See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/351223134

Effect of Preventive Maintenance on Machine Reliability in a

Beverage Packaging Plant

Article  in  International Journal of System Dynamics Applications · April 2021

DOI: 10.4018/IJSDA.2021070104

CITATIONS READS

0 50

3 authors, including:

Jacob Ben Kamalakanta Muduli

The Papua New Guinea University of Technology C. V. Raman College of Engineering
4 PUBLICATIONS   1 CITATION    70 PUBLICATIONS   904 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

HEALTH CARE MANAGEMENT View project

Investigating the Effect of Preventive Maintenance on Machine Reliability in a Beer Processing Plant. View project

All content following this page was uploaded by Jacob Ben on 17 July 2021.

The user has requested enhancement of the downloaded file.

International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Effect of Preventive Maintenance

on Machine Reliability in a
Beverage Packaging Plant
Jacob Ben, Papua New Guinea University of Technology, Papua New Guinea
Aezeden O. Mohamed, Papua New Guinea University of Technology, Papua New Guinea
Kamalakanta Muduli, Papua New Guinea University of Technology, Papua New Guinea
https://orcid.org/0000-0002-4245-9149

ABSTRACT

This paper investigates the effect of preventive maintenance on the reliabilities of devices in a
bottling plant. Six months of real-time maintenance data were analyzed quantitatively. Based on
the breakdown events obtained for each machine, mean time between failure (MTBF), mean time
to repair (MTTR), and failure rate (λ) values for individual equipment are calculated. The bottle
packer, empty bottle inspector (EBI), and palletizer are identified as the plant’s critical machines.
A breakdown analysis (BDA) is then performed on the bottle packer and from the failure mode of
all the reoccurring problems affecting this machine as a result of ineffective PM. An autonomous
maintenance (AM) team is set up as part of establishing an effective PM program to improve the
reliabilities of the critical machines that were continually falling. A significant reduction in machine
breakdowns is observed after two months of rolling out the AM program. As a result, the reliability
of bottle packer increased from 55.30% to 70.80%, while EBI and palletizer increased from 89.20%
and 87.20% to 92% and 90.50%, respectively.

Keywords
Corrective Maintenance, Failure Rate, Machine Breakdown, Plant Reliability, Preventive Maintenance

1. INTRODUCTION

With competition in the world continually increasing, organizations have been forced to adapt new
competitive strategy (Galli, 2020). One such strategy is how to minimize production loss due to
breakdown. Maintaining a breakdown-free production line is the goal of any manufacturing plant
in ensuring its operating costs kept as minimum as possible. To conduct a reliable maintenance
program has to be in place to maintain the machine and system in good working condition and that
all necessary repairs done on time in full. The type of maintenance program applied can be divided
into two categories: breakdown (emergency) maintenance and preventive maintenance. Breakdown
maintenance deals with machines or systems after it has broken down and failed to perform its
intended function. Preventive maintenance (PM), on the other hand, is time-based maintenance

DOI: 10.4018/IJSDA.2021070104
This article, published
This article as an Open
published as anAccess article article
Open Access on April 23, 2021 under
distributed in the the
goldterms
OpenofAccess journal,
the Creative International
Commons JournalLicense
Attribution of System Dy-
namics Applications (converted to gold Open Access January
(http://creativecommons.org/licenses/by/4.0/) 1, 2021),
which permits is distributed
unrestricted use,under the terms
distribution, andofproduction
the Creativein Commons Attribution
any medium,
License (http://creativecommons.org/licenses/by/4.0/) which
provided the author of the original workpermits unrestricted
and original use, distribution,
publication and production
source are properly in any medium, provided
credited.
the author of the original work and original publication source are properly credited.
50
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

aimed at reducing or avoiding breakdowns and involves carrying out maintenance activities on a
periodically scheduled basis.
One of the main challenge in developing a maintenance strategy with the constraints of
availability of resources, is using the right-thinking framework (Galli, 2018a). These challenges
are significant for small and medium sized plants those face lot of difficulty in their attempts
in innovating maintenance strategies due to their small size and limited resources (Nikabadi,
& Hakaki, 2018). If done right, preventive maintenance has an immense potential to increase
the reliability of a system, enhance machine up-time, minimize equipment breakdown, and
improve the efficiency of production plants at low operating costs. Various activities (cleaning,
inspection, lubrication, tightening, and replacement of worn parts) carried out under the PM
program must have one aim, and that is to increase both the availability and reliability of all
machines on the production plant. Machine reliability is increased by reducing its risk to fail,
hence allowing the device to work correctly to perform its desired functions during a specified
period under predetermined conditions (Kiyak, 2011). Availability is the proportion of time a
machine is available out of the time it should be possible. It increased by minimizing unplanned
downtime and changeover times. A well-maintained production plant will, therefore, have higher
equipment availability and reliability hence be more profitable.
With preventive maintenance, the aim is to detect and prevent faults before they create disturbances
in the machine and production system(Lagnebäck, 2007). As devices get old, their probability of
failure increases much as specific parts and components begin to wear. At this point, leaving the
equipment too long without any form of maintenance will result in a catastrophic failure. The desired
preventive maintenance to perform on a machine influenced by several factors, including the level of
technology, type of production processes, and the age and condition of the equipment. With a higher
degree of PM planning and execution, the total breakdown and time losses of the plant can reduce
significantly, thereby increasing the reliability and availability of the production system.
It is an accepted view that technological innovation is an extremely important factor to support
industrial growth(Omamo et al., 2018) and sustained economic development of any country and
perhaps the most important(Omamo et al., 2020). In this light study on development of suitable
maintenance strategies for the industries from all industrial sectors assumes importance. This
study is conducted to assess effect of PM on machine reliability in a beverage packaging unit
which is scarce in previous literature. Further, the study is conducted in Papua New Guinea,
hence, provides an opportunity to identify the importance of PM and relevant challenges over
the long-term because, resources are in the case of developing economies typically much more
limited, management much less committed and work methods much less organized than in the
industrialized world(Haddad & Otayek, 2019). Heineken South Pacific Brewery Limited in Papua
New Guinea conducted a study from June to December 2019. The purpose of the study was
to increase knowledge and evaluate the company’s preventive maintenance program to decide
if the level of planning, scheduling, and execution of maintenance activities was fulfilling the
requirements for reducing machine breakdowns and time losses in the beer packaging plant. The
research compiled quantitatively using real-time data collected from (1) the brewery’s online
production data recording software called IVAMs, and (2) historical breakdown data from the
maintenance team. These were then compiled, tabulated, and in-depth analysis performed. A
literature review was conducted as part of the research to compare the empirical findings with
peer-reviewed information to ensure the quality of the study.
Based on the results of the study, conclusions made on the relationship between failure
rates and ineffective or no preventive maintenance on machines on the production plant.
Significant factors contributing to high downtimes on the plant’s critical device identified,
and counter-measures developed to prevent breakdowns from reoccurring, hence improving
overall plant reliability.

51
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

2. MAINTENANCE STRATEGIES

Maintenance defined as the combination of all technical, administrative, and managerial actions
during the life cycle of an item intended to retain it in, or restore it to, a state in which it can perform
the required function (Mkemai, 2011). In manufacturing, any equipment subjected to plant operations
requires proper maintenance and repair to improve safety and production capacity. According to
Mkemai (2011), several maintenance strategies are available and divided into two major groups,
namely, preventive maintenance (PM) and corrective maintenance (CM). Preventive maintenance
tasks, as stated by Mobley (2002), are intended to prevent unscheduled downtime and premature
equipment damage that would result in corrective or repair activities. As discussed by Mkemai (2011),
the two groups can subdivide into subgroups shown in Figure 1.
Corrective maintenance strategy has the lowest investment, the highest operating cost, and
provides the lowest equipment availability (Adale, 2009). The sub grouped into deferred and immediate
maintenances were according to Márquez (2007). Deferred is one in which CM has not immediately
carried out after a fault is detected but is delayed based on given maintenance rules (e.g., lack of
spare parts, budget cycle, and climate condition). When maintenance deferred, component, or system
performance deteriorates rapidly, bringing the time at which failures are unacceptably frequent
earlier in the system’s life (Jessica, 2014). On the other hand, immediate maintenance is one where
CM is carried out without delay after a fault has detected to avoid the unacceptable consequences of
component or system failure (Adale, 2009).
Also known as time-based maintenance or planned maintenance, PM is a proactive type of
maintenance strategy aimed at reducing or avoiding breakdowns through a program of cleaning,
inspection, lubrication, tightening, and replacement of worm parts. The planned PM tasks, usually
carried out at regular intervals when the equipment is still functioning, are performed before a failure
has occurred(Hupjé,2018). If implemented correctly, PM can be very useful in preventing a machine
or component from catastrophic failure, thereby minimizing the number of emergency breakdowns
and machine downtime. However, if PM poorly planned and executed, equipment reliability will be
low, thus leading to higher maintenance costs due to frequent malfunctions and procurement and
replacement of spares(Olose, 2016).
Condition-based maintenance (CBM), also referred to as predictive maintenance(PdM), is
a preventive maintenance method carried out based on certain conditions of the machine. The
machine conditions can be monitor continuously or periodically depending upon the need for the
availability of the equipment(Morales, 2002). Real-time data indicating machine condition can acquire
through monitoring techniques such as vibration measurements and analysis, oil analysis, infrared
thermography, laser systems, and ultrasonic(Rao, 1996). Based on this data, maintenance activities

Figure 1. Classification of maintenance strategies (Mkemai, 2011)

52
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

and resources prioritized accordingly. Since CBM utilizes real-time data to predict machine failure
due to temperature, pressure, vibration, noise, humidity, and changes in the electrical and mechanical
characteristics of critical equipment) it is much more cost-effective than maintenance on failure
or scheduled maintenance(Mahmood, 2011). Scheduled preventive maintenance usually results in
repairing or replacing components too early or too late, leading to support that is too expensive and
ineffective in preventing breakdowns.
Predetermined maintenance, according to Swedish Standard Institute (2001), is maintenance
carried out by established intervals of time or number of units of use but without previous condition
investigation. The actions performed under this type of maintenance strategy are decided based on a
predetermined rule, usually involving timing or a measure of usage(Noriega, 2019). Thus, preventive
maintenance to be performed is either clock-based or usage-based actions. In clock-based keeping,
the preventive actions conducted at specified calendar times, for instance, overhaul every first
quarter of the year. With usage-based maintenance, scheduled PM activities based on the number
of specific production metrics like operating hours, name of products made, and several processed
parts (Nieminen, 2016).

3. MACHINE RELIABILITY AND MAINTAINABILITY

3.1 Reliability
Reliability engineering measures and improves resistance to failure over time, estimates expended
life, and predict time to failure of equipment or a component . There are two overlapping reliability
themes for modeling life and performance of items; these are data/evidence-driven and physics-
driven. Evidence-driven reliability analysis relies on probabilistic and statistical methods and based
on collecting data and assessing future behavior of things based on past occurrences of failure or
performances. On the other hand, physics-driven reliability analysis based on the physics of failure.
This model of failure has an empirical representation of the physics of how a machine or component
degrade and is established through a significant amount of knowledge and testing to understand the
underlying behaviors of failure.
Availability is the proportion of time a machine or a system is available out of the time it
should be possible. It increased by minimizing unplanned downtime and changeover times. In
contrast, the reliability of a system is the probability that it will perform or operate the required
functions without failure under a given condition for an intended operating period(Ghodrati, 2005).
Lower reliability means increased unplanned stoppages and, consequently, unscheduled repairs
and decreased availability(Tatis, 2012). Forgiven equipment, safety is measured by calculating
its mean time between failure (MTBF) or mean time to failure (MTTF), and mean time to repair
(MTTR)(Deka & Nath 2015). MTTF used for non-repairable components, for instance, electrical
equipment like drive motor, while MTBF used for repairable elements, for example, a rotating
shaft or a pulley belt(Kiyak, 2011). MTTR is the time it takes to run a repair after the occurrence
of a failure. MTBF and MTTR are two critical key performance indicators (KPIs) when it comes
to the availability of a system, facility, equipment, or process. The formulas for determining MTBF
and MTTR given in equations (1) and (2), respectively:

OperatingTime
MTBF = (1)
Number of Failures

Downtime
MTTR = (2)
Number of Failures

53
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Reliability, R(t), is the reliability of a system or a piece of equipment at time t (Modarras,2019).

To understand safety, let’s consider a production plant operating on 12 hours shift per day. In this
situation, t is 12 hours, and R(t) is the probability that the production plant will perform its desired
functions without any downtime due to machine failure. The general formula for the reliability function
of the exponential distribution is given in equation (3) Mitra(2016):

 t 
− 
(−λt )
R (t ) = e
 MTBF 
=e (3)

1
Where λ = = failure rate (4)
MTBF

The lambda, λ, in the reliability equation (3) represents the failure rate (failures/hour). There is
an assumption of random independent failures, which is generally useful for electronic components
in computers but might not apply to everything, especially mechanical components. An issue with λ
is that a constant failure rate is a bit of fiction. If we look at a graph of failure rate vs. component age
(Figure 2), often called a bathtub curve, we see that the constant failure rate is only valid during the
middle of component life. At the beginning (point A to B), according to Mitra(2016), components
have a high failure rate due to manufacturing defects that manage to pass the test but make them part
of wearing out very quickly. It is often called the “burning phase” and can last 1 hour, or 10 hours,
or 100 hours depending on the system. After burning, the useful product life does indeed provide
a more or less constant λ. Considering only the flat part of the curve (point B to C), equation (3) is
used to determine the reliability of the useful product life.

3.2 Maintainability
Another critical parameter is machine maintainability. Maintenance usually performed by skilled
personnel using prescribed procedures and resources (Reliability Analysis Center). It is typically
measured by mean time to repair (MTTR) calculation using equation (2).
Maintainability is a design attribute closely related to maintenance prevention and goes hand in
hand with reliability. According to Gulati&Smith(2009), both safety and maintainability are designed
into the assets to minimize maintenance needs and is done by using reliable components, simplify
replacements, and ease inspections. Generally expressed as the probability that a system will restore to
running conditions within a given time when the maintenance carried out according to recommended

Figure 2. Typical bathtub curve(Mitra, 2016)

54
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

procedures (Haider, 2007). An example of a stated maintainability goal is a 90% probability that
maintenance repair times will complete in 8 hours or less with a maximum repair time of 24 hours,
which will require a system MTTR of 3.48 hours (Barringer, 1997).
The maintainability function, M(t), given in equation (5), can be used to determine the probability
of performing a maintenance action for a system with the repair times exponentially distributed [23].
High reliability (few failures) and high maintainability (predictable maintenance times) result in a
highly effective production system:

 t 
 
M t   1  e  MTTR 
(5)

where:

t = allowable downtime
MTTR = expected downtime (mean time to repair)

The expected downtime for repair includes waiting time for repairs, time spent doing repairs,
and time spent testing and getting the equipment ready to resume operation.

4. EFFECT OF PREVENTIVE MAINTENANCE ON MACHINE RELIABILITY

4.1 Maintenance Strategy

Heineken South Pacific Brewery at Lae, Papua New Guinea, has two production departments, brewing
and packaging, and one support department, engineering. The packaging is the final manufacturing
stage where the beer is processed into cans and bottles and comprises two processing plants, the
canning, and the bottling plants. The plant fully automated with a capacity to produce up to 150,000
cartons of beer products every week to meet both local and international market demands. Both these
plants have the aim of increasing productivity by effectively utilizing their available production time.
To support and maintain a breakdown-free production line, the goal of the maintenance team
is to (1) effectively execute all PM activities during scheduled stops (planned downtime), and (2)
decrease the time used up to fix a machine or a particular component during breakdown (unplanned
downtime). The execution of all preventive maintenance activities is time-driven. The maintenance
tasks based on elapsed time or hours of operation; t is scheduled every Monday from 0800 to 1700
hours. Two sets of tasks performed: first are those immediate tasks based on work orders (WO)
created through failure tags raised by the production team, for instance, a ticket raised on a conveyor
lubrication spray nozzle not spraying during regular operation. The second set of tasks are those that
are performed based on recommendations in the manufacturer’s service manual, for instance, change
lubrication of rolling element bearings after 60 working days. All tasks aimed to maintain machines
or components at optimal conditions so that the likelihood of a failure reduced. Figure 3 shows the
maintenance strategy applied at the brewery.

4.2 Downtime Data

Figure 4 shows the layout of the major equipment on the bottling plant. Although there is an established
maintenance program to cater to the maintenance needs of this equipment, the production plant is
still plagued by delays due to frequent breakdowns, an increase in time taken to fix a failed machine
or component, and a low mean time to repair (MTTR). Information obtained from the brewery’s
online data recording software, IVAMs, and historical breakdown data from the maintenance team
shows that there has been a significantly high number of unplanned, disruptive stoppages that had

55
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Figure 3. Heineken South Pacific Brewery’s maintenance strategy at Lae manufacturing plant

a significant impact on production output. Based on the breakdown hours and several breakdown
events registered for each machine, the mean time between failures (MTBF), mean time to repair
(MTTR), and failure rate (λ) values for the individual device calculated using equations (1), (2),
and (4) respectively. These data then used to determine the reliability of each machine for a 12- and
24-hours operation using equation (3). Table 1 provides a summary of these calculations.
A critical machine in any production plant is one with the highest number of breakdown events
and has the most extended downtime. Based on Table 1, the bottle packer identified as the plant’s
critical machine with 127 breakdown events and 104.52 downtime hours. The empty bottle inspector
and palletizer machines are considered second and third essential machines with breakdown times
of 58.98 and 39.30 hours, respectively.

4.3 Reliability Analysis

Reliability, R(t), of the bottling plant is the probability that the plant will continue to operate for a
given time, for instance, 12 hours of a shift, without any downtime due to machine failure. The closer
the reliability value is closer to 1, the higher the chances of the plant’s ability to perform appropriately.
The reliability determined by calculating the reliability of all the proper equipment that makes up
the production plant.
From equation (4), lambda values for each machine that broke down for the six months determined,
and the result given in Table 1. With the λ values, the reliability of equipment calculated using equation
(3) for different time intervals (t = 12 and 24 hours). The reason behind finding the reliability on
a 12 hours interval is because the company currently has a 12 hours/shift, 24 hours/day, six days/
week continuous operation with only 9 hours of every Monday scheduled for executing preventive
maintenance tasks. Any disruptive stops that occur during production approached as breakdown
maintenance (or immediate corrective maintenance).
The implication of the exponential equation, (3), is that it is much harder to successfully sustain
and complete a long production uptime than a short one. What this means is that operating a production
plant with zero failure for several hours is a lot harder than running without failure for one hour. The
illustrated in Figure 5. The graph shows a decrease in the reliability of individual machines on the
plant as time goes from 12 to 24 hours, which ultimately results in low plant reliability. From Table
1, This shows that the longer the MTBF of a machine, the more reliable the equipment. For instance,
pasteurizer (MTBF=126.046 hours) is more reliable than the bottle packer (MTBF=20.248 hours).

4.4 Bottling Plant Reliability in 12 and 24 Hours

The overview of the bottling plant (Figure 4) shows that the machines in the production plant arranged
in series. Note that carton formers 1 and 2, carton check weighs 1 and 2, and carton sealers/coders

56
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Table 1. Machine downtime data include calculated λ and reliability values for t = 12 and 24 hours

R(t), R(t),
Breakdown Breakdown MTBF MTTR
Machine λ t=12 t=24
Time (hr) Events (hr) (hr)
(hr) (hr)
Bottle Packer 104.520 127.000 20.248 0.823 0.049 0.553 0.306
Empty Bottle Inspector 58.980 25.000 104.681 2.359 0.010 0.892 0.795
Palletizer 39.300 30.000 87.890 1.310 0.011 0.872 0.761
Pasteurizer 29.030 21.000 126.046 1.382 0.008 0.909 0.827
Bottle Labeler 28.150 32.000 82.745 0.880 0.012 0.865 0.748
Bottle Filler 28.070 35.000 75.655 0.802 0.013 0.853 0.728
Depalletiser 26.930 30.000 88.302 0.898 0.011 0.873 0.762
Bottle Washer 25.550 23.000 115.237 1.111 0.009 0.901 0.812
Empty Bottle Conveyor 2A 17.450 28.000 94.948 0.623 0.011 0.881 0.777
Label Coder 14.530 8.000 332.684 1.816 0.003 0.965 0.930
Shrink Packer 11.830 11.000 242.197 1.075 0.004 0.952 0.906
Depalletiser (pallet discharge) 8.980 10.000 266.702 0.898 0.004 0.956 0.914
Empty Bottle Conveyor 1A 8.530 12.000 222.289 0.711 0.004 0.947 0.898
Full Bottle Conveyor 4B 7.320 11.000 242.607 0.665 0.004 0.952 0.906
Heuft Inspector 2 (FBI 2) 5.900 6.000 445.017 0.983 0.002 0.973 0.947
Full Bottle Conveyor 3A 5.900 2.000 1335.050 2.950 0.001 0.991 0.982
Carton Former 1 4.620 8.000 333.923 0.578 0.003 0.965 0.931
Carton Coder 3.930 8.000 334.009 0.491 0.003 0.965 0.931
Full Bottle Conveyor 4A 3.900 5.000 534.420 0.780 0.002 0.978 0.956
Heuft Inspector 1 (FBI 1) 2.980 5.000 534.604 0.596 0.002 0.978 0.956
Check Weigher 2.850 2.000 1336.575 1.425 0.001 0.991 0.982
Carton Sealer 2.780 4.000 668.305 0.695 0.001 0.982 0.965
Full Bottle Conveyor 3B 2.150 2.000 1336.925 1.075 0.001 0.991 0.982
Bottle Crowner 1.730 2.000 1337.135 0.865 0.001 0.991 0.982
Full Bottle Conveyor 2B 1.650 4.000 668.588 0.413 0.001 0.982 0.965
Full Bottle Conveyor 1B 0.880 2.000 1337.560 0.440 0.001 0.991 0.982
Empty Bottle Conveyor 2B 0.670 1.000 2675.330 0.670 0.0004 0.996 0.991
Empty Bottle Conveyor 3A 0.400 2.000 1337.800 0.200 00.001 0.991 0.982
Empty Bottle Conveyor 2C 0.370 1.000 2675.630 0.370 0.0004 0.996 0.991
Empty Bottle Conveyor 5 0.370 1.000 2675.630 0.370 0.0004 0.996 0.991
Packer 0.350 1.000 2675.650 0.350 0.0004 0.996 0.991
Coder 0.350 1.000 2675.650 0.350 0.0004 0.996 0.991
Empty Bottle Conveyor 4 0.230 1.000 2675.770 0.230 0.0004 0.996 0.991
TOTAL 451.180 461.000 4.826 0.979

57
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Figure 4. Overview of bottling plant

Figure 5. Reliability vs. bottling plant machines at 12 and 24 hours

1 and 2 are considered as one unit because the design of the line is such that if one of the pair stops
due to a component or system failure, the performance of the plant is severely affected. For instance,
if carton former one fails, then package former 2 is the only one available to support bottle packer
with erected boxes. This results in the bottle packer operator continuously transferring constructed
boxes from lanes 1 to 2, and in the process, there is a delay in meeting the required quantity per hour.
Figure 6 is a simplified block diagram of the line for reliability calculations.
Where the numbers represent (1) bottle packer, (2) empty bottle inspector, (3) palletizer, (4)
pasteurizer, (5) bottle labeler, (6) bottle filler, (7) depalletize, (8) bottle washer, (9) empty bottle

58
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Figure 6. Series arrangement of machines for reliability assessment

conveyor 2A, (10) label coder, (11) shrink packer, (12) depalletize (pallet discharge), (13) empty bottle
conveyor 1A, (14) full bottle conveyor 4B, (15) heuft inspector 2 (FBI 2), (16) full bottle conveyor
3A, (17) carton former 1, (18) carton coder, (19) full bottle conveyor 4A, (20) heuft inspector 1 (FBI
1), (21) check weigh, (22) carton sealer, (23) full bottle conveyor 3B, (24) bottle filler crowner, (25)
full bottle conveyor 2B, (26) full bottle conveyor 1B, (27) empty bottle conveyor 2B, (28) empty
bottle conveyor 3A, (29) empty bottle conveyor 2C, (30) empty bottle conveyor 5, (31) packer, (32)
coder, (33) empty bottle conveyor 4.
Equation (6) used to calculate the reliability of machines or components in series. It is the
multiplication of the reliabilities of the individual devices in the series:

ReliabilitySeries = R1 * R2 * R3 * …Rn (6)

where R1, R2, R3, Rn are the reliabilities of machines 1, 2, 3 through to (nth) machine. Therefore, using
Table 1 the reliability of the plant at t=12 hours is:

R(12) = 0.553*0.892*0.872*0.909*0.865*0.853*0.873*0.901*0.881*0.965*0.952*0.956*0.947*
0.952*0.973*0.991*0.965*0.965*0.978*0.978*0.991*0.982*0.991*0.991*0.982*0.991*0.996*
0.991*0.996*0.996*0.996*0.996*0.996
= 0.1223*100
= 12.23%

Reliability at t=24 hours:

R(24) = 0.306*0.795*0.761*0.827*0.748*0.728*0.762*0.812*0.777*0.930*0.906*0.914*0.898*
0.906*0.947*0.982*0.931*0.931*0.956*0.956*0.982*0.965*0.982*0.982*0.965*0.982*0.991*
0.982*0.991*0.991*0.991*0.991*0.991
= 0.0149*100
= 1.49%

4.5 Machine Breakdown Analysis

As an approach to understanding why equipment or components on the bottling plant failed, a
breakdown analysis (BDA) performed on the machine with the highest breakdown events. The bottle
packer, with downtime events of 127, was selected for the study. The purpose of a BDA is to identify
the root cause behind a failed component to implement countermeasures that will prevent failures
from reoccurring. Machine downtimes lead to low overall equipment effectiveness (OEE); therefore,

59
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

it is essential that once a component fails, the failure modes identified and correct preventive actions
are put in place to eliminate the problem. Table 2 lists the ingredients of the bottle packer that failed
during the six months, the basic description of the function of each component, and the failure mode
that resulted in a machine breakdown.
From catastrophe mode repeatedly affecting bottle packing machine, results indicated ineffective
preventive maintenance. Because the plant set to operate on 12 hours per shift, 24 hours per day,
continuous operation for almost every single week, the maintenance team hardly has enough time to
complete all PM tasks. Additionally, a lack of understanding of the root cause of equipment failure
from the planned maintenance team results in ineffective planning and execution of maintenance
tasks during the 9 hours of scheduled downtime every Monday. Hence, almost all failure modes are
associated with either damaged or worn parts. For instance, the issue with cylindrical grippers not
picking up bottles during production is an on-going problem related to damage grippers caused by
fluctuation in working air (compressed air). The maintenance team goes into fire-fighting mode every
time the gripper issue arises, which always results in heavy downtime to the plant.
Lack of spare parts for the major components of the bottle packer is another factor that contributed
to increased downtime. Its fact, in the case of the rectangular grippers on the packing head for 4x6
shrink packs. The issue with a fallen six-pack is so common that it has now accepted as a norm. A
closer look at the component shows severe damage to the rubber-protected edges of the grippers.
Many order tags have raised by production operators indicating worn parts needing immediate
replacement, but the response they get from the maintenance team is the unavailability of spare parts.
A robust preventive maintenance program is one that has an effective tracking system for spare part
availability to support a continuous operation. Since the operation of the bottle packing machine is
continuous and functions on a 24/7 basis, the role of the maintenance planning team is critical to
improving machine operation in achieving improved plant reliability.

4.6 Relationship Between Failure Rate and Maintenance

The failure rate of the machines on the bottling plant increased with time during the six months of
operation. The effect of performing less preventive maintenance tasks on the machines because the
plant is operated continuously for 24 hours, six days a week, the machines work for longer times,
whereby the failure rate of their components increases to the point that results in equipment failure.
As machine or component failure rate increases, more corrective actions are required hence reducing
the reliability of the production plant.
Figure 7 shows the cumulative failure rate of machines on the plant. There are two curves shown,
red and green. The red curve indicates a cumulative failure rate of machines before autonomous
maintenance (AM), and countermeasures introduced as part of developing an effective PM program
for the components of critical machines that were prone to reoccurring breakdowns due to wear
and tear. Based on BDA analysis performed on the bottle packer, root causes of failed components
identified. Maintenance tasks then planned and executed accordingly. Detailed checklists designed for
periodic inspection lead to the effective planning of parts maintenance and replacement on the bottle
packing machine as well as the other two critical machines. The countermeasures (regular inspection,
replacement of worn parts, detailed PM, lubrication, and tightening) planned and performed over
two months. Improvement made, with a decline in plant cumulative failure rate (green curve), as
shown in Figure 7.

5. RESULTS AND DISCUSSION

Recommendations were put forward for countermeasures to developed for the bottle packer as part
of establishing an effective PM program for plant reliability improvement. All achieved by setting up
an autonomous maintenance (AM) team to carry out continuous improvement activities periodically
daily on the main components of the bottle packer. The team comprised three highly experienced

60
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Table 2. Components of bottle packer and their failure modes that resulted in machine breakdown

Machine component Component function Functional failure Failure mode

A touch screen that displays information about Loose ethernet

Human machine interface
machine state. The application software accepts and HMI screen freezed. connection cable on
(HMI)
implements operator’s control instructions. Communication module.

Moves over the bottle table and the cylindrical

grippers pick up the bottles. Then moves with the Between 4 to 12 cylindrical
Gripper head (for individual Damaged or worn
bottles over the carton conveyor and places the bottles grippers not picking up
bottles) suction heads.
in the cartons. The grippers return, cartons discharge, bottles (total 192 grippers)
and the cycle continues.

Moves over the shrink six pack table, and the

rectangular grippers pick up the packs. Then moves 2 to 4 rectangular grippers
Packing head (for 4x6 shrink with the packs over the carton conveyor and places continuously dropping six Damaged or worn
packs) the packs in the cartons. The grippers return after packduring transfer (total rectangulargrippers.
releasing the packs, cartons discharge, and the cycle 32 rectangular grippers)
continues.

Actuators that enable all lanes on the bottle table to be

Switching flags not
completely filled with bottles, ensuring that the full Damaged or broken
Switching flags actuated, continuously
number of bottles for the packing process is reached flag.
stopping the machine.
before the grippers pick up the bottles.

A metal frame with plastic aprons that centers the

Empty cartons not
empty cartons in the packing zone before the gripper Damaged centering
Correction frame positioned or centered
(or packing) head deposits the bottles (or six packs) aprons.
correctly.
into the cartons.

Secures the production area of the machine against

Unable to lock/unlock Loose and damaged
unauthorized entry. A message is sent to the central
main sliding door. CPU cable for operating
Safety door processing unit (CPU) if a safety door to the secured
not permitting machine for controls on actuation
area is opened, and the CPU triggers an emergency
production. housing.
stop for the components in the secured area.

Mat conveyor not running,

Six pack conveyor that moves packs into the packing Broken mat links due to
Infeed mat conveyor unable to deliver six packs
zone (receiving position) of the packing head. wear and fatigue stress.
to the receiving table.

Transports the cartons into and out of the packing

zone. The conveyor is made of plastic chains and are
Worn, damaged chain
Carton conveyor rotated by chain wheels seated on a primary and tail Conveyor stopped running.
links.
shafts. A motor drives the primary shaft via a gearing
mechanism.

Motor failed to supply

Supplies motive power to a primary shaft that drives
Infeed geared motor for carton motive power to primary
the conveyor chains to transport cartons into and out Ceased bearing.
conveyor shaft. Motor makes high
of the packing zones.
vibrating noise.

Pneumatically controlled device on the packing head

Shaft on the pneumatic
that control the movement of grippers (cylindrical
Pneumatic cylinders on cylinders unable to open Faulty reed switch on
cups or rectangular grippers). They control the
packing head and close thereby unable to pneumatic cylinders.
grippers to open and close pneumatically to pick up
control gripper movement.
and deposit bottles or six packs.

Compressed air supply connected to the maintenance

Packing head dropping
unit of the bottle packer. The maintenance unit Dirty air filter on the
Air supply unit bottles (or six packs) due to
supplies the pneumatic machine components with maintenance unit.
insufficient air supply.
working air.

Toothed belt pulleys that connect lifting frame,

counterweights, and lifting unit. The lifting frame Lifting unit not working. Worn application rollers
Synchronous belts is raised or lowered according to the lifting unit’s Unable to raise or lower causing belt to slip off
direction of rotation. The belt tension is monitored by Lifting frame. position.
synchronous belt controls.

Unable to transport cartons

System network for real-time distributed control for into and out of the packing
Faulty field bus
Field bus cable the bottle packer. Field bus connection for bottle and zone. HMI indicated
connecting cable.
carton transport. faulty bus slave cable and
conveyor stops running.

61
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Figure 7. Cumulative failure rate of equipment on the bottling plant

operators from the production team, one electrical specialist and one mechanical shift technician from
the engineering team, and one production supervisor as a team leader to provide overall supervision.
Checklists planed for the machine’s critical components that prone to reoccurring breakdowns due to
wear and tear. Additional preventive maintenance activities were also planned for two other essential
machines, EBI, and palletizer, having the second and third highest breakdown hours. Apart from
the 9 hours of PM stop every Mondays, for periodic inspection and maintenance carried out during
planned stops (changeover, daily cleaning, inspection, lubrication, and tightening) for two months
and the results showed improvement with a noticeable decrease in breakdown events and downtime
and an increase in MTBF on these machines. Figure 8 shows these results.
From Table 1 and Figure 8, the total breakdown time for the bottle packer reduced from 104.52
hours to 93.41 hours. The same trends also observed on the empty bottle inspector and palletizer.
EBI breakdown was reduced from 58.98 hours to 47.29 hours, while palletizer from 39.30 hours to
31.06 hours. Table 3 shows the results. A noticeable increase in MTBFs on the critical machines
observed. MTBF of bottle packer increased from 20.248 hours to 34.730 hours, EBI and palletizer
machines had their MTBFs increased from 104.681 and 87.890 hours to 143.550 and 120.160 hours,
respectively. The higher the MTBF, the more reliable the machine.

Figure 8. Breakdown time vs machines on the bottling plant

62
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Table 3. Critical machine downtime data

R(t), R(t),
Time Breakdown MTBF MTTR
Machine λ t=12 t=24
(hr) Events (hr) (hr)
(hr) (hr)
Bottle Packer 93.41 102.00 34.730 0.920 0.029 0.708 0.501
Empty Bottle Inspector 47.29 25.000 143.550 1.890 0.007 0.920 0.846
Palletizer 31.06 30.000 120.160 1.040 0.008 0.905 0.819

Under AM program, operators of critical machines selected based on the expert on those machines
bottle packer, EBI, and palletizer, were empowered to perform adjustments and minor maintenances
(cleaning, inspection, lubrication, tightening) on a day to day basis. Analysis of machine reliability
delivered after two months of rolling out the AM program, and the result showed improvement in
reliabilities of the critical machines. The reliability of bottle packer increased from 55.30% to 70.80%,
while EBI and palletizer from 89.20% and 87.20% to 92% and 90.50%, respectively. Figure 9 results
show some improvements observed in other machines on the plant. Red and green bars represent
machine reliabilities before and after the AM program, respectively.

6. CONCLUSION

The regular adoption and use of modern technology, the change inlatency of business processes
makes it more challenging to decide suitable maintenance strategies (Galli, 2018b).In this light this
paper investigated the effect of preventive maintenance on the reliabilities of machines on the bottling
plant of a beer manufacturing company. The research compiled quantitatively using six months of
real-time data. Within this period, the plant was in operation for 111.5 days, an equivalent of 2676
hours of production. The analysis showed a total of 461 breakdown events that caused the company
451.18 hours to lose time in unplanned stoppages. Based on the breakdown events obtained for each
machine, mean time between failure (MTBF), mean time to repair (MTTR), and failure rate (λ)
values for individual equipment calculated. The bottle packer identified as the critical machine with
the highest of 104.52 hours of breakdown time. A breakdown analysis (BDA) then performed. From
the failure mode of all the reoccurring problems affecting this machine, the result indicated mainly
ineffective preventive maintenance. The results showed that there is a lack of understanding of the

Figure 9. Machine reliabilities before and after autonomous maintenance

63
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

root cause of equipment failure from the planned maintenance team; hence the level of planning,
scheduling, and execution of maintenance activities was not fulfilling the requirements for reducing
machine breakdowns and time losses on the plant. Furthermore, the analysis showed that the machines
on the plant stretched to operate for longer times with less or no scheduled PM activities; hence failure
rates of components increased to the point that resulted in unplanned breakdowns.
An autonomous maintenance (AM) team set up as part of establishing an effective PM program to
improve the reliabilities of the critical machines that were continually falling. Experienced operators
with unique knowledge and skills on several machines were selected to perform adjustments and minor
maintenances (cleaning, inspection, lubrication, tightening) on a day to day basis using specifically
designed checklists as countermeasures to preventing machine failure. Noticeable reduction in machine
breakdowns observed after two months of rolling out the AM program. The total breakdown time
for the bottle packer reduced from 104.52 hours to 93.41 hours. The same trends also observed on
the empty bottle inspector and palletizer machines. EBI breakdown was reduced from 58.98 hours
to 47.29 hours, while palletizer from 39.30 hours to 31.06 hours. As a result, the reliability of bottle
packer increased from 55.30% to 70.80%, while EBI and palletizer from 89.20% and 87.20% to 92%
and 90.50%, respectively.

64
 International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

REFERENCES

Adale, M. A. (2009). Evaluation of Maintenance Management Through Benchmarking in Geothermal PowerPlants

(Master of Science Thesis). Department of Mechanical and Industrial Engineering, University of Iceland.
Barringer, H. P. (1997). Availability, Reliability, Maintainability, and Capability. Barringer& Associates Inc.
Deka, D., & Nath, T. (2015). Breakdown and Reliability Analysis in a Process Industry. InternationalJournal
of Engineering Trends and Technology, 28(3), 150–156. doi:10.14445/22315381/IJETT-V28P229
Galli, B. J. (2018a). Systems Thinking and Systems Analysis in Six Sigma: A Relational Review. International
Journal of System Dynamics Applications, 7(4), 98–112. doi:10.4018/IJSDA.2018100106
Galli, B. J. (2018b). Application of system engineering to project management: How to view their relationship.
International Journal of System Dynamics Applications, 7(4), 76–97. doi:10.4018/IJSDA.2018100105
Galli, B. J. (2020). Measurement System Analysis and System Thinking in Six Sigma: How They Relate
and How to Use Them. International Journal of System Dynamics Applications, 9(1), 44–62. doi:10.4018/
IJSDA.2020010103
Gulati, R., & Smith, R. (2009). Maintenance and Reliability Best Practices. Industrial Press.
Ghodrati, B. (2005). Reliability and Operating Environment Based Spare Parts Planning (Doctoral Thesis).
Division of Operational and Maintenance Engineering, Luleå University of Technology.
Haddad, M., & Otayek, R. (2019). Assessing the Sustainment of a Lean Implementation Using System Dynamics
Modeling: A Case Study of Apparel Manufacturing in Lebanon. International Journal of System Dynamics
Applications, 8(4), 14–29. doi:10.4018/IJSDA.2019100102
Haider, S. J. A. (2007). Role of Maintenance in Component and System Reliability: A Diesel Power GenerationCase
Study (Master of Engineering Thesis). Department of Mechanical Engineering, University of Alberta.
Hupjé, E. (2018). How to Choose the Right Maintenance Strategy. https://www.roadtoreliability.com/types-of-
maintenance/#Corrective_Maintenance_CM
Jessica, R. (2014). Modeling Preventive Maintenance in Complex Systems. Open Access Theses. 244. Retrieved
Dec, 2019 from: https://docs.lib.purdue.edu/open_access_theses/244
Kiyak, E. (2011a). The Importance of Preventive Maintenance in terms of Reliability in Aviation Sector.
Academic Press.
Kiyak, E. (2011b). Recent Advances in Manufacturing Engineering. Retrieved Dec, 2019 from: https://www.
researchgate.net/publication/266492733
Lagnebäck, L. (2007). Evaluation of wayside condition monitoring technologies for condition-based maintenance
of railway vehicles (Licentiate Thesis). Department of Civil, Mining and Environmental Engineering, Luleå
University of Technology.
Mahmood, S. T. (2011). Use of Vibrations Analysis Technique in Condition Based Maintenance (Master of
Science Thesis). Department of Production Engineering, Royal Institute of Technology, Sweden.
Márquez, A. C. (2007). The Maintenance Management Framework: Models and Methods for Complex Systems
Maintenance. Springer-Verlag.
Mitra, A. (2016). Fundamentals of quality control and improvement. John Wiley & Sons.
Mkemai, R. (2011). Maintenance Procedures and Practices for Underground Mobile Mining Equipment (Master
of Science Thesis). Department of Civil, Environmental and Natural Resources Engineering, Luleå University
of Technology.
Mobley, R. K. (2002). An Introduction to Predictive Maintenance. Butterworth-Heinemann Publications.
Morales, D. K. (2002). CBM Policy Memorandum. Washington, DC: Deputy under Secretary of Defense for
Logistics and Material Readiness.

65
International Journal of System Dynamics Applications
Volume 10 • Issue 3 • July-September 2021

Nieminen, H. (2016). Improving Maintenance in high-volume manufacturing. Case: Ball Beverage Packaging
Europe (Master’s Thesis). Lahti University of Applied Sciences.
Nikabadi, M. S., & Hakaki, A. (2018). A dynamic model of effective factors on open innovation in manufacturing
small and medium sized companies. International Journal of System Dynamics Applications, 7(1), 1–26.
doi:10.4018/IJSDA.2018010101
Noriega, M. (2019). Review and Classification of Industrial Boilers Maintenance and a Reliability-Centered
Maintenance Methodology Proposal for Production Plants (Master of Science Thesis). Department of Electronics,
Information and Bioengineering, Polytechnic University of Milan.
Omamo, A. O., Rodriguez, A. J., & Muliaro, J. W. (2018). A Systems Dynamics Model for Mobile Industry
Governance in the Context of the Kenyan Vision 2030. International Journal of System Dynamics Applications,
7(2), 81–100. doi:10.4018/IJSDA.2018040105
Omamo, A. O., Rodrigues, A. J., & Muliaro, W. J. (2020). A System Dynamics Model of Technology and
Society: In the Context of a Developing Nation. International Journal of System Dynamics Applications, 9(2),
42–63. doi:10.4018/IJSDA.2020040103
Olose, E. O. (2016). Effective Maintenance and Reliability Program in the Production of Crude Oil and Natural
Gas. International Journal of Scientific and Engineering Research, 7(2), 1048–1056.
Rao, B. K. N. (1996). Handbook of Condition Monitoring. Elsevier Advanced Technology Publications.
Reliability Analysis Center (RAC). (n.d.). Introduction to Operational Availability (A0). Author.
Swedish Standard Institute. (2001). SS-EN 13306 – Maintenance terminology. Swedish Standard Institute.
Tatis, R. (2012). Vibration Measurement for Rotatory Machines – Importance of maintenance practices
(Bachelor’s thesis). HAMK University of Applied Sciences, Finland.

Jacob Ben is currently working fulltime as a Packaging Supervisor with Heineken South Pacific Brewery Limited,
Lae Morobe Province, Papua New Guinea. He has a Bachelor degree in Mineral Processing and is working towards
completing his masters in mechanical engineering at Papua New Guinea University of Technology. His thesis is
about investigating the effect of preventing maintenance on machine reliabilities.

Aezeden Mohamed has B.Sc., M.Sc., and Ph.D. degrees in Mechanical and Manufacturing at the University of
Manitoba, Canada. His areas of research are experimental in nature includes but not limited; mechanical properties,
materials characterizations, corrosion and corrosion control, and biomedical engineering. He has carried out
research and taught at the University of Manitoba and Memorial University in Canada. In addition to his technical
research interests, he earned a diploma in Higher Education Teaching from the University of Manitoba, Canada.
He has published over 10 papers in the Canadian Engineering Education Association. Currently, he is a Senior
Lecturer at the University of Technology, Papua New Guinea.

Kamalakanta Muduli is presently working as Associate Professor in Department of Mechanical Engineering, Papua
New Guinea University of Technology, Lae, Morobe Province, Papua New Guinea. He has obtained PhD from
School of Mechanical Sciences, IIT Bhubaneswar, Orissa, India, in the area of Green Supply Chain Management.
He has obtained Master’s degree in Industrial Engineering. He has over 14 years of academic and research
experience. He has published 27 papers in peer reviewed international journals and more than 21 papers in
National and International Conferences.

66
View publication stats