IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS

Optimization preventive maintenance of equipment production on palm

oil mill
To cite this article: Ina Siti Hasanah and Ratih Wulandari 2019 IOP Conf. Ser.: Mater. Sci. Eng. 505 012155

View the article online for updates and enhancements.

This content was downloaded from IP address 181.214.188.55 on 05/07/2019 at 02:23

1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

Optimization preventive maintenance of equipment production on

palm oil mill

1
Ina Siti Hasanah
2
Ratih Wulandari
1,2
Industrial Engineering Departement, Faculty of Industrial Technology University Gunadarma
Jl. Margonda Raya no.100, Depok, Jawa Barat, Indonesia
inash@staff.gunadarma.ac.id
ratih_wulandari@staff.gunadarma.ac.id

ABSTRACT
Maintenance is an important issue that must be considered, especially in the agroindustry company. This
paper is to identify and prioritize the most critical component in the production of palm oil using FMEA, and schedule
maintenance with optimal total cost of maintenance. RPN obtained score from the highest to the lowest to be scheduled
maintenance that is a component of the worm screw press machine, V-belt thresher drum equipment, V-belt separator
clarification equipment, steel cable sling tracklier equipment, door seals equipment sterilizer, wheel lorry with ad on lorry
equipment, V-belt boiler, screw pn clarification separator equipment, friction pad clarification separator equipment and V-
belt equipment at the seed stasion. Optimization of the total cost of maintenance is the reliability level of 90% at a cost Rp.
131.443.940,2/year. Scheduling maintenance can save as much as 47,4% compared to no maintenance scheduling.

1. INTRODUCTION

Maintenance is all action or a combination of the various activities carried out in order to maintain or restore an
equipment/ machinery on conditions acceptable to earer (Ricardo et all, 2009; Adolfo C., 2007; Ebeling EC,
1997). Maintenance has become the common issues and important faced by the industry and is one of the factors
that determine the productivity of the industry.
Sharma et al. (2008) conducted a study that FMEA (Failure Mode and Effect Analysis) to identify and
prioritize the causes of failure in a system. FMEA is a method that can be used to identify the critical components
that have failed resulting in undesirable events such as loss of production. The goal is to determine and prioritize
the potential failure modes by counting each RPN (Risk Priority Number) which may have detrimental effects on
a system.
Combination of Petri net and techniques FMECA (Failure Mode Effects and Criticality Analysis) is a new
solution for modeling complex industrial system reliability (Bertolini, M., et al. 2006). Fuzzy modeling FMECA
can calculate the index of fuzzy failure mode, by using Mamdani type fuzzy inference mode and implications min
max aggregation (Zafiropoulos and Dialynas, 2005).
Maintenance system plays an important role in achieving corporate goals. It is instrumental in reducing
equipment downtime, improve quality and increase productivity. Election strategy improper system maintenance
can affect the company's operating budget for unplanned maintenance costs will reduce the productivity and
profitability (Shyjith et al. 2008). Operating budget especially plant maintenance costs and component costs are
very high when compared to machines other processing costs (Pahan, 2008). It can be seen from the average
percentage of the cost of care palm oil mill equipment PKS X from 2006-2010, 30.21% is almost close to the
average cost of processing palm oil at 31.88%. The average percentage of the costs incurred for the maintenance

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

of equipment in 2010 amounted to 46.04% of the total cost of palm oil processing. Because of the magnitude of
the maintenance costs, the maintenance of the system is needed so that maintenance costs can be reduced.
According Kurniawan (2006), a total investment of palm oil mills Rp. 103.73 billion for the PKS with installed
capacity by 60 tonnes FFB / hour. The total investment is higher than the average cost of maintaining the
equipment PKS X 24% of the total investment of PKS. Production equipment maintenance budget submitted to
PKS X fell in stages ie 3 months in a year so that necessary preventive maintenance scheduling optimal
production equipment at minimum cost.
Ebeling (1997), perform preventive maintenance on 90% reliability rate. PKS X wants preventative
maintenance on 75%. This study aims to identify and define the components of the most critical production
equipment in the production of palm oil millers X. The next goal is to schedule the maintenance component of
the cost of maintaining the optimum.

2. METHOD
This study uses secondary data and primary data. Primary data were collected through interviews to obtain
information about the scheduling of maintenance activities of the field supervisors, maintenance chief, head of
production, plant managers, and maintenance staff.
Secondary data is data that is used all kinds of production equipment used, the components of the equipment
used, the failure interval, and time replacing failure components. Goodness of fit test methods used to examine the
distribution of component failures that are suitable for the equipment used. Bartlett's Test is used to test the
exponential distibusi, Mann's test was used for testing the Weibull distribution, and the Kolmogorov Smirnov Test
was used to test the normal and lognormal distributions. Having obtained a suitable distribution of failure, then
the value of MTTF (Mean Time To Failure) and MTTR (Mean Time To Repair) can be calculated by the
following formula (Ebeling, 1997; Ricardo et al, 2009):
1. Distribusi Eksponensial
MTTF = MTTR = 1 (1)

R(t) = e  t (2)
2. Distribusi Weibull
MTTF = MTTR = 1 (3)
1 
 
R(t) = e  (t / ) (4)
3. Distribusi Normal
MTTF = MTTR =  (5)
R(t) = 1   t    (6)
  
4. Distribusi Lognormal
2 /2
MTTF = MTTR = t med e s (7)

R(t) = 1   1 ln t  (8)
s 
 t med 
To prioritize the most critical component FMEA method is used to determine the risk priority number (RPN)
to perform the necessary corrective actions to prevent failures. Ben Power et al. (2009) using a numerical scale
ranging from 1 to 10 to rank the severity of the failure rate, the possibility of failure modes and the possibility of
failure is detected. Scale used ranges from 1 to 10, where 1 indicates very low scale and high scale 10 states.
Higher RPN values are given higher priority than the lower value of the RPN. RPN is calculated by multiplying
the rate of failure severity (S), the possibility of failure (O = Occurrence) and the probability of failure is detected
(D = Detection) to determine the risk level of a process, as follows:
RPN = S x O x D (9)
According to Sharma and Sharma (2010), Severity is related component failure effects mean time between
repair (MTTR) of each component, Occurrence rate of failure is derived from the average time between failures
(MTTF) of each components. Detection is the quantification of control that makes a failure can be detected and

2
1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

the results obtained from the interviews with the head of engineering maintenance section. Scheduling method
based on the failure replacement intervals specified level of reliability. If the lognormal distribution component
replacement intervals failure to the reliability level R:
1
tR = tmed e s( ( R)) (10)
If the weibull distribution component replacement intervals failure to the reliability level R:
 
tR =  ln( R) (1 / ) (11)
Reliability of the system after preventive maintenance is as follows (Ebeling, 1997):
Rm(t) = R(T)n x R(t-nT) (12)
where :
Rm(t) = reliability of systems with preventive maintenance
n = number of maintenance
t = time
T = time interval preventive replacement
R(T)n = probability of reliability to n intervals
R(t-nT) = probability of reliability for time t-nT of the last preventive maintenance actions.
Calculation of the total cost of preventive maintenance is as follows (Jardine, 2001):
Total cost = C p  R(t p ) C f  (1  R(t p )) (13)
t p  R(t p ) t f  (1  R(t p ))
In this study calculates the total cost of maintenance for all components in order to obtain optimal maintenance
costs, ie
n C pi  R(T pi ) C fi  1  R(T pi ) 
Minimization: Total cost =   T k (14)
i 1 T pi  R(T pi ) t fi  1  R(T pi ) pi pi

with constraints: 0< R(Tpi) < 1

Cpi , Cfi , tfi , Tpi > 0
i = 1, 2, 3, ...., n
where :
Cpi = cost of preventive cycles in the i-th component
= (cost technician x tpi + of i component costs)
Cfi = cost of failure cycles in the i-th component
= (cost technician x Tfi + of i component cost + cost of lost production x Tfi)
tfi = value of the i-th component MTTF
Tpi = time interval preventive maintenance on the i-th component
R(tpi) = probability of the expected reliability of the i-th component

3. RESULTS AND DISCUSSION

Based on analysis OEE (Overall Effectiveness Equipment Corporation), there are ten component in its eight
equipment which experiencing mischief at months August. Components-component which experiencing mischief
the said is components screw worms press machine, V-belt equipment thresher drums, V-belt equipment
clarification separator, cables sling fertilizer equipment tracklier, seal has door equipment sterilizer,, wheel truck
with a peace be at equipment lorry, V-belt boiler, screw pn equipment clarification separator, Friction pad
equipment clarification separator and V-belt equipment on the station seeds.
The first step is to choose the preventive maintenance scheduling which components will be prioritized for
scheduled, using FMEA (Failure Mode and Effect Analysis). Rating occurence is the quantification of the level of
failure events derived from the MTTF and the scale used ranges from 1-10. Rating of severity is the quantification
of the level of impact due to the occurrence of failure associated MTTR and scale used ranges from 1-10 (Pande,
et al. 2000). From the calculation of RPN values obtained from the highest to the lowest that will be scheduled
maintenance that worm screw press machine, V-belt thresher drum equipment, V-belt separator clarification
equipment, steel cable sling peralatam tracklier, seal stew door sterilizer equipment, wheel dolly + lori as
equipment, V-belt boiler, screw pn clarification separator equipment, friction pad separator clarification
equipment and appliances on the V-belt seed station.

3
1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

Table 1. Costs of failure and Cost Prevention in a single cycle

component technician Production component Tf = tp failure cost, preventive
cost loss cost cost (Rp) MTTR (hour) Cf (Rp) cost, Cp (Rp)
(Rp/hour) (Rp/hour) (hour)
Screw worm 10.000 525.000 100.000 6,53 6 3.592.471,2 160.000
V-belt thresher drum 10.000 525.000 50.000 5,57 5 3.027.335,6 100.000
V-belt separator 10.000 525.000 200.000 6,03 6 3.426.456,3 260.000
Kabel sling baja 10.000 525.000 150.000 4,75 4 2.690.742,3 190.000
Seal pintu rebusan 10.000 525.000 425.000 4,31 4 2.728.243,4 465.000
Roda lori+as 10.000 525.000 150.000 5,26 5 2.965.437,5 200.000
V-belt boiler 10.000 525.000 250.000 6,14 6 3.533.562,5 310.000
Screw pn separator 10.000 525.000 970.000 2,74 2 2.436.865,1 990.000
Friction pad separator 10.000 525.000 2.840.000 4,40 4 5.196.573,7 2.880.000
V-belt stasiun biji 10.000 525.000 92.000 6,42 6 3.526.351,8 152.000

Calculation of total cost of maintenance prior to preventive maintenance for one year shown in Table 2.
Based on Table 1 and Table 2, the components are calculated the total cost of preventive maintenance on a variety
of levels of reliability to obtain the optimal total cost of maintenance for the 10 components. Calculation of total
maintenance costs with the help of MS Excel. In Figure 1 can be seen the total cost of preventive maintenance is
the most optimal reliability level 90% at a total cost of preventive maintenance Rp 131443940.2 per year.
Table 2. Calculation of Total Failure Cost before preventive maintenance for one year
component atf = Cf (Rp) Tf = kf Tc (Rp/jam) expectation cost Total failure
MTTF MTTR (Rp) costyear/
(hour) (hour) (Rp/year)
Screw worm 1.071,1 3.592.471,2 6,53 8,2 3.333,6 3.570.709,8 29.201.981,7
V-belt thresher drum 1.818,4 3.027.335,6 5,57 4,8 1.659,7 3.018.098,9 14.539.355,7
V-belt separator 1.485,7 3.426.456,3 6,03 5,9 2.297,0 3.412.603,7 20.121.559,9
Kabel sling baja 1.185,9 2.690.742,3 4,75 7,4 2.260,0 2.680.009,6 19.797.436,7
Seal pintu rebusan 1.193,2 2.728.243,4 4,31 7,3 2.278,2 2.718.435,5 19.956.946,8
Roda lori+as 2.580,9 2.965.437,5 5,26 3,4 1.146,7 2.959.403,2 10.044.825,3
V-belt boiler 1.955,5 3.533.562,5 6,14 4,5 1.801,3 3.522.506,9 15.779.490,3
Screw pn separator 452,2 2.436.865,1 2,74 19,4 5.357,1 2.422.177,0 46.927.940,4
Friction pad separator 1.494,2 5.196.573,7 4,40 5,9 3.467,5 5.181.300,0 30.375.399,6
V-belt stasiun biji 709,6 3.526.351,8 6,42 12,3 4.925,0 3.494.736,5 43.143.013,0
Total 249.887.949,6

210.000.000,0
99%; 207.157.129,3

200.000.000,0

190.000.000,0

180.000.000,0
Total biaya
Rp/tahun

170.000.000,0

160.000.000,0

150.000.000,0

140.000.000,0 90%; 131.443.940,2

97%; 132.570.102,3
130.000.000,0
70% 75% 80% 85% 90% 95% 100%
Reliability

Figure 1. The total cost for the ten components of preventive maintenance for one year

Scheduling maintenance for the ten components at the level of 90% reliability can save as much as 47.4%.
After the scheduled preventive maintenance then the costs that affect the total cost of maintenance for the ten
components of the sensitivity analysis. Parameter sensitivity analysis conducted by engineers and component

4
1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

costs. The increase in the cost parameters technician will decrease by 50% to 88% level of reliability with optimal
total cost for all 10 components of Rp 135,567,151.30 per year. When the parameters of the component costs
increased by 16%, the rate dropped to 88% reliability with optimal total cost of maintenance for the 10
components of Rp 140935405.4 per year.
The decline parameter technician costs by 37% would raise the level of reliability of 97% to the total cost of
maintaining optimal for the 10 components of Rp 128361802.5 per year. When the parameters of the component
costs reduced by 13%, the rate increased to 97% reliability with optimal total cost of maintenance for the 10
components of Rp. 123649800.8 per year.

4. CONCLUSION
FMEA can be used to identify failures in the components and calculate the value of the RPN to prioritize
components. Priority consecutive components namely the worm screw press components, parts v-belt on drum
thresher, v-belt component in the separator, the component cables on tracklier steel slings, components of the
sterilizer door seals, wheel dolly + components as the lorry, the component v -belt on the boiler, screw pn
component in the separator, the component separator and friction pad on the v-belt components on seed station.
Preventive maintenance scheduling can improve the reliability of the components. Worm screw components
scheduled preventative maintenance every 617 hours, the component v-belt on drum thresher every 755 hours, the
component separator v-belt at every 727 hours, the components of steel cable sling on tracklier every 651 hours,
the door seal components sterilizer every 328 hours , components wheels and axles on lorry lorries every 604
hours, v-belt components boiler every 662 hours, the component separator screw pn at every 452 hours, the
separator pad friction components every 1,483 hours and v-belt component on the boiler station every 119 hours.
At the 90% level of reliability, maintenance scheduling the 10 components can save costs for companies operating
budgets of 47.4%. Increase in the cost parameters technician will decrease by 50% to 88% level of reliability.
Decrease in the cost parameters technicians will increase by 37% reliability rate of 97%. If the parameters of the
component costs increased by 16%, the rate drops to 88% reliability. When the parameters of the component costs
decreased to 13%, the rate increased to 97% reliability.
5. REFERENCES
[1] Adolfo C. 2007. The Maintenance Management Framework: Models and Methods for Complex Systems
Maintenance. Springer Series in Reliability Engineering.
[2] Ben-Daya, M., Duffuaa, S.O., Raouf, A., Knezevic, J., Ait-Kadi, D. 2009. Handbook of Maintenance
Management and Engineering. 1st Edition. Springer.
[3] Bertolini, M, Bevilacqua, M., Mason, G. 2006. Reliability design of industrial plants using Petri nets. Journal
of Quality in Maintenance Engineering. Vol.12 No.4. pp.397-411.
[4] Ebeling, E., C. 1997. An Introduction to Reliability and Maintainability Engineering. Mc.Graw Hill,
Singapore.
[5] Jardine. 2001. Maintenance Excellence Optimizing Equipment Life Cycle Decisions. Marcel Dekker. Inc.
[6] Kurniawan Ambar. 2006. Biaya Investasi Industri Kelapa Sawit. Potensi dan Peluang Investasi Industri
Kelapa Sawit di Indonesia. Pusat Penelitian Kelapa Sawit. Medan.
[7] Pahan, Iyung. 2008. Panduan Lengkap Kelapa Sawit. Cet.5. Jakarta: Penebar Swadaya.
[8] Pande, P.S., Neuman, R.P., Cavanagh, R.R. 2000. The Six Sigma Way: How GE, Motorola, And Other Top
Companies Are Honing Their Performance. Mc Graw-Hill. New York.
[9] Riccardo M., Alberto R. 2009. Maintenance for Industrial Systems. Springer. London.
[10] Sharma, R., Kumar, D. and Kumar, P. 2008. Predicting uncertain behavior of industrial system using FM - A
practical case, Applied Soft Computing, Vol. 8 No. 1, pp. 96-109.
[11] Sharma, R.K, Sharma P. 2010. System failure behavior and maintenance decision making using, RCA,
FMEA and FM. Journal of Quality in Maintenance Engineering, 16.1. p.64-88
[12] Shyjith K., Ilangkumaran K., Kumanan S. 2008. Multi criteria decision making approach to evaluate
optimum maintenance strategy in textile industry. Journal of Quality in Maintenance Engineering, Vol.14
No.4, p.375-386.
[13] Zafiropoulos, E.P., Dialynas, E.N. 2005. Reliability prediction and failure mode effects and criticality
analysis (FMECA) of electronic devices using fuzzy logic. The International Journal of Quality & Reliability
Management. Vol.22, Iss.2/3, 183-201.

5
1st International Conference on Industrial and Manufacturing Engineering IOP Publishing
IOP Conf. Series: Materials Science and Engineering 505 (2019) 012155 doi:10.1088/1757-899X/505/1/012155

Authors’ background
Your Name Title* Research Field Personal website

Ina Siti Lectur Maintenanc inash@staff.gunadarma.ac.

Hasana e e id
h

*This form helps us to understand your paper better,the form itself will not be published.

*Title can be chosen from: master student, Phd candidate, assistant professor, lecture, senior lecture, associate
professor, full professor