Vol. 5, No. 9, September 2024
E-ISSN:2723 – 6692
P-ISSN:2723– 6595
http://jiss.publikasiindonesia.id/
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2191
Taguchi Approach to Defect Analysis on Electric Motor
Conversion at PT. Electric Vehicle Trimotorindo
Hendri Simon Siregar, Prihantoro Syahdu Sutopo
Universitas Buddhi Dharma, Indonesia
Email: hendrisimons427@gmail.com, prihantoro.[email protected]m
Corespondence: hendrisimons4[email protected]m*
KEYWORDS
ABSTRACT
Production Process; Taguchi;
QC
Maintaining product quality can reduce the risk of defective
products. The survival of a company is heavily reliant on product
quality. In order to improve performance and reduce defects in the
product process, PT. Electric Vehicle Trimotorindo employs the
Taguchi approach in conjunction with the QC system. The goal of
this research is to determine the product process of PT.
Trimotorindo Electric Vehicle, identify the highest rejection rate,
evaluate process capability, and propose improvements using the
Kaizen system. This study uses the Taguchi system to optimize the
product process of PT. Trimotorindo Electric Vehicles. After
collecting data from the taguchi system, it can be concluded that
the highest position of blights in a conversion process is in the
electric current string.
Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
Introduction
Indonesia is one of the most populous countries in Asia, and it has the opportunity to create
more jobs and support economic development. Electric motors are essential components in a
variety of industrial, vehicle, and household appliance applications. Quality control involves
identifying defects or non-conformities in the production process and implementing corrective
actions to minimize or eliminate those defects (Aldi & Rahmatullah, 2023; Tirtayasa et al., 2021).
The implementation of Industry 4.0 has enabled a flexible manufacturing system capable of
producing different types of products at a lower cost (Durakovic & Halilovic, 2023; Guo et al., 2021;
Javaid et al., 2022; Teja et al., 2022). The objectives of this study include analyzing defects using the
Taguchi method that affect the production process in the conversion of electric motors and looking
for factors that cause defects in electric motor components.
The high-efficiency motor design reduces friction and heat generation during operation. The
materials used also affect the performance of the engine. For example, using magnetic materials
with high magnetic permeability can improve motor performance. Operating conditions, such as
machines operating at heavy loads or high temperatures, can also affect the machine's efficiency
(Germann et al., 2021; Hasan et al., 2022; Quan et al., 2021).
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2192
Materials and Methods
The Taguchi method is a new methodology in the field of engineering that aims to improve the
quality of products and processes and to reduce costs and resources to a minimum. This method is
very effective for quality improvement and cost reduction, improvement in product manufacturing
and reduction of product development costs (Setyo Pradana and Sulistiyowati 2022).
Taguchi's definition of quality is the loss received by the public since the product was shipped.
The Taguchi concept is to improve the quality of manufactured goods by designing products or
processes before they reach the production stage (Pramudita et al., 2022; Saleem et al., 2022;
Sanjeevannavar et al., 2022).
Results and Discussions
From the results of observations and interviews with related parties at PT. Electric Vehicle
Trimotorindo finally obtained information about the process of corruption of electric motorcycles,
from the results of interviews and direct observations in the field, in the end it resulted in an electric
motorcycle data and not too many people have explored the process of corruption from gasoline or
conventional motorcycles to electric engines, therefore the author explores about electric
motorcycles in the field of corruption.
Fishbone identification
This identification took from literature reviews and interviews with stakeholders, particularly
in the Quality Control (QC) department and the electric motor conversion section, resulting in the
identification of factors that affect quality characteristics. From the results of the observation of the
data on the number of defects obtained in the 5-day observation time of the electric motor
conversion process, it can be seen in the table below:
Figure 1 Fishbone diagram
The results of the observation of the data on the number of defects obtained in the 5-day
observation time of the electric motor conversion process it can be seen in the table
Table 1 Total of damages per day
No
Day
Total of Damage
1
Monday
8
2
2
Tuesday
7
2
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2193
3
Wednesday
9
3
4
Thursday
6
5
5
Friday
5
10
From this data, it can be concluded that the most damage is on Wednesday can be seen in the
diagram below:
Figure 2 diagram of the amount of damage
Determining factor levels
The level setting is determined by considering the company's current operational limitations
and potential changes to ensure the test results are as realistic as possible. The level settings
specified for each control factor are described below.
Table 2 Design Variables
Design Variable
Variations
Level 1
Level 2
Level 3
Cable diameter
1 mm
2,5 mm
3 mm
Types of socket sizes
4 pin holes medium
socket type
6 large socket-type pin
holes
3 small socket type pin
holes
Cable length
2 meters
3 meters
1 meter
Waktu wiring
15 minutes
25 minutes
10 minutes
Determination of orthogonal array matrix
The determination of the orthogonal array (OA) matrix in the Taguchi method is one of the
important steps in the experimental design process (Mensah et al., 2019). OA is used to determine
the level combination of factors to be tested. Table 4 is a calculation of the degree of freedom of
control factors in this study. This experiment uses four factors in a three-level design. The number
of columns in an orthogonal matrix can be determined from the number of levels and factors
present. Using orthogonal array matrix analysis, the orthogonal array calculation is obtained as
follows:
Table 3 alculation of the degree of freedom of the orthogonal matrix Array
Code
Factor
Explanation
DF
A
Cable Diameter
(3-1)
Total of damages per day
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2194
B
Types of socket sizes
(3-1)
C
Cable length
(3-1)
D
Wiring time
(3-1)
Total
8
From the calculation table above, it can be seen that the factor of this writing is eight (8) To
find the degree of freedom of the orthogonal array, multiply the degree of freedom of each column
by the number of columns. Based on the explanation in Table 8 This study should have been carried
out with an orthogonal arrangement =, but because the Taguchi experiment did not show its
existence, the number of orthogonal arrangements required was increased to = to carry out the
research according to Taguchi's rules as illustrated in the Table below :
Table 4 Standard orthogonal array matrix from taguchi
2 Level
3 Level
4 Level
5 Level
Level gabungan
L9 (34)
L18 (21x37)
(27)
L27 (311)
L32 (21x39)
(211)
L81 (340)
L36 (211x312)
L36 (23x313)
L54 (21x323)
Source: (Gao, Xu, and Xu 2022)
Implementation of Taguchi's Design of Experiment (DOE) Calculation
It is carried out based on the results of calculating the degree of freedom. The result of the
calculation is that the orthogonal matrix used in this experiment is =. While the experimental data
are:
Table 5 Taguchi Design of Experiment (DOE) Calculation
Eksperimen
Control Factors
Result
A
B
C
D
I
II
III
1
1
1
1
1
1.15
0.04
0.02
2
1
2
2
2
1.25
0.06
0.03
3
1
3
3
3
1.10
0.03
0.01
4
2
1
2
3
2.10
0.04
0.03
5
2
2
3
1
2.15
0.06
0.01
6
2
3
1
2
2.25
0.03
0.02
7
3
1
3
2
3.25
0.04
0.01
8
3
2
1
3
3.10
0.06
0.02
9
3
3
2
1
3.15
0.03
0.03
The number of rows indicates the number of experiments performed. This experiment was
repeated three times (repeated experiments), so it required a total of 9 attempts to be carried out.
The level code and its value are shown in table 5. From the calculation using the equation, the
average and SNR tables from the 1st to the 9th data are obtained as follows:
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2195
Table 6 Average
and SNR
No
Y1
Y2
Y3
mean
1
1,15
0,04
0,02
0,4033333
30,17
2
1,25
0,06
0,03
0,4466667
26,65
3
1,10
0,03
0,01
0,38
35,68
4
2,10
0,04
0,03
0,7233333
27,62
5
2,15
0,06
0,01
0,74
35,34
6
2,25
0,03
0,02
0,7666667
30,8
7
3,25
0,04
0,01
1,1
35,49
8
3,10
0,06
0,02
1,06
29,66
9
3,15
0,03
0,03
1,07
28,69
The data above will be analyzed in four ways: calculating the effect of the mean, calculating
the SNR effect and calculating for each replication, and calculating the variant analysis (ANOVA).
Calculating the average impact of these factors is done by subtracting the average of the largest
responses from the average of the largest responses. The results of the calculation of the average
influence value and the influence value of each factor are obtained in Table 6.
Table 7 Response Table for Means
Level
Cable diameter
Socket Type
Cable Length
Wiring time
1
0,4100
0,7422
0,7443
0,7378
2
0,7433
0,7489
0,7467
0,7711
3
1,0767
0,7389
0,7400
0,7211
Delta
0,6667
0,0100
0,0067
0,0500
Rank
1
3
4
2
Calculation of the effect of SNR
The calculation of the effect of SNR on factors is done by subtracting the average value of the
largest response from the average value of the largest response. So that the average response value
and SNR effect value are obtained in the table below:
Table 8 Calculation of the effects of SNR
Level
Cable diameter
Socket Type
Cable Length
Wiring time
1
-4,085
-4,338
-4,341
-4,326
2
-4,392
-4,421
-4,239
-4,302
3
-4,511
-4,214
-4,409
-4,360
Delta
0,426
-4,437
0,170
0,058
Rank
1
3
4
2
Calculate the effect of each factor for each replication
The first step of this calculation is to find the average response of each factor level for each
replication. The calculation of the response for each level of each factor using the equation from the
calculation using the equation above, the response table of each factor for each replication is
obtained as follows:
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2196
Table 9 Calculation of the effect of each factor for each replication
Factor
Y1
Y2
Y3
Rangking
A1
30,83
31,25
31,28
1
B1
31,09
30,55
30,55
3
C1
30,21
27,65
27,65
4
D1
31,4
30,98
30,98
2
S/N Prediction
S/N Exiting
Parameter
S/N
Parameter
S/N
Average S/N
31,12
Average S/N
31,12
A1/A2/A3
31,28
A1
30,83
B1/B2/B3
31,72
B2
30,55
C1/C2/C3
35,5
C2
27,65
D1/D2/D3
31,4
D2
30,98
Total
36,54
total
26,65
Prediction
Exiting Design
26,65
Optimum Design
36,54
gain
9,89
From the calculation of the average response to the replication, we can determine the effect of
each factor for each replication, namely by subtracting the average value of the largest response by
the average value of the smallest response.
ANOVA variant analysis calculation
From the data of the experiment results to find out the contribution of each controlled factor,
then a variant analysis was carried out with the following calculation results:
Table 10 Analysis of Variance
Source
DF
Sum
Square
%
Adj SS
Mean
Square
F-Value
P-Value
A
2
2,0000
5,99%
2,0000
1,00000
0,57
0,573
B
2
0,0005
0,00%
0,0005
0,00023
0,00
1,000
C
2
0,0002
0,00%
0,0002
0,00010
0,00
1,000
D
2
0,0117
0,03%
0,0117
0,00583
0,00
0,997
Error
18
31,3765
93,97%
31,3765
1,74314
Total
26
33,3888
100,00%
Table 10 provides a comparative summary of the four calculation methods that have been
carried out. So that a ranking table of the influence of each factor can be obtained as follows:
Table 11 Ranking of the Influence of Each Factor
Rangking
Mean
SNR
Repikasi
Anova
1
A
A
A
A
2
D
B
D
D
3
B
C
B
B
4
C
D
C
C
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2197
From Table 11 ANOVA above shows that by comparing the f-value of each factor and
interaction with a value of 0.05, we can find out the factors or interactions that significantly
influence the defect rate in the electric motor assembly line. If the f-value of the factor or interaction
is more than 0.05, then it can be concluded that the factor or interaction has a significant influence
on the response variable .
Conclusion
As a result of the report and the data processing process, several conclusions can be made: We
can see the largest error rate in the production process from Table 13 of ANOVA by comparing the f-
value of each factor and interaction with the value of 0.05, we can find out the factors or interactions
that have a significant influence on the rate of defects in the assembly of electric motors that occur
in electric motor cables. The types of defects that are produced from the assembly process are
often found in socket products and pin cables on sockets that are detached from the housing. The
pin bit is not connected to the cable or the pin eye is separated from the socket housing which is
caused by negligence during the process when checking the electrical cable. The quality of the
electrical socket decreases due to the long delivery process and is not properly placed so that it
experiences an impact.
References
Aldi, I. D., & Rahmatullah, A. (2023). Analisis Pengendalian Kualitas Produk Sepatu Adida Dengan
Metode DMAIC dan FMEA DI PT. Parkland World Indonesia-Cikande. Jurnal Taguchi: Ilmiah
Teknik Dan Manajemen Industri, 3(1), 142–148.
Durakovic, B., & Halilovic, M. (2023). Thermal performance analysis of PCM solar wall under
variable natural conditions: An experimental study. Energy for Sustainable Development, 76,
101274.
Gao, G., Xu, F., & Xu, J. (2022). Parametric Optimization of FDM Process for Improving Mechanical
Strengths Using Taguchi Method and Response Surface Method: A Comparative Investigation.
Machines, 10(9). https://doi.org/10.3390/machines10090750
Germann, T., Martin, D. M., Kubik, C., & Groche, P. (2021). Mastering Uncertain Operating Conditions
in the Development of Complex Machine Elements by Validation Under Dynamic Superimposed
Operating Conditions (pp. 236–251). https://doi.org/10.1007/978-3-030-77256-7_19
Guo, D., Li, M., Lyu, Z., Kang, K., Wu, W., Zhong, R. Y., & Huang, G. Q. (2021). Synchroperation in
industry 4.0 manufacturing. International Journal of Production Economics, 238, 108171.
https://doi.org/10.1016/j.ijpe.2021.108171
Hasan, I., Hakim, L., & Denur. (2022). Desain Pengganti Penggerak Motor Bakar Torak (110 CC) pada
Sepeda Motor Otomatic dengan Motor Listrik Type Bldc (Brushless DC). Jurnal Surya Teknika,
9(2), 516–524. https://doi.org/10.37859/jst.v9i2.4382
Javaid, M., Haleem, A., Singh, R. P., & Suman, R. (2022). Enabling flexible manufacturing system
(FMS) through the applications of industry 4.0 technologies. Internet of Things and Cyber-
Physical Systems, 2, 49–62. https://doi.org/10.1016/j.iotcps.2022.05.005
Mensah, R. A., Kirton, S. B., Cook, M. T., Styliari, I. D., Hutter, V., & Chau, D. Y. S. (2019). Optimising
poly(lactic-co-glycolic acid) microparticle fabrication using a Taguchi orthogonal array design-
of-experiment approach. PLOS ONE, 14(9), e0222858.
https://doi.org/10.1371/journal.pone.0222858
e-ISSN: 2723-6692 🕮 p-ISSN: 2723-6595
Journal of Indonesian Social Sciences, Vol. 5, No. 9, September 2024 2198
Pramudita, D. T., Gunawan, N. F., Ningsih, M. C., & Adilah, R. (2022). Determinasi Kepuasan
Pelanggan Dan Loyalitas Pelanggan: Harga Dan Kualitas Produk (Literature Review
Manajemen Pemasaran). Jurnal Manajemen Pendidikan Dan Ilmu Sosial, 3(1), 424–436.
Quan, Z., Ge, L., Wei, Z., Li, Y. W., & Quan, L. (2021). A Survey of Powertrain Technologies for Energy-
Efficient Heavy-Duty Machinery. Proceedings of the IEEE, 109(3), 279–308.
https://doi.org/10.1109/JPROC.2021.3051555
Saleem, S., Haider, H., Hu, G., Hewage, K., & Sadiq, R. (2022). Continuous performance improvement
of aquatic centres: A Taguchi-based optimization approach towards sustainability. Journal of
Building Engineering, 54, 104576. https://doi.org/10.1016/j.jobe.2022.104576
Sanjeevannavar, M. B., Banapurmath, N. R., Soudagar, M. E. M., Atgur, V., Hossain, N., Mujtaba, M. A.,
Khan, T. M. Y., Rao, B. N., Ismail, K. A., & Elfasakhany, A. (2022). Performance indicators for the
optimal BTE of biodiesels with additives through engine testing by the Taguchi approach.
Chemosphere, 288, 132450. https://doi.org/10.1016/j.chemosphere.2021.132450
Setyo Pradana, E., & Sulistiyowati, W. (2022). Literature Review: Use of the Taguchi Method for
Quality Improvement. PROZIMA (Productivity, Optimization and Manufacturing System
Engineering), 6(2), 85–96. https://doi.org/10.21070/prozima.v6i2.1575
Teja, S., Ahmad, A., & Salomon, L. L. (2022). Peningkatan Kualitas Produksi Pakaian Pada Usaha
Konveksi Susilawati Dengan Berbasis Metode Six Sigma. Jurnal Ilmiah Teknik Industri, 10(1), 9–
20.
Tirtayasa, S., Lubis, A. P., & Khair, H. (2021). Keputusan pembelian: sebagai variabel mediasi
hubungan kualitas produk dan kepercayaan terhadap kepuasan konsumen. Jurnal Inspirasi
Bisnis Dan Manajemen, 5(1), 67–86.
