failure play an eminent role also where both
demonstrate the failure rate and capability to identify it. From all three factors, the
numbers were then multiplied in order to obtain the risk probability number (RPN). RPN is
used to assess the risk of each failure. The most crucial part of FMEA is the recommended
action section that suggests the improvement activities that should be taken for
eliminating the failures. All recommended action would be based on failure mode, failure
effects, and causes to the failures.
However, the practice of FMEA as a failure analysis approach in Phase III of OPTOMS
model will omit the determination of severity, occurrence, detectability as well as the
calculation of RPN. This is because the focus is to have a list of failure modes, effects, causes
and actions taken in order to make amalgamation with PM activities identified in the
 Identify
Opportunistic
Potential Failure maintenance
Mode policy

Identify
Determine
Potential Effects
Severity
143
of Failure Mode

Identify
Determine
Potential Causes
Occurance
of Failure Mode

Evaluate Current
Controls or Design
Verification Process

Determine Determine
Detectability RPN

Identify Actions Figure 7.

Leading to Basic FMEA
Improvement process flow

previous step of this phase. The ranking of failure risk is not significant as opportunities to
conduct OM policy are random.
The identification of failure modes is based on historical data means not all but only
certain components in a system to be used to apply OM policy. This step eliminates the
impossible tasks to monitor and observe every single component in the selected system, to
only focus on components that have experienced failures resulting from a machine
downtime. The listing of components and its failure mode also enables analysis to be
conducted on components that may share similar failure type, making a selection of PM
tasks becomes easier in the next phases. The failure mode will become a one-off definite
factor for decision-making process in the OPTOMS model.
As an inference, Step 2 focuses on the analysis of failures experienced by the critical
machine based on historical data. The problematic components of the machines are
identified and structurally listed out. The compilation of failures experienced by the critical
machine will be useful for conducting the machine performance measurement in the next
phase as well as improved planning. Further discussion on how to use the information
gathered in this phase for the application of OM policy will be presented in the next section.

7. Phase IV: opportunistic maintenance (OM) principle and rule selection

In the previous phases of OPTOMS model, first maintenance objective was derived from
the company’s mission and vision. Then, a machine with the highest downtime was
JQME selected for analysis and application of OM policy. Subsequently, PM activities and
26,1 previous failures underwent by components in the machine were listed and studied to find
the component’s relation to one another. Now, the fourth phase aims to deliberate the
rules, assumptions and limitations that are needed to be observed when conducting OM
policy. First, Step 1 premeditates on the application of a component proximity rule based
on a dependency factor that will be used in decision making. Then, Step 2 lists the
144 assumptions and limitations in applying the OM policy in the OPTOMS model. The rules,
assumptions, and limitations work as information and guidelines in using the model. In
the model, this information is significant for the proper and structured decision-making
process when performing maintenance tasks. The steps in this phase will incorporate OM
principle and concept during analysis of data. Further descriptions are provided in the
next subsections.

7.1 Step 1: component proximity for PM plan selection

The principle behind OM is that every component in a system has the tendency of being
dependent on one or another. When one component fails, there is a high possibility one or
more components are affected and need to be maintained. Thus, the opportunity may be
taken during the shutdown or CM activities to carry out PM on other maintenance-
significant components that have the potential to fail in the near future (Savic et al., 1995).
As the result, total equipment downtime and maintenance cost can be reduced.
OM policy works with the principle of opportunity utilization like system shutdown or
component failure to perform maintenance tasks on the system that may be required in
the future and subsequently saves a substantial amount of downtime (Samhouri, 2009).
Figure 8 illustrates opportunistic zones between failure occurrence (noted with f ) and the
limit of PM activities planned for the system (noted with T), whereas in Figure 9, it can be
observed that OM policy with occurrences of failure means CM will be conducted along
with related PM activities even if planned PM time is not reached yet. The maintenance
conducted during the opportunity will reduce the duration as well as a number of machine
downtime or stoppages.
As an opportunity occurs, the concept of component dependency needs a definite factor
that can be used to effectively select the un-failed component related to the failed component
without compromising the system’s operation as well as avoiding malpractice of replacing a

PM Interval

t
Good Zone Opportunistic
Figure 8. 0 f T
Zone
Opportunistic zone in (failure)
OM policy
Source: Koochaki et al.

CM with PM Planned PM
No PM
Infinite
Figure 9.
Opportunistic 0 f x component T
maintenance (OM) failed
policy
Source: Pham and Wang (2000)
component before its life limit. Another reason for the use of a component proximity Opportunistic
measure is because, inconsistent with the rapid advancement of technology, maintenance maintenance
has become more and more complex because manufacturing system consists of more and policy
more processes and machines.
Machines are also built with many components and some depend on one or another. The
interaction between the components makes the optimization of maintenance to be very
difficult. According to Ozekici (1988), the lifetime of a component generally has the 145
characteristic of stochastic dependency on another component in the same system/
equipment because they all operate in similar environmental conditions like temperature,
humidity and vibrations. The relation between the components here can be defined as
physical dependency where:
(1) the next component should not be far from the first one (physical proximity),
contained in a space of interest; and
(2) it should be maintained with the available technical and human resources (technical
proximity) used for the first maintenance action within the time left.
These dependencies make practical optimization of a multi-component system an impossible
task quantitatively to be achieved. However, optimal maintenance is not a simple
juxtaposition of optimal strategies of the components (Castanier et al., 2005). For the OPTOMS
model, the focus is on the physical/structural dependency between the components. In
practice, the decision of component dependency and proximity can be decided based on PM
checklist and failure analysis available in the company. PM activities are usually divided
according to component and system function. Therefore, the relation between failed
components and other un-failed components in the same machine is needed to be identified to
achieve optimal maintenance. This step is about rearranging the failures identified in the
previous step based on component’s proximity and closeness.
A machine consists of various components with different working function and purpose.
Sometimes a machine breakdown occurs because of one component failure. Therefore,
based on the information from the machine operating system, manual book and components
list, the first step in Phase IV is to list out each component function of the critical machine.
Machine function in this identification process is defined as the task assigned to a
component of the critical machine to accomplish a specific process. This step is aimed to
simplify and focus the analysis on components that experienced most failures as identified
in the previous phase. This way, direct and accurate maintenance solution can be planned
for the critical machine based on the functionality of its components.
The process can be done by first constructing functional block diagram (FBD) of the
critical machine. FBD is a simple, yet powerful tool to structure and order a complex
machine in its functional term. In general, FBD is constructed to diagrammatically show the
breakdown of the machine into components that are required to achieve successful
operation. A basic structure of FBD is depicted in Figure 10, whereas the terms used in the
FBD are defined in Table III (Benjamin and Fabrycky, 2006).
Several sets of FBD can be drawn for the selected critical machine. The relations between
components can be based on failure analysis documented in the FMEA method conducted
in the previous phase. These FBD will be referred when a component experiences a failure.
When an opportunity arises, maintenance personnel need to identify the failed component
and the FBD that contains the component. Then, maintenance activities can be conducted to
all the components in the selected FBD. The other component can be decided based on the
indentures in the FBD drawn. During CM on the failed component, PM checklists of all
components involved will be checked and PM conducted. The maintenance activities can be
conducted while paying attention to OM rule, assumptions and limitations set in the next
step of this Phase IV.
JQME Boundary
26,1
Function block
Numbering

146
1.0 2.0

Figure 10. Input Output

Basic functional block
diagram (FBD)
Source: Benjamin and Fabrycky (2006)

Term Definition

Function block The item is presented in single box enclosed by a solid line. The item in the function block
may include machine functions or components which depend on the FBD indenture level
Numbering A uniform numbering system is developed for the breakdown order of the functional
system. It is essential to provide traceability through each level of indenture. Functions
that are identified in FBD at each level should be numbered in a manner which preserves
the continuity of functions and provides information with respect to the function origin
throughout the system. The top level FBD should be numbered by sequence, 1.0, 2.0, 3.0,
etc. Further indenture of these top level FBDs is contained by the same parent whose are
identified and coded at the next decimal level for each indenture. The first indenture of the
machine functional will be numbered as 1.0 (machine functional A), 2.0 (machine functional
B) and 3.0 (machine functional C). The second indenture level is the decomposition from
machine functional A which has been numbered as 1.1 (Sub-function X), 1.2 (Sub-function
Y) and 1.3 (Sub-function Z)
Input and output The input and output of the system in FBD are presented in single box enclosed by a
dashed line
Table III. Flow connection Lines connecting function block indicate the flow of the input and output
Terms and definitions Boundary Border line between the systems functional is represented in single box enclosed by a solid line
in FBD Source: Benjamin and Fabrycky (2006)

7.2 Step 2: opportunities principle decision

As discussed earlier, production lines in the industry are commonly established with a
number of machines to complete a process or product. This means there will be a multitude
of different maintenance activities needed to be conducted each day. Also, failures or
machine downtime that occurs randomly means there is a need for a decision support
system with instructions and guidelines in conducting maintenance activities. For that,
Step 2 of Phase IV has the objective to list out assumptions and limitations that can be used
when using the OM principle and component proximity elaborated in the previous step.
7.2.1 OM policy assumption. There is a vast gap between the theories of OM policy with
the practice of the policy in the real industrial environment. Therefore, assumptions for the
OPTOMS model need to be set. In a publication by Zhou et al. (2009), Ding and Tian (2012)
and Zhou et al. (2015), PM is argued as an imperfect maintenance when in practice. The
statement was backed up with the fact that PM activities like a component replacement will
not restore the whole system to as-good-as-new condition because there will be some Opportunistic
cumulative wear on adjacent components. This fact is in coloration with the component maintenance
dependency concept used in OM policy. policy
This step is developed to minimize the gap between theories of OM and the practice in
the industry. Assumptions in this model are useful in providing the basis for action and in
mimicking real-life situations with “what if” scenarios. Assumptions make it easier to
simulate and create situations like in real companies. Throughout the whole conduct of this 147
model development, a few assumptions needed to be followed are:
(1) Maintenance planning horizon is infinite.
(2) The component state is binary (either operating or failed).
(3) No constraints from production or maintenance department for when maintenance
can be performed.
(4) Any maintenance activities conducted is perfect maintenance and machine in the
condition of as-good-as-new.
(5) Maintenance activities start immediately and without any delay when it occurs;
therefore, all required resources are made to be available.
(6) Spare part cost is neglected for component replacement. This is for the parts that are
actually planned to be replaced in the near future when a breakdown occurs.
Minor parts like screw and nuts are considered in storage; therefore, no additional
cost are involved.
(7) In terms of maintenance cost, replacing more than component at the same time is
cheaper than replacing them separately.
7.2.2 OM policy limitation. Aside from assumptions, another element in the implementation
of OM policy is to set limitations in accordance with a practical point of view. In order to
effectively utilize the concept and principle of OM policy, a maintenance decision should
follow certain control-limit threshold or rule. Therefore, some limitations need to be
observed in order to have the optimal result from the application of the policy. Limitations
for the application of OPTOMS model is suggested as the act to allow only specific
situations to be considered when applying OM policy. The limitations set in this model are
based on the observation of the case study company as well as various research works
published in this area. The limits set will be corresponding with OM principle, rule and
assumptions set earlier.
The limitations are:
(1) PM is conducted on an un-failed component that shares the same or similar failure
mode/effect with the failed component;
(2) PM on an unbroken component involves cleaning, lubricating, tightening and repair
tasks (but not replacement task);
(3) PM tasks involved in OM policy are only limited to daily, weekly and monthly
plans; and
(4) PM tasks scheduled in quarterly and yearly basis are excluded from OM application
and the tasks are to be conducted as planned.
For explanation, the scope of PM tasks in this research is limited to include PM performed
daily, weekly, or monthly. The rationale for the limited number (3) and (4) are because
quarter and yearly PM tasks are for an overhaul, replacement of a component, a calibration
that is major maintenance activities. In some cases, these tasks are subcontracted to a
JQME specialist or the manufacturer of the machine due to the complexity and high level of
26,1 maintenance’s requirement. Overhauls, for example, require a long duration of machine
stoppages; therefore, it is considered as too large of a task for opportunities (Dekker and van
Rijn, 2003). Therefore, it is important to stick to the plan and exclude the task when
implementing OM policy.
The information obtained from this phase will be the input for the application of OM
148 policy in achieving the optimal maintenance system. The next phase of OPTOMS model will
now focus on how this information is used to achieve the optimal maintenance system based
on maintenance objectives set in the first phase.

8. Phase V: measurement and control of optimal maintenance system (OMS)

Maintenance performance is defined as the state or condition of the action or process in
conducting maintenance function when measured from time to time. Levels of maintenance
effectiveness toward manufacturing operation illustrate the performance, and it is necessary
to establish appropriate metrics for the purpose of measuring the maintenance performance
(Chan et al., 2005). Suitable and effective maintenance performance measurement is crucial
to monitor the maintenance activities and the planning for more successful improvement.
And, maintenance tasks should be conducted effectively to reduce downtime and increase
machine availability without compromising its reliability and at the lowest cost possible.
In this research, the OPTOMS model is a five-phase decision support system to achieve
an optimal maintenance system using the OM policy. After conducting Phase I–IV, the final
phase in the OPTOMS model is to measure the performance of the maintenance system to
see whether it achieved optimal maintenance system or not. Optimal maintenance actions
for a component depend on the states of the other components (Pham and Wang, 2000). For
an optimal system, its functionality should be maintained with balanced trade-offs between
all its performance features or elements that relate to other components and its operation.
However, what defines an optimal maintenance system definitely varies from one
company to another. It even varies from one machine to another. Not to mention how each
performance measure ranges and limits are not absolute for all. Due to the uncertainty and
probability that can be achieved when implementing OM policy in the industry, control
chart is selected for the OPTOMS model for its function in showing process changes with a
limit line that is effective in monitoring maintenance performance over time. Instead of
giving a definite value for performance of an optimal maintenance system, a control chart
allows for variation of acceptable performance value.
The usage of control chart is recognized in maintenance management because of their
common goal in achieving optimal product quality, with a reduction of downtime and cost
by controlling process variations (Zhou and Zhu, 2008). Control chart is used to efficiently
analyze performance data to evaluate a process. It is not only a statistical process control
mainly used for quality analysis, but it also has many other functions to identify and
understand process variation (Al-Oraini and Rahim, 2002). For example, control chart can be
used in the manufacturing process to test if machinery is producing products within
specifications. Three major functions of control chart are to analyze process variables, to
determine process capabilities and finally to monitor the effects of analyzed variables on the
difference between actual performances with a targeted value.
Drawn like a line chart, control chart has the distinctive feature in the form of three
straight-horizontal lines between its plotted data. The three lines represent a central line for
the average value, an upper control line for maximum accepted value and lower control limit
for a minimum value of the acceptable value in a process. These lines are determined from
historical data for the analyzed system. By comparing current data to these lines,
observation can be done to see whether the variations are consistent (in control) or are
unpredictable (out of control). Shewhart Charts or commonly known as control chart will be
used for the performance measurement of maintenance system tool in this phase Opportunistic
(Panagiotidou and Nenes, 2009). The full process flow is shown in Figure 11. The next maintenance
subsections in this paper will explore step-by-step approach for this tool. policy
8.1 Step 1: data collection based on maintenance objective
The first step of Phase V is to collect performance data for the selected critical machine
observed. For this step, performance measure needs to be consistent with maintenance 149
objective selected based on the company’s mission and vision in Phase I of the OPTOMS
model. Analysis of maintenance system performance can be divided into three parameters
as listed in Figure 12. The relation between maintenance objectives and performance factors
measured will be further explained in the next three sub-subsections.

Step 1

• Data Collection Based on Maintenance Objective

– Objective A: Increase Machine Availability
– Objective B: Increase Machine Reliability
– Objective C: Minimize Maintenance Cost

Step 2

• Means of Variance Calculation

Step 3

• Standard Deviation, Upper Control Limit (UCL) and

Lower Control Limit (LCL) Setting

Step 4

• X-bar of Control Chart Construction

Figure 11.
Step 5 Steps in developing
control chart for
• Performance of Maintenance System Observation maintenance system’s
– Within control limit performance
– Over control limit measurement

• Maintenance Objective 1: Machine

Failure Rate Availability

• Maintenance Objective 2: Machine

Age/Machine Operation Reliability
Figure 12.
Maintenance
objectives and its
• Maintenance Objective 3: Maintenance performance
Cost Cost measurement factor
JQME 8.1.1 Analysis of objective 1: system availability. The first option of maintenance objective
26,1 suggested in Phase I of OPTOMS model is the failure rate of a machine. Failure rate
represents the number of failures that happened on the identified machine. In a
manufacturing plant, unavailable time (the complement of available time) is commonly
termed as machine downtime. The data can be collected from Production Data that
records production scheduling and operation. Planned production time (Tplan) is the total
150 time a machine supposed to be available to produce a product. Therefore, the planned
production time for the selected machine can be calculated by multiplying the days of work
in a month with total minutes the machine expected to be operating in a day as in the
following equation:

 X1
Planned production time T plan ¼ d  t0: (1)
t¼1

Here, δ is the number of working days. This can be determined on the weekly, monthly or
yearly basis, whichever preferred during the data analysis process. The recommended time
frame would be on a monthly basis. ť is the daily production time planned for the machine to
operate and is converted into minutes. Actual production time (Tact) is calculated by
eliminating downtime losses such as machine failures and setup and adjustments.
This is basically the time planned to have an available machine. It is well known that the
duration of Tplan is the maximum time of machine operation, and yet, it is rarely achieved
because of both of unplanned and planned downtime. Therefore, Tact is used. Calculation for
Tact is as follows:
X
1

Actual production time ðT act Þ ¼ T plan  T updt þT pdt : (2)
t¼1

Tupdt in Equation (2) symbolizes the duration of unplanned downtime that occurs during the
whole Tplan. The example of Tupdt that usually happen is, when machine experience failures
or breakdowns. Tpdt is the duration of downtime planned on the machine for maintenance
actions or breaks such as:
• implementation of PM or routine check-ups and calibrations of the machine;
• machine trials and process improvement activities;
• machine stoppages for a change of components to produce a different product; and
• machine stoppages for software installation.
The mathematical model for availability calculation is as follows:

Actual Production TimeðT act Þ

Machine Availability ¼
  100: (3)
Planned production Time T plan

8.1.2 Analysis of Objective 2: machine reliability. The second maintenance objective as listed
in Figure 11 is related to the age of component or machine. When a component fails, it is
considered as the end of usage or lifetime. It is universally known that all products and
systems can be unreliable because they degrade over time and experience failure. It can be
expressed that an increase of deterioration in a component or machine is the same term as
the decrease of the system’s value.
 Therefore, for the OPTOMS model, the optimal performance measures for the second Opportunistic
maintenance objective are the maximization of machine reliability in operation. maintenance
Reliability can be defined as the ability of a product or machine to perform a specified policy
function, under given conditions and for a given period of time without experiencing any
failure (Thomas, 1986).
With known time-to-failure distribution function and component lifetime, maintenance
should be planned to increase machine reliability and maximize its useful life. Since the 151
effect of maintenance is to increase the mean time to failure, when less failure is reported,
any maintenance conducted is considered suitable and effective. According to Pande et al.
(2002), the biggest dilemma with multiple component systems is how to calculate the
system’s reliability with the probability of every component is operational. It assuredly will
involve complex computations to satisfy the variation in order to find an optimal
maintenance solution. Correspondingly, the reliability calculation as maintenance
performance measure should be focused on one critical component or set of components
identified using FBD in the previous phase. By applying the OM policy, maintenance
interval and duration of the critical component will reflect the machine’s reliability and these
can be controlled and improved.
8.1.3 Analysis of Objective 3: maintenance cost. The third performance measure to
analyze the performance of OM policy is in regard to money or cost. The changes in a
company’s profits are the reflection of the changes in productivity and price recovery of that
company. These changes are, however, are determined by the effectiveness and efficiency of
the production process. Indirectly, a productive maintenance will ensure a smooth production
process that will lead to an increase in profits. With a company that focuses mainly on
productivity, the maintenance system is not paying as much attention as to the production
activity. Most of the resources are allocated to the production line than to the maintenance
system. The perception is that maintenance does not contribute to the company’s profitability.
However, costing in the industry is not under the responsibility of the maintenance
department as it involves other departments in the company. Therefore, for the purpose of
calculating the maintenance cost of an optimal maintenance system, two direct costing
factors can be used. The first one is a cost for preventive tasks, and the second one is the
cost of corrective tasks. The concept used for minimization of maintenance cost is conducted
PM tasks of un-failed component whenever a system stopped due to a component failure.
The optimal maintenance system should result in a reduction in a number of CM task
conducted and directly reduce overall maintenance cost.
The approach for calculating maintenance cost to measure the optimality of OM policy
conducted is by multiplying the number of maintenance task conducted per week on the
selected machine, as shown in the following equation:


Maintenance Cost ¼ np  C p þ ðnc  C c Þ; (4)

where np is the number of PM task conducted, nc the number of CM task conducted, Cp the
cost of PM task conducted and Cc the cost of CM task conducted.
As maintenance varies in many elements, the value for Cp and Cc are to be decided by
maintenance personnel responsible for maintenance planning. Focus in this maintenance
objective is reduction in maintenance cost, reducing the number of maintenance activities
when applying OM policy.

8.2 Step 2: means of variance calculation

The main features of control charts include data points, a center line, upper line and lower
line. The second step to create a control chart for performance measure in the OPTOMS
JQME model is to calculate centreline that is the means of the variance of the data collected in the
26,1 previous step. It is the average of the performance value. To find the mean value, all
measurement data collected earlier need to be added up and then divided by the number of
measurements conducted in the process. The equation to be used is as follows:
X 1 þX 2 þX 3 þ    þX i
Means of variance; m ¼ ; (5)
n
152
where Xi is the individual measurements and n the number of measurements conducted.

8.3 Step 3: standard deviation, upper control limit (UCL) and lower control limit (LCL) setting
All processes have some kind of variation, and maintenance system also has a variation that
may be caused by natural process variation like a different type of component, material,
activities, and duration of maintenance conducted. There might also be special cause
variation, generally caused by some extraordinary occurrence in the system. Therefore, it is
not practical to base any performance level on a certain value only. The next step in Phase V
is to set the standard deviation value based on the mean value calculated in the previous
step; then the standard deviation to set upper and LCL for the control chart is used.
Standard deviation (σ) is explained as a statistical way to describe how much variation
exists in a set of data (Alexander et al., 1995). The equation to calculate σ is as follows:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u n
u1 X
Standard deviationðsÞ ¼ t ðxi mÞ2 ; (6)
n i¼1

where Xi is the individual measurements, n is the number of measurements conducted, and µ

is the means of variance.
In a state of statistical control, UCL and LCL indicate the range of performance value that
are acceptable during maintenance system. UCL and LCL define the constraints of natural
cause variations. All these limits and line will help to detect the trend of all plotted value
during analysis. And, where to put these limit lines will determine the risk of undertaking
the improvement process with a variation of performance measured. In general, both UCL
and LCL can be set using the following equations:
UCL ¼ mþks; (7)

LCL ¼ mks; (8)

where μ is the mean of the variable, σ the standard deviation of the variable, and k
the distance of the control limits from the center line, expressed in terms of standard
deviation units.
The symbol k in the Equations (7) and (8) represents the distance of UCL and LCL from
central line or means of variance. The distance indicates the allowable variation of the
measured maintenance system. As one of tools widely used in Six Sigma research, the k
value can be in the value of 1 until six. Historically, k ¼ 3 has become an accepted standard
in the industry (Ozekici, 1988). Therefore the control limits are set at a distance of 3-sigma
above and below the means of variance. However, the decision of k’s value can also be up to
maintenance engineer or any personnel using the OPTOMS model.

8.4 Step 4: X-bar of control chart construction

With all available data collected and calculated, Step 4 is where a chart is constructed.
Microsoft Excel software can be used to create a simple control chart. Control chart comes
in many types; each differs according to its usage. For the OPTOMS model, X-bar or also Opportunistic
called averages chart will be used because it is a variable control chart applied to data maintenance
with continuous distribution over time. X-bar is the most commonly used type of control policy
chart for controlling and determining the state of situation measured (Ben-Daya and
Rahim, 2000). In completing Step 4, scales and plot the control limits, centreline and data
points, in each plotting area need to be selected. The scales must be determined before
the data points and centreline can be plotted. Once the UCL and LCL have been computed, 153
the easiest way to select the scales is to have the current data take up approximately
60 percent of the vertical (Y ) axis. The scales for both the upper and lower plotting
areas should allow for future high or low out of control data points. Figure 13 shows the
example of a control chart that created using Microsoft Excel software.

8.5 Step 5: performance of maintenance system observation

Now with the created control chart based on historical data, it is important to understand
how to interpret it. Therefore, the final step at the end of OPTOMS model is to observe the
current maintenance system using control chart plotted in previous steps. From time to
time, maintenance performance needs be calculated and plotted as a new data point, and
then it needs to be checked for out-of-control signal. If the data points plotted are within the
control limits, then variations of the value calculated are acceptable and application of OM
policy in the system is considered in control.
Otherwise, if the data points fall outside the control limits set, then optimization is
considered failed because the process is out of the acceptable value of performance
measures. For this type of situation, the application of OM policy needed to be stopped and
re-checked. The causes of imbalance should be analyzed in order to have an optimal
maintenance system. Improvement can be done to eliminate or minimize the out-of-control
value by referring back to FMEA and maintenance checklist in Phases III and IV of the
OPTOMS model and check for any misconduct of maintenance activities.
Another usage of a control chart is it can be used to predict the expected range of
outcomes from the application of OM policy toward an optimal maintenance system. The
chart will aid a company, especially the maintenance department to gauge their
maintenance system. The result from the measurement process can then be used to plan
improvement activities and to provide better maintenance activities in the future. Therefore,
maintenance performance can be in its effective level as possible with continuous
monitoring and improvement.

Averages
Lower control limit
Upper control limit
Center line
X-bar Chart
80
70
60
Averages

50
40
30
20
10 Figure 13.
Example of an X-bar
0 chart in the
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
control chart
Sample number
JQME 9. Model validation in a case study company
26,1 In order to use the concept and principle of OM policy in the industry, the best research
method is to conduct a series of a case study in the industry as to test the developed model in
real situations. A case study is a useful method to test theoretical models by using it in
real-world situations. Operating as a subcontracting company, Company A offers different
packages of processes for product manufacturing based on customer demand and
154 specifications (packages). The company comprises a full range of support and services from
circuit design, prototype fabrication, and mass production up to board assembly. Most of
the products produced in this company are in the form of a panel and can be commonly
found in electronic products like car headlamps and taillight and antenna. The company’s
mission and vision are to focus on productivity as demand is high for their product.
Then, as suggested in Phase II of OPTOMS model, Figure 14 shows total unscheduled
machine downtime at a Drilling Department. Here, data from three major machines that
share the same process specification and machine capability were analyzed for 12 months.
Meanwhile, Table IV contains the basic PM checklist that was used to maintain the
machines by a week or by month depending on component’s specification. There is also
scheduled downtime of PM activities drawn in Figure 15. All this information was collected
as suggested in Phase II of OPTOMS model.
Subsequently, Phase III of failure analysis using FMEA was validated and resulting of
information in Table V. The worksheet of failures was filled based on information in failure
analysis data as well as verbal input by technician and operators. One point to take note is
that failure components and functions listed in the table are the same for the three machines
(Hitachi Nos. 1–3) in the Drilling Department. Table VI and Figure 16 contain details of
unscheduled downtime of the analyzed machines from January until December. As
suggested in Phase IV of OPTOMS model, the regularity and occurrences of the failures can
be properly observed in Figure 17. Further application of OPTOMS model using the
gathered information will be presented in the next section of this paper.

9.1 Maintenance performance measurement

The final stage of the conducted validation process that focused on a single component
system is to measure the performance of the maintenance system when applied to the OM
policy using the OPTOMS model framework. As mentioned, the maintenance objective is to
increase machine reliability to ensure product demands can be achieved as scheduled. The
ideal capability and lifespan of a reliable product is one that did not experience any failure
during operation and been used until the end of its lifetime. Age or reliability of the
component as optimal criteria in OM policy research works is widely covered and results of
many mathematical models. For example, a lot of methods like Weibull analysis (Cassady
et al., 2001), Poisson Distribution Function (Dekker and Dijkstra, 1992), Genetic Algorithm

Machine Donwtime at Drilling Department

Hitachi No. 3 12.3

Machine

Hitachi No. 2 114.33

Hitachi No. 1 21.72

Figure 14.
Total machine Mega 16
downtime at the
drilling department 0 20 40 60 80 100 120 140
for 12 months’ period
Total Unscheduled Downtime
 Opportunistic
maintenance
policy

155

Table IV.
A sample of PM
checklist for Hitachi
Nos. 1–3

(Savic et al., 1995) and Markov Chain (Amari and McLaughlin, 2004) were used when
discussing the system’s lifetime and age.
In the similar area of research, historical data on a machine like failure time and
maintenance cost to develop reliability analysis based on the Weibull distribution were used
(Metwalli et al., 1998). Weibull parameters were analytically achieved to determine the
reliability and hazard functions for each component and system in the machine. The plotted
data were then analyzed and used for an optimization technique for effective maintenance
planning. Using the probabilistic representation of a deterioration process on a machine
through discrete stages, a conceptual model called Asset Management Planner (AMP) and a
mathematical model were proposed to relate maintenance tasks to machine reliability
(Endrenyi and Anders, 2006). The models described the impact on the reliability of the
gradually deteriorating machine of periodic inspection, which can lead to various possible
maintenance policies.
JQME 3/1
WW1
10/1
WW2
17/1
WW3
24/1
WW4
31/1
M1
Legend

26,1 15/10
Weeky PM
15/10 31/12
Monthly PM Yearly PM

1/1 1/2
156
Figure 15. 31/1 28/2 31/3 30/4 31/5 30/6 31/7 31/8 30/9 31/10 30/11 31/12
Scheduled downtime M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 31/12
Yearly
for weekly, monthly
and yearly PM
schedules for Hitachi
machine
1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 1/9 1/10 1/11 1/12 31/12

Research by Wiksten and Johansson (2006) discussed that failure function that is the basic
measurement of reliability will increase if improper maintenance is conducted on the
machine. The calculation used is on mean time between maintenance, mean time between
overhaul, maintenance-free operating period, mean time between critical failure and mean
time between unscheduled removal. Looking at all the previous research works discussed, it
can be understood that the main factors in reliability principle are the failure time and
failure duration. And, one principle that uses both pieces of information is the calculation of
mean time between failures (MTBF). This is commonly used in the industry with the
equation of:
Machine Downtime
Mean time between failure ðMTBFÞ ¼ : (9)
Number of failures
However, reliability is not equal to MTBF. The value of machine reliability is the inverse of
MTBF or 1 minus the MTBF value. The equation is as in Equation (10). To ease
understanding and performance measure, the reliability value is presented in percentage
(%). Table VII contains the information of failures of Pneumatic System for Hitachi No. 1
machine. The failures are extracted from Table VII. Only seven out of 23 failures happen at
the Pneumatic System of Hitachi No. 1.
From the information in Table VII, machine reliability at each time of failure can be
calculated as the benchmark or performance measurement of maintenance system conducted
on the machine. Then, to draw a control chart, means of variance and standard deviations are
calculated as in Equations (11) and (12):
Machine Reliability ðf 8Þ ¼ ð1MTBFÞ  100 ¼ ð10:0471Þ  100 ¼ 95:29%; (10)

X 1 þX 2 þX 3 þ    þX i
Means of variance; m ¼ ¼ 79:08142857; (11)
n
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u n
u1 X
Standard deviation ðsÞ ¼ t ðxi mÞ2 ¼ 26:0256799; (12)
n i¼1

with the K value set as 1, the UCL and LCL are:

UCL ¼ m þks ¼ 79:08142857 þ ð1  26:0256799Þ ¼ 105:1071085; (13)
 Opportunistic
maintenance
policy

157

Table V.
FMEA worksheet
for failures of the
HMS machine

LCL ¼ m þks ¼ 79:08142857ð1  26:0256799Þ ¼ 53:05574867: (14)

The control chart for machine reliability is as in Figure 17.

Based on the values of LCL and UCL in the control chart, the maintenance system on
the Hitachi machines should be improved with the application of OM policy and the
performance measure to achieve 100 percent reliability. Taking the failure number f14 as
an example, system reliability due to the failure of Pneumatic system drop to 21.43 percent
because the failure takes a lot of time to be recovered. The machine was stopped for 5.5 h
compared to an average of half an hour for another failure in the same system. Moreover,
the same failure occurs again after two weeks as recorded in f17. This shows that
maintenance activities conducted on the machine were ineffective and create high
JQME Total
26,1 Failure downtime
No. Date Machine Failure types Related components (hr)

1. January 21 Hitachi No. 2 Spindle problem Spindle 72.0

2. February 5 Hitachi No. 1 Spindle problem Spindle 1.5
3. June 22 Hitachi No. 1 Spindle problem Spindle 1.0
158 4. July 22 Hitachi No. 2 Servo out of control Motor 9.0
5. July 24 Hitachi No. 2 Vacuum problem Vacuum 21.0
6. July 25 Hitachi No. 2 Pressure foot user switch error Belt 4.5
7. October 3 Hitachi No. 1 Spindle problem Spindle 4.0
8. October 4 Hitachi No. 1 Low air pressure Pneumatic system 0.33
9. Hitachi No. 2 Low air pressure Pneumatic system 0.33
10. Hitachi No. 3 Low air pressure Pneumatic system 0.33
11. Hitachi No. 2 Collect problem Spindle collect 2.0
12. October 7 Hitachi No. 1 Low air pressure Pneumatic system 0.83
13. Hitachi No. 3 Push alarm error Alarm 5.0
14. October 14 Hitachi No. 1 Low air pressure and spindle off Pneumatic system 5.5
and spindle
15. Hitachi No. 2 Low air pressure and spindle off Pneumatic system 5.5
and spindle
16. Hitachi No. 3 Low air pressure Pneumatic system 5.5
17. October 30 Hitachi No. 1 Low air pressure Pneumatic system 1.0
18. November 2 Hitachi No. 1 Low air pressure Pneumatic system 0.42
19. November 15 Hitachi No. 1 Spindle problem Spindle 2.0
Table VI. 20. November 16 Hitachi No. 1 Low air pressure Pneumatic System 0.67
Unscheduled 21. November 23 Hitachi No. 1 Spindle problem Change spindle 5 3.0
downtime for HMS 22. November 28 Hitachi No. 1 Low air pressure Pneumatic System 1.5
(Hitachi Nos. 1–3) 23. December 28 Hitachi No. 3 Low air pressure Pneumatic system 1.5
machines for a year Total downtime 148.41

Legend
15/11
20/1
f19
Machine Failure, f
25/7
2/11
f6 16/11
f18
f20
7/10
f12 and f13
23/11
14/10 f21
24/7 f14, f15 and
f5 4/10 f16
f8, f9, f10 and f11 28/11
21/1 30/10 f22
5/2
Figure 16. f1 f2
22/6 22/7
3/10 f17 28/12
f3 f4 f23
Unscheduled machine f7
downtime at Hitachi
Nos 1–3 for January
until December
1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 1/9 1/10 1/11 1/12 31/12

unscheduled machine downtime. For improvement, maintenance system should be

applied to OM policy by following the principle, assumptions and limitations suggested in
the OPTOMS model.

10. Discussion
In proving the viability of the OPTOMS model in the real industrial application, a validation
process is conducted in a case study company with a multi-component system to test the
component dependency and proximity principles in the OM policy. The maintenance policy
 Control Chart for Reliability at Hitachi No. 1 Machine Opportunistic
120
maintenance
100
85.71
94 90.43 policy
95.29 88.14
80 78.57
Reliability

60
159
40

20 21.43

0 Figure 17.
f8 f12 f14 f17 f18 f 20 f 22 Control chart of the
Month reliability of Hitachi
Reliability Mean (CL) No. 1 machine when
failures occur
Upper Control Limit (UCL) Lower Control Limit (LCL)

No. Failure of pneumatic system Machine downtime (hrs) MTBF Machine reliability (%)

1. f8 0.33 0.0471 95.29

2. f12 0.83 0.1186 88.14
3. f14 5.5 0.7857 21.43
4. f17 1.0 0.1429 85.71 Table VII.
5. f18 0.42 0.0600 94.00 Values of MTBF and
6. f20 0.67 0.0957 90.43 reliability at Hitachi
7. f22 1.5 0.2143 78.57 No. 1 machine

measured in this process is related to the machine’s availability by reducing machine failure
rate. The validation of the OPTOMS model in the multi-component system is hoped to
mirror the real situation in the industry. It is worth to be noted that OM applications
described in the literature are based on a single component or a few components. Yet, the
real industry uses equipment with the multi-component and multi-unit system. Because
more maintenance activities plan for more components in a system, and then, more OM
activities can be applied when opportunities arise.
For this validation, a case study is conducted in Cluster 1: circuit formation. The
maintenance objective in focus is the maximization of machine availability during operating
hours. The direct solution is by reducing scheduled downtime frequency by conducting it
together with CM as suggested in OM policy. Based on the information of machine
downtime, the critical machine selected is the HMS machine. With 54 weekly PM, 12
monthly PM and one yearly PM including daily inspection and cleaning, it is an
overwhelming maintenance schedule of the machine. However, the abundance of PM
activities is ineffective as the machine still has a high failure rate. A total of 17 unscheduled
downtimes recorded on the machine hinder production to meet the customer demand.
In proving the robustness and applicability of the OPTOMS model, the model is developed
in five phases with each phase consisting of additional steps and tools. Phase 1: identification
of maintenance objective based on the company’s mission and vision suggests that mission
and vision statements of a company are analyzed to find the aim and future plan of
the company so that maintenance system can be planned and improved to help the
company achieving the ideal. Check sheet with a list of performance measure elements for
three different maintenance objectives is used to tally with the company’s mission and vision.
JQME From the case study, it is found that the company can achieve success with their maintenance
26,1 system by focusing improvement effort to reduce the machine’s failure rate so that more
quality product can be manufactured.
The relation between Phases II and III is in the data collection and analysis. Results from
Phase II are the input for analysis in Phase III. In the case study, first unscheduled and
scheduled machine downtimes for one of the clusters in the case study company are
160 calculated. Based on the information, a stacked-bar chart is drawn and machine with the
highest downtime is chosen as a critical machine. The selection of a critical system is as a
step to prioritize improvement system to one machine among many machines and
equipment in the company. In the phase, FMEA is also carried out to find the cause, effects
and maintenance actions conducted whenever a failure occurs. The usage of FMEA as an
analysis tool is practical in the industry and can provide the information required for the
improvement of the maintenance system.
Succeeding is Phase IV: OM principle and rule selection. At this point, in order to apply
dependency, opportunity and component proximity principles in OM policy, FBD of critical
components in the OKUNO machine is drawn. The usage of FBD suggested in the OPTOMS
model is to connect the components in a system with one another based on the failure type it
experienced. By having an FBD, maintenance activities can be planned to the failed
component as well as operating yet related component in the same system. This way,
machine downtime can be reduced both in frequency and total duration. The FBD is also
helpful to avoid a common problem in OM policy which is over-maintenance and wasting a
useful component lifetime. Besides that, Phase IV is also focused on the assumptions and
limitations suggested in the OPTOMS model by observing the maintenance system in the
case study company and compare the situations.
Finally, the Phase V of the OPTOMS model is an observation on the performance
of the maintenance system. Here, control chart is introduced. The chart is used to
determine whether a process is no state of control or not. The chart provides information
for decision making with regards to acceptable values for the performance of the
maintenance system. The usage of control chart also provides a graphical description to
evaluate a given process, whether it is in a state of statistical control or is out of control.
Any maintenance conducted under OM policy that been analyzed using the control chart
and is outside the upper or lower control limits warrants an immediate investigation to
determine the cause of the tasks ineffective. Then, appropriate corrective measures
can be planned to maximize the positive traits from the OM policy application. When an
opportunity occurs, the failure is to be analyzed and maintenance activities have to be
decided. Then, the performance is measured on the given machine by using the
historical data calculated and drawn as a control chart. The chart aids in decision making
in the maintenance system. The control chart is suggested in maintenance work to
control the effectiveness of the work performed on the road to achieve the optimal
maintenance system.
Based on the FBD and maintenance planned of the OM activities, the schedule can be
simplified as the time spent for the maintenance work that can be reduced and a proper
maintenance management can be enabled to facilitate the problem occurring in the
manufacturing facilities. In this process, multiple components in one machine will be
observed to find the opportunity or machine breakdown as a time to conduct PM tasks.
And, the measurement factor to be used is how long a machine is available to work when
applying the OM policy. The results from this validation show the reduction by 20 percent
of the number of machine stoppages for both scheduled and unscheduled downtime by
combining the maintenance activities conducted whenever the opportunity arises.
Calculations on the machine availability proved that with the suggested improvement in
the maintenance activity, much time could be saved on the time spent on maintenance
work, thus allowing technicians to focus the time on doing thorough maintenance on the Opportunistic
critical machines. maintenance
For the OM policy discussed in this research, OM is planning and scheduling of an policy
optimal maintenance system that prospectively conducted PM activities on dependent/
related components when a component failed with the consideration of lowest maintenance
cost possible and without sacrificing its reliability. OM can be practiced to reduce the
number of machine breakdowns and downtime especially for the continuous system. From a 161
practical point of view, the competitive industry and fast-changing technology make the
operating systems in a company to become more and more complex. Therefore, the
prospective approach in OM can be widely studied and applied in the real industry as it is an
effective and optimal maintenance policy. The savings that can be achieved from
application of principle and concept in OM policy are in the form of:
• less number of production disruption for maintenance;
• lower failure rate as component with high failure potential is replaced or repaired
before it failed;
• lower maintenance cost because less manpower is needed to conduct maintenance
activities; and
• higher system or machine availability and reliability as components are replaced
before it experience any major failure.
The most important thing to be addressed by the OPTOMS model is the practicality of OM
policy in the industry. The development of the model takes into consideration the common
maintenance policy in the industry and how it was implemented (Colledani et al., 2018).
From all the case studies conducted, the benefits that a company could reap from the
implementation of the OPTOMS model are:
• having systematic phases and steps in conducting the prospective OM policy and an
improvement plan in production line according to company’s achievement target and
future goal;
• having a number of factors to choose for maintenance performance analysis and
improvement by looking at the various elements that contribute to failures, for
example from the component’s reliability and machine availability;
• having a simple yet accurate tools in measuring and improving the maintenance
system using the data and information commonly available in the company
especially for complex and multiple equipped system; and
• having a prospective and optimal maintenance approach that will reduce operating
cost and increase production rate.

11. Conclusion
An OPTMS model is developed as a decision support system for the application of OM
policy based on practical theory. As the effort to reduce the gap between theory and
practice, the model can be used in the decision-making process toward the idea of having an
optimal maintenance system. The optimal system can be catered to suit the company
mission and vision. The concept opportunity in the model is the time or the moment at
which PM tasks can be carried out on an un-failed component without having adverse
effects of a unit shutdown being incurred. As discussed, this research provides the detail
analysis of OM as a maintenance policy. OM was found to be originated from ARP and BRP
concepts, and then developed based on both concepts of age-related system performance
JQME with component relation and proximity to decide for maintenance activities and
26,1 opportunities function. However, this concept is not directly addressed in publications
and the study of the evolution both in theory and in real industry application is very much
lacking. Therefore, more research works are needed on the issue so that OM can be more
practical and easy to be implemented.
The study and simulation of OM activities are also needed in finding the optimal trade-
162 off between system availability, reliability and costs. The simulation of maintenance system
is required to reduce the gap between theories and practical implementation. As there are a
lot of numerical analyses published and models introduced in OM research, it should be
simulated to find loopholes and test its practicality. For future research on the term of OM
approach, studies should also be directed to the question of how to choose which group of
component to be replaced or restored each time this policy is applied. Artificial intelligence
methods such the genetic algorithm, Fuzzy Logic and Poisson distribution can be applied to
achieve an optimized OM system. Aside from that, more extensive analysis should be done
based on real data from companies to test the practicality of OM concepts in the industry.
Another point for future work is, the OPTOMS model developed in this research can be
jointly conducted by maintenance, production and marketing department to portray the
numbers and figures in the model into financial figures. With that, the higher management
can visualize and understand the full picture of the situation in the company. Since money
wise is deemed most relevant by the management, future work may involve the quantitative
measures like operating time or cost savings. Furthermore, future work can also involve the
development of computerized maintenance performance measurement model. The manually
developed OPTOMS model can be upgraded to the computerized system that is more
compatible and accurate for industrial application. With the aid of technology, data
collection and analysis would be easier to be conducted.

References
Ab-Samat, H. and Kamaruddin, S. (2014), “Opportunistic maintenance (OM) as a new maintenance
policy: a review”, Journal of Quality in Maintenance Engineering, Vol. 21 No. 2, pp. 98-121.
Alexander, S.M., Dillman, M.A., Usher, J.S. and Damodaran, B. (1995), “Economic design of control
charts using the Taguchi loss function”, Computers and Industrial Engineering, Vol. 28 No. 3,
pp. 671-679.
Al-Oraini, H.A. and Rahim, M.A. (2002), “Economic statistical design of X control charts for systems
with Gamma (λ, 2) in-control times”, Computers and Industrial Engineering, Vol. 43 No. 3,
pp. 645-654.
Al-Turki, U. (2011), “A framework for strategic planning in maintenance”, Journal of Quality in
Maintenance Engineering, Vol. 17 No. 2, pp. 150-162.
Amari, S.V. and McLaughlin, L. (2004), “Optimal design of a condition-based maintenance model”,
Reliability and Maintainability, 2004 Annual Symposium-RAMS, pp. 528-533.
Basri, E.I., Abdul Razak, I.H., Ab-Samat, H. and Kamaruddin, S. (2017), “Preventive maintenance (PM)
planning: a review”, Journal of Quality in Maintenance Engineering, Vol. 23 No. 2, pp. 114-143.
Ben-Daya, M. and Rahim, M.A. (2000), “Effect of maintenance on the economic design of x-control
chart”, European Journal of Operational Research, Vol. 120 No. 1, pp. 131-143.
Benjamin and Fabrycky (2006), System Engineering and Analysis, Prentice Hall, Essex.
Besnard, F., Patriksson, M., Stromberg, A.-B., Wojciechowski, A. and Bertling, L. (2009), “An
optimization framework for opportunistic maintenance of offshore wind power system”, IEEE
Bucharest Power Tech Conference, pp. 1-7.
Bevilacqua, M. and Braglia, M. (2000), “The analytic hierarchy process applied to maintenance strategy
selection”, Reliability Engineering & System Safety, Vol. 70 No. 1, pp. 71-83.
Cassady, C., Murdock, W.P. Jr and Pohl, E.A. (2001), “Selective maintenance for support equipment Opportunistic
involving multiple maintenance actions”, European Journal of Operational Research, Vol. 129 maintenance
No. 2, pp. 252-258.
Castanier, B., Grall, A. and Bérenguer, C. (2005), “A condition-based maintenance policy with
policy
non-periodic inspections for a two-unit series system”, Reliability Engineering & System Safety,
Vol. 87 No. 1, pp. 109-120.
Chan, F.T.S., Lau, H.C.W., Ip, R.W.L., Chan, H.K. and Konga, S. (2005), “Implementation of total
productive maintenance – a case study”, International Journal Production Economics, Vol. 95,
163
pp. 71-94.
Colledani, M., Magnanini, M.C. and Tolio, T. (2018), “Impact of opportunistic maintenance on
manufacturing system performance”, CIRP Annals, Vol. 67 No. 1, pp. 499-502.
Crocker, J. and Kumar, U.D. (2000), “Age-related maintenance versus reliability centred maintenance: a
case study on aero-engines”, Reliability Engineering & System Safety, Vol. 67 No. 2, pp. 113-118.
Cui, L. and Li, H. (2006), “Opportunistic maintenance for multi-component shock models”, Mathematical
Methodology of Operational Research, Vol. 63, pp. 493-511.
Dekker, R. (1996), “Applications of maintenance optimization models: a review and analysis”, Reliability
Engineering & System Safety, Vol. 51, pp. 229-240.
Dekker, R. and Dijkstra, M.C. (1992), “Opportunity-based age replacement: exponentially distributed
times between opportunities”, Naval Research Logistics, Vol. 39, pp. 175-190.
Dekker, R. and Smeltink, E. (1991), “Opportunity-based block replacement”, European Journal of
Operational Research, Vol. 53 No. 1, pp. 46-63.
Dekker, R. and van Rijn, C. (2003), “PROMPT, a decision support system for opportunity-based
preventive maintenance”, available at: http://people.few.eur.nl/rdekker/pdf_files/paper_
PROMPT.pdf (accessed November 27, 2011).
Derigent, W., Thomas, E., Levrat, E. and Iung, B. (2009), “Opportunistic maintenance based on fuzzy
modelling of component proximity”, CIRP Annals - Manufacturing Technology, Vol. 58 No. 1,
pp. 29-32.
Ding, F. and Tian, Z. (2012), “Opportunistic maintenance for wind farms considering multi-level
imperfect maintenance thresholds”, Renewable Energy, Vol. 45, pp. 175-182.
Endrenyi, J. and Anders, G.J. (2006), “Aging, maintenance, and reliability”, IEEE Power and Energy
Magazine, Vol. 4 No. 3, pp. 59-67.
Gary Teng, S., Ho, S.M., Shumar, D. and Liu, P.C. (2006), “Implementing FMEA in a collaborative
supply chain environment”, International Journal of Quality & Reliability Management, Vol. 23
No. 2, pp. 179-196.
Jardine, A.K. and Tsang, A.H. (2006), Maintenance, Replacement, and Reliability: Theory and
Applications, CRC Press, Boca Raton, FL.
Jiang, R. and Ji, P. (2002), “Age replacement policy: a multi-attribute value model”, Reliability
Engineering & System Safety, Vol. 76 No. 3, pp. 311-318.
Kimura, Y. (1997), “Maintenance tribology: its significance and activity in Japan”, WEAR, Vol. 207,
pp. 63-66.
Kmeta, S., Fitch, P. and Ishii, K. (1999), “Advanced failure modes and effects analysis of complex
processes”, Proceedings of the 1999 ASME Design Engineering Technical Conferences,
Las Vegas, Nevada, September 12-15.
McCall, J.J. (1963), “Operating characteristics of opportunistic replacement and inspection policies”,
Management Science, Vol. 10 No. 1, pp. 85-97.
Mann, L. Jr, Saxena, A. and Knapp, G.M. (1995), “Statistical-based or condition-based preventive
maintenance?”, Journal of Quality in Maintenance Engineering, Vol. 1 No. 1, pp. 46-59.
Mechefske, C.K. and Wang, Z. (2001), “Using fuzzy linguistics to select optimum maintenance
and condition monitoring strategies”, Mechanical Systems and Signal Processing, Vol. 17 No. 2,
pp. 305-316.
JQME Metwalli, S.M., Salama, M.S. and Taher, R.A. (1998), “Computer-aided reliability for optimum
26,1 maintenance planning”, Computers & Industrial Engineering, Vol. 35 No. 3, pp. 603-606.
Ozekici, S. (1988), “Optimal periodic replacement of multicomponent reliability system”, Operations
Research, Vol. 36 No. 4, pp. 542-552.
Panagiotidou, S. and Nenes, G. (2009), “An economically designed, integrated quality and maintenance
model using an adaptive Shewhart chart”, Reliability Engineering and System Safety, Vol. 94
164 No. 3, pp. 732-741.
Pande, P.S., Holpp, L. and Pande, P. (2002), What is Six Sigma?, Vol. 1, McGraw-Hill, New York, NY.
Parida, A., Kumar, U., Galar, D. and Stenström, C. (2015), “Performance measurement and management
for maintenance: a literature review”, Journal of Quality in Maintenance Engineering, Vol. 21
No. 1, pp. 2-33.
Pham, H. and Wang, H. (2000), “Optimal (t, T) opportunistic maintenance of a k-out-of-n: G system with
imperfect PM and partial failure”, Naval Research Logistics, Vol. 47, pp. 223-239.
Rander, R. and Jorgenson, D.W. (1963), “Opportunistic replacement of a single part in presence of
several monitored parts”, Management Science, Vol. 10 No. 1, pp. 70-83.
Rao, A.N. and Bhadury, B. (2000), “Opportunistic maintenance of multi-equipment system: a case
study”, Quality and Reliability Engineering International, Vol. 16, pp. 487-500.
Samhouri, M.S. (2009), “An intelligent opportunistic maintenance (OM) system: a genetic algorithm
approach”, IEEE Toronto International Conference on Science and Technology for Humanity
(TIC-STH), pp. 60-65.
Saranga, H. (2004), “Opportunistic maintenance using genetic algorithms”, Journal of Quality in
Maintenance Engineering, Vol. 10 No. 1, pp. 66-74.
Savic, D.A., Walters, G.A. and Knezenic, J. (1995), “Optimal opportunistic maintenance policy using
genetic algorithms, 1: formulation”, Journal of Quality in Maintenance Engineering, Vol. 1 No. 2,
pp. 34-49.
Scarf, P.A. and Deara, M. (2003), “Block replacement policies for a two-component system with failure
dependence”, Naval Research Logistics (NRL), Vol. 50 No. 1, pp. 70-87.
Schultz, J.R. (2006), “Measuring service industry performance: some basic concepts”, Performance
Improvement, Vol. 45 No. 4, pp. 13-17.
Sharma, K.R., Kumar, D. and Kumar, P. (2005), “FLM to select suitable maintenance strategy in process
industries using MISO model”, Journal of Quality in Maintenance Engineering, Vol. 11 No. 4,
pp. 359-374.
Smith, R. and Hawkins, B. (2004), Lean Maintenance: Reduce Costs, Improve Quality and Increase
Market Share, Butterworth-Heineman Publication, Burlington, MA.
Thomas, L.C. (1986), “A survey of maintenance and replacement models for maintainability and
reliability of multi-item systems”, Reliability Engineering, Vol. 16 No. 4, pp. 297-309.
Vu, H.C., Do Van, P., Barros, A. and Bérenguer, C. (2012), “Maintenance activities planning and
grouping for complex structure systems”, Annual Conference of the European Safety and
Reliability Association, PSAM11 & ESREL 2012, Helsinki.
Wiksten, J. and Johansson, M. (2006), “Maintenance and reliability with focus on aircraft maintenance
and spares provisioning”, Bachelor’s thesis, Luleå University of Technology.
Xia, T., Xi, L., Pan, E. and Ni, J. (2017), “Reconfiguration-oriented opportunistic maintenance policy
for reconfigurable manufacturing systems”, Reliability Engineering & System Safety, Vol. 166,
pp. 87-98.
Yang, L., Zhao, Y., Peng, R. and Ma, X. (2018), “Opportunistic maintenance of production systems
subject to random wait time and multiple control limits”, Journal of Manufacturing Systems,
Vol. 47, pp. 12-34.
Yin, R.K. (1994), Case Study Research: Design and Methods, 2nd ed., Sage Publications, Thousand
Oaks, CA.
Zhou, B., Yu, J., Shao, J. and Trentesaux, D. (2015), “Bottleneck-based opportunistic maintenance Opportunistic
model for series production systems”, Journal of Quality in Maintenance Engineering, Vol. 21 maintenance
No. 1, pp. 70-88.
Zhou, W.H. and Zhu, G.L. (2008), “Economic design of integrated model of control chart
policy
and maintenance management”, Mathematical and Computer Modelling, Vol. 47 No. 11,
pp. 1389-1395.
Zhou, X., Xi, L. and Lee, J. (2009), “Opportunistic preventive maintenance scheduling for a multi-unit
series system based on dynamic programming”, International Journal of Production Economics,
165
Vol. 118 No. 2, pp. 361-366.

Further reading
Oyebisi, T.O. (2000), “On reliability and maintenance management of electronic equipment in the
tropics”, Technovation, Vol. 20 No. 9, pp. 517-522.

Corresponding author
Shahrul Kamaruddin can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]