RESEARCH ARTICLE

Filtered EMG Signal
0.25

0.2

amplitude (mV) 0.15

0.1

0.05

00 0.5 1 1.5 2 2.5

sample x 104

Figure 5: Filtered EMG signal for bicep muscle contraction

Table 2: Angular separation result of 17 features extracted from EMG signal

Feature Angular Separation Angular Separation Result
Result Between Between Different Trial
IEMG Different Subject for a subject
RMS 0.64 0.70
WL 0.64 0.70
SSI 0.62 0.60
MAV 0.51 0.63
MMAV1 0.64 0.70
MMAV2 0.64 0.70
AR 0.62 0.72
TTP 0.82 0.87
MNF 0.51 0.62
MDF 0.77 0.71
MMNF 0.51 0.62
FR 0.76 0.65
MinWT 0.63 0.39
MaxWT 0.98 0.97
MeanWT 0.46 0.46
SDWT 0.91 0.95
0.47 0.49

43

5. CONCLUSION

This research paper presents method to determine optimum feature

extraction of sEMG signals acquired from five healthy subjects performing

fundamental arm movement. Seventeen popular feature extractions from time

domain, frequency domain and time frequency domain have been tested. By

using angular separation technique, the result shows that MinWT gives the

optimum result and can be used for further upper limb rehabilitation activity.

These two results will now be used in further research work of upper limb

rehabilitation such as in developing rehabilitation system using virtual reality

environment.

REFERENCES

1. "Stroke in Malaysia: What is a stroke?," Retrieved from
http://www.nasam.org/english/prevention-what_is_a_stroke.php , 2017.

2. W. Chuanchu, P. Kok Soon, A. Kai Keng, G. Cuntai, Z. Haihong, L. Rongsheng, K. S.
G. Chua, A. Beng Ti, and C. W. K. Kuah, "A feasibility study of non-invasive motor-
imagery BCI-based robotic rehabilitation for Stroke patients," in 4th International
IEEE/EMBS Conference on Neural Engineering, 2009., 2009, pp. 271-274.

3. G. S. Rash, “Electromyography Fundamentals,” Retrieved from
papers3://publication/uuid/F8E3A390-2C75-42E6-9F0C-768A97087B38 , 2008.

4. Phinyomark, A. , Phukpattaranont, P. & Limsakul, C (2012). Feature reduction and
selection for EMG signal classification. International Journal of Physical Sciences.
39(1), 7420-7431.

5. Wan Daud, W. A. B., Yahya, A. B. , Horng, , C. S., Sulaima, M. F. & Sudirman, R.
(2013). Feature Extraction of Electromyography Signals in Time Domain on Biceps
Brachii Muscle. International; Journal of Modeling and Optimization. 3(6), 3-7.

6. Fernando D Farfán, Julio C Politti and Carmelo J Felice, “Evaluation of EMG
processing techniques using Information Theory,” published in BioMedical
Engineering OnLine, 2010, DOI: 10.1186/1475-925X-9-72

7. C. J. DeLuca. (1997) Surface Electromyography Detection and Recording
[Online]http://www.delsys.com/library/tutorials.htm

8. S. Kumar and A. Mital, Electromyography in Ergonomics. London, U.K.: Taylor and
Francis, 1996.

9. J. S. Lee. (1998, Oct.) “EMG page”. [Online] Available:
http://home.earthlink.net/~leejs/EMG.html

44

10. Dinesh Kant Kumar, Nemuel D. Pah, and Alan Bradley, “Wavelet Analysis of Surface

Electromyography to Determine Muscle Fatigue,” published in IEEE

TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION

ENGINEERING, 2003, Vol. 11, ISSN Print: 1534-4320.

11. M.H. Pope, et. al., “Evaluation of low back muscle surface EMG signals using

wavelets,” published in Clinical Biomechanics, 2000, Vol. 15,

https://doi.org/10.1016/S0268-0033(00)00024-3.

12. F. A. Mahdavi, S. A. Ahmad, M. H. Marhaban, M. R. Akbarzadeh-T, “Surface

Electromyography Feature Extraction Based on Wavelet Transform,” International

Journal of Integrated Engineering, Vol. 4, No. 3, 2012, pp. 1-7.

13. Sachin Sharma, Gaurav Kumar, Sandeep Kumar, Debasis Mohapatra, “Techniques for

Feature Extraction from EMG Signal,” published in International Journal of Advanced

Research in Computer Science and Software Engineering, 2012, Vol. 2, ISSN: 2277

128X.

14. A. Phinyomark, S. Hirunviriya, C. Limsakul, P. Phukpattaranont, “Evaluation of EMG

Feature Extraction for Hand Movement Recognition Based on Euclidean Distance and

Standard Deviation,” International Conference on Electrical Engineering/Electronics

Computer Telecommunications and Information Technology (ECTI-CON), 2010, Vol.

1, Issue 1, 2009, ISSN: 2151-9617.

15. Nizam Uddin Ahamed, K. Sundaraj, R. B. Ahmad, S. A. M. Matiur Rahaman, M. A.

Islam, M. A. Ali, “Variability in surface electromyography of right arm biceps brachii

muscles between male adolescent, vicenarian and tricenarian with distinct electrode

placement,” published in IEEE Conference on Sustainable Utilization and Development

in Engineering and Technology (STUDENT), 2012, ISSN Print: 1985-5753.

16. R. Suhaimi, Aswad A.R, Nazrul H. ADNAN, Fakhrul Asyraf, Khairunizam WAN, D.

Hazry, Shahriman AB, Juliana A. Abu Bakar, Zuradzman M. Razlan, “Analysis of

EMG-Based Muscle Activity for Stroke Rehabilitation,” 2nd International Conference

on Electronic Design (ICED 2014), pp.167-170, 19-21 August 2014, Penang, Malaysia.

17. A. Phinyomark, A. Nuidod, P. Phukpattaranont, and C. Limsakul, “Feature Extraction

and Reduction of Wavelet Transform Coefficients for EMG Pattern Classification,”

Elektronika IR Elektrotechnika, Vol. 122, No. 6, 2012, ISSN Print: 1392 – 1215, ISSN

Online: 2029 – 5731.

18. A. Phinyomark, C. Limsakul, and P. Phukpattaranont, “A Novel Feature Extraction for

Robust EMG Pattern Recognition,” Journal of Computing, Vol. 1, Issue 1, 2009, ISSN:

2151-9617

19. K. Mahaphonchaikul, D. Sueaseenak, C. Pintavirooj, M. Sangworasil, S.

Tungjitkusolmun, “EMG Signal Feature Extraction Based on Wavelet Transform,” The

2010 ECTI International Conference on Electrical Engineering/Electronics, Computer,

Telecommunications and Information Technology, 2010, ISBN Electronic: 978-1-

4244-5607-9.

20. Kardi Teknomo, “Angular Separation,” Retrieved from

http://people.revoledu.com/kardi/tutorial/Similarity/AngularSeparation.html, 2015.

45

21. M. G. Benedetti, V. Agostini, M. Knaflitz and P. Bonato, “Applications of EMG in
Clinical and Sports Medicine: Muscle Activation Patterns During Level Walking and
Stair Ambulation,” book edited by Catriona Steele,2012, ISBN 978-953-307-798-7.

[Publication from Journal of Telecommunication, Electronic and Computer Engineering JTEC]
46

2018 M.H Mohd Hashim et al.

Comparative Measurement of RF Link Budget
between SUI and Theoretical Models in a Densely

Populated Area

Mohd Hasruzairin Bin Mohd Hashim*, Ir. Dr Khairul Nizam Bin Puniran**,
Mohd Hazman Bin Ibrahim*#

*Department of Industrial Mechatronics,
Kolej Kemahiran Tinggi MARA Balik Pulau,

**Petrofac Engineering Services,
*#Telekom Malaysia, Jalan Bangsar

[Email : [email protected]]

Abstract – This research presents the analysis of the RF link budget
measurement performed for LTE network in a densely populated area (i.e.
Kuala Lumpur) by comparing the result of propagation between theoretical
model and Standard University Interim (SUI). The aim of this paper is to
establish assumptions of system design at the eNodeB (transmitter) and the
user equipment (UE) for both uplink and downlink transmissions.
Consequently, the maximum allowable path loss (MAPL) in the LTE system
and the maximum coverage area based on the above propagation model is
determined and estimated accordingly.

[Keywords: LTE, Link Budget, SUI Model, MAPL, Link Budget, eNodeB, UE, Free Space
Loss, EIRP]

1. INTRODUCTION

The MCMC is recently granted a license to service provider to operate
LTE network in 3GPP Band 7, with consideration of uplink frequency in
between 2500 MHz to 2570 MHz, and downlink frequency at the range of

47

2620 MHz to 2690 MHz. The LTE will be in operation in year 2013. This
paper is to establish the RF link budget and to determine the maximum
allowable path loss (MAPL) in LTE and estimate the maximum coverage area
due to propagation model for the site.

A. Problem Statement
All gains and losses at transmitter (eNodeB) and receiver (UE) for

both uplink and downlink transmissions shall be considered in the RF link
budget calculation. From the link budget’s tabulated result for both uplink and
downlink as shown in this paper, the value of calculated MAPL will
determine the maximum coverage area based on the chosen propagation
model. The MAPL is determined based on the difference of maximum RF
power output of the transmitter and the maximum sensitivity of the receiver5.
The consequences of limiting the coverage area will cause an additional base
station which is required to cover the target area. Thus, MAPL is one of the
factors which involves in determining the performance of the model.

The SUI model is selected mainly due to the fact that the model has
been developed and benched marking for the frequency bands below 11
GHz3. In the US, this model is defined for Multipoint Microwave
Distribution System (MMDS) frequency band from 2.5 GHz to 2.7 GHz
which is suitable for the LTE network frequency in Malaysia, for both uplink
and downlink transmissions. The SUI model has better performance in
analytical process and has been widely used in propagation model.
Furthermore, the model also considers a different type of terrains with
different type of parameters and correction factors, which makes the result
more accurate to determine and design the RF link budget modelling for LTE
network3.

48

B. Assumption
In designing a cellular network which is involving the high speed data

link such as LTE network, there are two common questions which are the
governing points on how to determine the performance of the network. The
questions are how far can it go and what will the throughput be? There are
several factors that may impact the performance of the network specifically in
a wireless communication. Available and permitted output power, available
bandwidth, receiver sensitivity, antenna gains, radio technology, and
environmental conditions are some of the major factors that may impact
system performance.

A link budget involves a relatively simple addition and subtraction of
gains and losses within a RF link. When these component gains and losses are
determined and summed, the result is an estimation of end-to-end system
performance 4.

To arrive at an accurate answer, factors must be taken into account
such as:

 The frequency bands
 The uplink power
 Amplifier gain and noise factors
 Transmit antenna gain
 Path loss without assuming any relative error
 Receive antenna and amplifier gains and noise factors
 Other attenuation

In this research, the values and figures in relative to the above factors
have been tabulated in the methodology and result sections. The figures are
taken from various LTE products specification particularly the antenna for
both transmitter and receiver.

49

2. METHODOLOGY

A. Pathloss Model
The propagation model using theoretical model to determine the path

loss between eNodeB and UE is described in below equation. The path loss is
derived by the transmission path from an eNodeB to UE with the
consideration of all the losses1- 2.

PL (dB) = PTX + GTX – LTX – PRX – LRX – LM (1)

Where:

PTX = Transmitted output power (dBm)
GTX= Transmitter antenna gain (dBi)
LTX = Transmitter losses (dB)
PRX = Receiver power (dBm)
LM = Miscellaneous losses (dB)

The Free Space Path Loss (FSL) is a distinguished and prevailing loss
in cases where there are no obstacles along the path. The path loss according
to theoretical model (FSL) can be calculated by 1- 2.

LFSL (dB) = 32.44 + 20log10d (km) + 20log102600 (MHz) (2)

Where:

f = Operating frequency
d = Separation distance between eNodeB and UE

B. Stanford University Interim (SUI) Model
The propagation model selected in comparing the theoretical model as

mentioned in Section (2.A) is Standard University Interim (SUI) model. In
this paper, a Terrain A which is associated with a maximum path loss and
moderate to a highly dense populated area is considered. The expression of

50

path loss propagation with correction factors according to SUI model is
shown as per below equation 1- 2.

PL SUI (dB) = A+ 10γ log10 ( ) + Xf + Xh + s for d > do (3)
Where:

d = the distance between the eNodeB and UE antennas
d0 = 100m
s = a log normally distributed factor that is used to account

for the shadow fading owing to trees and other clutter and

has a value between 8.2dB and 10.6dB. The other
parameters are defined as 1- 2.

A = 20log10(4 ) (4)
(5)
γ = a – bhb +
ℎ

Where:

hb = base station height above ground
a, b and c = constant values given in Table 1
γ = equal to the path loss exponent, for a given terrain type the

path loss exponent is determined by hb

Table 1: Terrain Types and Parameters for SUI Model3

Model Parameter Terrain A Terrain B Terrain C

a 4.6 4.0 3.6

b(m-1) 0.0075 0.0065 0.005

c(m) 12.6 17.1 20

The correction factor for the operating frequency above 2GHz and for

the receiver antenna height are defined in below equation 1- 2.

Xf = 6.0log10(20 0 0) (6)

51

The expression of type A and B terrain is determined per below

equation 3- 4.

Xh = -10.8log10(20ℎ0 0) (7)
While for type C terrain 3- 4

Xh = -20 log10(20ℎ0 0) (8)

Where:

f = Frequency (in MHz)
hr = Receiver antenna height

The shadowing correction, S is calculated using the equation 1- 2.

S = 0.65 (log f) 2 – 1.3(log f) + α (9)

Where:

α = 5.2dB for urban and suburban area
α = 6.6dB for rural area

To determine the maximum allowable path loss (MAPL), the EIRP has to be
calculated 1- 2.

E.I.R.PPTX + GTX – Loss 
Where:

PTX = Transmitter output power (dBm)
GTX = Transmitter antenna gain (dBi)

Thus, the MAPL equation as described below 1- 2. (11)

MAPL = EIRP + GRX – PTRmin – I – Co

Where:

GRX = Receiver antenna gain (dBi)
PTRmin = Receiver sensitivity (dBmW)
I = Interference margin (dB)
Co = Loss of cable (dB)

52

3. RESULT AND FINDING

The RF link budget for uplink and downlink with 2600 MHz of
carrier operating frequency are tabulated in below section. The results are
obtained based on the theoretical propagation model and SUI propagation
model.

A. Theoretical Propagation Model
Table 2 shows the result obtained for the transmitter and receiver data

with all parameters required in downlink and uplink link budget.
Assumptions have been made in calculating the link budget by taking into
consideration of antenna gain, height and all the losses. The losses may due
to the cable loss, medium loss or loss due to the building penetration.

Table 2: Link Budget Design Specification for LTE Network

Description LTE (2600 MHz)
Transmitter (eNodeB)
Tx (dBm) Downlink Uplink
Tx Antenna Gain (dBi)
Cable Loss (dB) 46 23
EIRP (dBm)
Receiver (UE) 18 0
UE Noise Figure (dB)
Thermal Noise (dB) 20
Receiver Noise Floor (dBm)
SNR (dB) 62 23
Receiver Sensitivity (dBm)
Interference Margin (dB) Downlink Uplink
Control Channel Overhead (%)
Rx Antenna Gain (dBi) 72
Body Loss (dB)
Max Allowable Path Loss (dB) -104.5 -118.4

-97.5 -116.4

-9 -7

-106.5 -123.4

41

20 0

0 18

00

163.5 163.4

53

B. SUI Propagation Model
Table 3 shows the result obtained from SUI model. Several

parameters have been assumed such as type A terrain, the base station height,
receiver height and distance between transmitter and receiver, d.

Table 3: Link Budget Design Specification for SUI Model

Parameter Downlink Uplink
π 3.1429 3.1429
100
do (m) 1000 100
d (m) 4.6 1000
0.0075 4.6
a 12.6 0.0075
b (m-1) 3.0E+08 12.6
c (m) 3.0E+08
C (m/s) 2.60E+09 2.60E+09
1.65 40
f (hz) 40 1.65
hr (m) 1.15E-01
hb (m) 1.15E-01 8.5
8.5
λ 80.74
80.74 12.22
s (dB) 4.615 0.68
0.68 18.35
A 33.30 230.52
γ 169.38
Xf
Xh
PLSUI (dB)

4. ANALYSIS AND DISCUSSION

A. Downlink Analysis
Figure 1 shows the comparison of path loss obtained by using SUI

model and theoretical model namely as Free Space Loss (FSL) for downlink
transmission.

54

Figure 1: Comparison of Path Loss using SUI Model and Theoretical Model
From the downlink graph, the estimated maximum distance coverage

area is approximately d = 0.7Km. However, for the ideal model, the MAPL is
doubled the difference by computing below equation:

MAPL =32.44 + 20log10d (km) + 20log102600 (MHz)
Then, d = 1374.1km
B. Uplink Analysis
Figure 2 shows the comparison of path loss obtained by using SUI
model and theoretical model namely as Free Space Loss (FSL) for uplink
transmission.

55

Figure 2: Comparison of Path Loss using SUI Model and Theoretical Model

From the uplink graph, the estimated maximum distance coverage
area is approximately d = 0.29Km. However, for the ideal model, the MAPL
is shown per below equation:

MAPL = 32.44 + 20log10d (km) + 20log102600 (MHz)
Then, d = 1364.6km

C. Discussion
The estimated link budget analysis is engineered and designed by

network engineer to investigate the maximum allowable path loss (MAPL)
between eNodeB and UE in any wireless technology specifically LTE.
MAPL is used to determine maximum allowable attenuation between
eNodeB and UE. The assumption for the case study is to take sample starting
from 100m up to 1000m. The assumptions made in determining the link
budget are as per below:

 The estimated isotropic radiated power (EIRP) to be 62dBm for
downlink and 23dBm for uplink

56

 The gains for transmitter antenna and receiver antenna are as per
Table 2

 Standard eNodeB height is 40m from the sea level and UE height
is 1.65m

 The MAPL obtained from standard LTE design is 163.5dB for
downlink while 163.4dB for uplink

The result of path loss for ideal model is compared with SUI model to
determine which equation gives more accurate result in relative to the
maximum coverage area. By using SUI model, maximum PL for downlink is
169.38dB while 230.52dB for uplink.

From the downlink graph shown in Figure 1, the maximum cell size
for LTE network coverage by using SUI model is 0.7Km (from eNodeB to
UE). In this case study that is implemented in Kuala Lumpur, about 268 cells
are required for LTE network. Besides, from the graph shown in Figure 2, the
maximum coverage for uplink is 0.29Km (from UE to eNodeB). As for the
comparison, the coverage area for downlink is higher than uplink. It is due to
the fact that the UE power transmitter is lower than the eNodeB power
transmitter. The UE is required a small amount of power to transmit the
signal.

5. CONCLUSION
Most of service providers in cellular technology are implementing

cell splitting and frequency reuse techniques in order to have a better
coverage and to double the capacity. However, careful consideration and
technique shall be employed in order to establish a good design by
minimizing and eradicating issues especially the interference issue. As an
initial part of the design, the MAPL and maximum coverage area in the

57

network shall be determined.
In this research, the aim is to develop the assumptions of system

design for uplink and downlink transmission at transmitter and receiver sides.
A key part of the work is to find the MAPL and the maximum coverage area
for signal transmission in LTE network. Standard University Interim (SUI)
model at densely populated area with maximum path loss is considered in
this work. From the design obtained, it is found that SUI model has greater
path loss as compared with theoretical model.
REFERENCES

1. V.S. Abhayawardhana, I.J. Wassell, D. Crosby, M.P. Sellars and M.G. Brown.
"Comparison of Empirical Propagation Path Loss Models for Fixed Wireless Access
Systems", 2005.

2. J.Milanovic, S. R. Drlje and K.Bejuk. "Comparison of Propagation Models Accuracy
for WiMAX on 3.5 GHz", 2007.

3. H. Holma and A. Toskala, "LTE for UMTS; Evolution to LTE-Advanced, 2nd ed",
2011.

4. F. Khan, LTE for 4G Mobile Broadband, Cambridge University Press, 2009.
5. S. Sesia et al (eds.), LTE, The UMTS Long Term Evolution: From Theory to Practice,

Wiley, 2009.

58

2018 Yusup M.Z

Relationship between Manufacturing Sustainability
and Lean Production Practices by Malaysian
Manufacturers – A Preliminary Study

Dr. Muhamad Zaki Yusup
Department of Mould

Kolej Kemahiran Tinggi MARA Balik Pulau
[Email : [email protected]]

Abstract – A diversity of techniques and practices in Lean Production (LP) as
a tool for cultivating a culture of continuous improvement has successfully
produced a holistic approach in achieving a high level of sustainability in
manufacturing. Therefore, this study was undertaken to investigate the current
performance of manufacturing sustainability in manufacturing industry in
Malaysia, and the contribution of Lean in manufacturing sustainability. The
findings show that manufacturing industry in Malaysia has a high
sustainability in social competency, followed by economic and environmental
competency. Even though the environmental sustainability is ranked lowest,
the result reveals that aiming to have better control in environment and labour
safety is crucial in producing high quality products. The correlation analysis
reveals that a high competency in adapting Lean does not just increase the
manufacturing efficiency, but also has a substantial influence in producing a
sustainable product. This indicates that Lean is not just suitable in fulfilling
the manufacturing function, but has the potential to be adapted
comprehensively, primarily in achieving high manufacturing sustainability.

[Keywords: Manufacturing sustainability, Lean Production, Malaysia]

59

1. INTRODUCTION

Currently, the ability to achieve manufacturing sustainability that
covers elements such as environmental, economic and social competency is
very crucial in ensuring the continuity of a business in a new era of
manufacturing environment. The consideration of using new technology that
is consistent with a comprehensive manufacturing practice potentially
improves the level of manufacturing sustainability. This further encourages
the formation of new manufacturing management style that potentially raises
the ecological safety and good economic returns1. In addition, it also allows a
new paradigm in the manufacturing sector be developed through a
comprehensive transformation in managing a complex manufacturing
operations that cover changes in the behavior and technologies used2, 3.

The adaptation of Lean Production (LP) practices in achieving high
efficiency in production operation has a significant influence on each of the
elements needed to establish a sustainable manufacturing practice. The
integration of this practice can provide high added value in managing the
production operations in a more effective manner4. This eventually promotes
a more proactive approach in improving the quality and the sustainability of
the product in meeting the needs and opportunities in the future market.

Therefore, this initial study was conducted to investigate the relation
between the performance of manufacturing sustainability with the
performance of current practice of LP in Malaysia’s manufacturing industry
based on three components namely Sustainability in Environment (SEP),
Sustainability in Economic (SCP) and Sustainability in Social Competency
(SSC). The findings can be used in shaping a more comprehensive actions to
achieve high manufacturing sustainability. Besides that, the outcomes from

60

this study can be used by the manufacturer, particularly to identify the
initiatives that are required in enhancing LP in order to achieve high
sustainability in production operations. Moreover, academicians can also use
the evidences from this study to develop a new manufacturing framework to
strengthen the current practice in manufacturing operations. Organisation of
this article is based on the following sub-titles: research methods, result and
discussion, and conclusion.

2. RESEARCH METHOD

The analysis of this study was based on data obtained from
questionnaires. The questionnaires were distributed by mail to several
manufacturing industries in Malaysia. The respondents were all staff at
management level who have more than 2 years of experience in the same
organisation. The questionnaire comprises 25 items on LP, 19 items on
sustainability of environmental performance (e), 21 items on sustainability of
economic performance (c) and 23 items on sustainable of social competency
performance (s). For each item, the respondents were asked to rate each
performance based on a seven-point Likert’s scale (e.g.: 1=strongly disagree
to 7=strongly agree). From a total of 340 questionnaires, only 42 were
returned, at a response rate of 12.4 percent. As for the analysis, only 40
questionnaires on LP, 41 on SEP, 39 on SCP and only 39 for SSC were
considered valid for the analysis in this initial study.

61

3. RESULT AND DISCUSSION

The highest number of company ownership was by Malaysia (64.3%),
followed by other Asian countries (26.2%) while 9.5% are the US and Europe
companies. Nearly half of the respondents were from the mechanical
products group (42.9%). Others are from electric and electronic products
group (19%), automotive (16.7%), chemical (14.3%) and 7.1% are from
other product group. The majority of the respondents work for Malaysian
companies with less than 150 employees (35.9%), while the lowest numbers
of respondents were from the foreign companies that have more than 750
employees (17.9%). Meanwhile, 45.2% of the respondents have experience
with ISO14001 while 78.9% have more than 10 years of experience with this
management system. Furthermore, 73.6% of the respondents had obtained
other management certification such as ISO9001 (73.8%), OHSAS 18001
(11.9%), TS 16949 (21.4%) while 4.8% possess other management
certifications such as ISO 13485 and QS9000.

A. Manufacturing Sustainability Performance
The Cronbach’s alpha analysis shows that all items in the

questionnaires are consistently reliable at a value of 0.980 for SEP, 0.984 for
SCP and 0.979 for SSC. For SEP, the respondents agreed that all 19 items
have fulfilled the current practice where the ability to have better control of
environmental management (e19) has the highest mean score of 5.83. This is
followed by the ability to increase the practice, management and
environmental performance (e9), the action to reduce the waste of material
(e15), and the consideration of direct effects on the environment from
operation (e18) at a mean score of 5.71, 5.67 and 5.64 respectively. However,
the ability to adopt reusing and recycling the design of a product (e6) is still
less popular as this item gets the lowest score with a mean score of 4.90.

62

As for SCP, the ability to increase the quality of products (c5) has the
highest mean score value at 6.02, followed by the ability to reduce non-value
added activities (c6) and to increase added value activities (c7) at a mean
score value of 5.98 respectively. Respondents also agreed that they often
purchase materials, parts, and manage the resource required based on demand
(c16) and this reported a mean score value of 5.88. In contrast, the reduction
of the number of parts in the product (c20) has less influence as the
respondents appraised this with the lowest mean score value of 4.98.

SSC on the other hand shows that, high competency in complying
with environmental regulations and safety issues (s14), and the ability to
improve the housekeeping and safety of labour (s1) have the highest mean
score value at 6.24. This is followed by the claim that they have a holistic
environmental policy statement (s19) at a mean score value of 6.12. The
respondents also agreed that the consideration of workforce engagement in
the process (s17) and the consideration of current technology feasibility and
labour safety in the operation (s15) are also significant in SSC with a mean
score of 6.02 respectively. Meanwhile, collaboration with the communities,
government and non-government agencies regarding environmental issues is
at the lowest rank at a mean score value of 5.59. The dispersal of the mean
scores for all three sustainability components measured is shown graphically
in Figure 1.

63

(b) Sustainability of (c) Sustainability of (a) Sustainability of
Environmental Economic Social Competency

Performance (SEP) Performance (SCP) Performance (SCP)

Figure 1: The Performance of Components in Manufacturing
Sustainability

This result shows that generally the manufacturing industry in
Malaysia’s has a high focus on improving the environmental performance.
This is not surprising because the focus on dealing with environmental issues
is crucial to remain competitive in a modern manufacturing era. This is
evidently shown in Figure 1 where SSC has the biggest dispersal for mean
score compared to the other components measured. As for SCP, the
performance is influenced by the ability to increase the quality of product,
and reduction of non-value added activities. This subsequently leads to
reduction of non-value added cost of the production operation, and increase
the business and the environment5. This is in line with Awudu and Zhang6
where the ability to react in this situation can reduce the total production cost,
where the uncertainties of price in the market are closely monitored.
Although the mean score for SEP has the smallest dispersion, the results
show that the manufacturers are aware of the need to have better control in
environmental management. The current practice that focuses on reducing the
waste of material in production operation, potentially reduces the pollutants
produced, as stated by Hongbing and Haiyan7. This situation can generate

64

opportunities in business strategies, in producing a more sustainable product8.
This fact indicates that Malaysia’s manufacturing industry nowadays has a
good platform in achieving high sustainability in manufacturing, particularly
to remain competitive in a new manufacturing environment.

B. Relationship between Manufacturing Sustainability and Lean
Production
In this study, from a total of 1575 matrices of relationship created

from Spearman’s correlation analysis between all three components of
sustainability namely SEP, SCP and SSC with the LP, 32.8% of the matrices
has produced a strong positive correlation relationship ranging from 0.600 to
0.864, at a significant level of 0.01, as tabulated in Table 1. The analysis also
shows that, 19 items in the LP section have a strong relationship with SEP,
where a significant correlation relationship exists between the ability to
increase of opportunities in preventing pollution (e13) with the ability to
reduce the lead time in production operations (LP23) at a value of 0.757.

As for the relationship between SCP and LP, 96% of the items in LP
have produced a strong relationship with SCP, where 5 items in SCP and 4
items in the LP have a very strong relationship with each other at a value
ranging from 0.801 to 0.864. Furthermore, there seems to be a very strong
correlation at a value of 0.864 between the ability to relocate the resources
based on requirement (c13) with the ability to reduce the setup time (LP13).

65

Table 1: Spearman correlation between LP against Sustainability of
Environmental, Economic and Social Competency Performance

Sustainability of Sustainability of Economic Sustainability of Social
Item Environmental Performance (c) Competency

Performance (e) c3, c4, c7, c9, c10, c11, c12, Performance (s)
c13*, c14, c15, c21,
LP1 e4, e5, e9 c13, c15, c18, s5, s6

LP2 e9 c3, c15, c18, s1, s2, s4, s5, s6, t12, s15, s16
s1, s2, s3, s4, s5, s12, s15, s16,
LP3 - c1, c3, c4, c7, c9, c10, c11, c13, s17
c14, c15, c17, c18, c19, c21 s1, s2, s3, s4, s5, s6, s7, s8, s9,
LP4 e10, e11, e14, e16, e17, e19 s10, s11, s12, s13, s17, s20, s21,
c7, c11, c13, s23
LP5 e18, e19 s1, s3, s4, s5, s6, s15
LP6 e9, e18, e19 c3, c7, c11, c13, c14, s1, s2, s3, s4, s5, s6, s12, s15,
s16, s17
LP7 - c6, c7, c9, c11, c13, c14, c19,
c21 s2, s3, s4, s5, s6, s7, s8, s12, s21
LP8 - c3, c7, c9, c13, c14, c17, c18,
LP9 e9, e18 c19, c21 s2, s3, s4, s5, s6, s8, s9, s10, s11,
LP10 e9, e19 c3, c9, c21 s12, s17, s20, s21, s22
s3, s4, s5, s8, s16, s17
LP11 e1, e5, e8, e9, e11, e14, c1, c2, c3, c9, c11, c13, c21 s2, s3, s4, s5, s6, s8, s15, s16,
e15, e17, e18, e19 s17, s21
c1, c2, c3, c4, c7, c9, c10, c11, s2, s3*, s4, s5, s6, s7, s8*, s10,
LP12 e9, e18, e19 c12, c13, c14, c15, c16, c17, s11, s12, s13, s15, s17, s20, s21,
c18. c19, c20, c21 s23
LP13 e8, e9, e11, e14, e16, e17, c3, c4, c6, c7, c9, c10, c11, c13, s1, s2, s3, s4, s5, s6, s8, s9, s12,
e18, e19 c14, c19, c21 s14, s15, s16, s17
c1, c2, c3*, c4, c6, c7*, c8, c9*, s2, s3*, s4*, s5*,s6*, s7, s8* , s9,
LP14 e1, e3, e8, e9, e10, e14, c10, c11, c12, c13*, c14, c16, s10, s11, s12, s15, s17, s20, s21,
e16, e17, e18, e19 c17, c18. c19, c20, c21* s23
s1, s2,s3*, s4, s5, s6, s7, s8, s11,
LP15 e1, e3,e8, e17, e19, c3, c4, c7, c9, c10, c11, c13, s12, s13, s15, s16, s17, s20, s21,
c14, c15, c17, c18, c19, c21 s23
LP16 e9 s2, s3, s4, s5, s6, s7, s11, s13,
c1, c3, c4, c5, c6, c9, c10, c11, s17, s21, s23
LP17 e19 c12, c13, c14, c15, c19 s3, s4, s5, s6, s7, s21
c1, c2, c5, c16
LP18 - c1, c4, c6, c7, c10, c11, c12, s2, s3, s4, s5, s6, s7, s15, s16, s21
c13, c14, c19
LP19 e14, e17, e18, e19 c6, c7, c12, c13, -
c1, c3, c4, c6, c*, c9, c10, c11*,
LP20 - c12, c13*, c14, c21 s2, s3, s4, s5, s6, s7, s8,
LP21 - -
LP22 e19 c6, c7, c11, c13 -
c1, c2, c6, s7
LP23 e1, e2, e3, e8, e9, e14, e15, c1, c2, c3, c4, c6, c7, c8, c9, s1, s3, s4, s5, s6, s7, s15, s16, s17
e16, e17, e18, e19 c10, c11, c12, c13*, c14, c16, s2, s3*, s4, s5, s6*, s7, s8, s9,
c17, c18, c19, c21 s10, s11, s12, s13, s15, s17, s20,
LP24 e9, e18, e19 c3, c9, c10, c11, c13, cp19 s21, s23
s2, s3, s4, s5, s15
LP25 e9, e18, e19 c3, c4, c11, c14, c15, s1, s2, s3, s4, s5*, s6, s7, s8, s12,
s15, s16, s17, s21

* Item with a very strong relationship

66

Meanwhile, there are 5 items in SSC and 5 items in the LP section
that shows a very strong relationship with each other at a score range of
0.801 to 0.848. From this relationship, the ability to improve the
manufacturing capability and flexibility (s6) has a high significant positive
correlation with the ability to reduce the setup time (LP13), and the ability to
reduce the lead time in production operations (LP23) with a value of 0.848.

The results indicate that LP not only successfully improves
operational efficiency, but also positively influences matters concerning the
environment. This is because LP have always had a close relation with the
objectives set in the environmental management9. This has been proven by
the results that show that the ability to reduce the production lead times
through the adaptation of LP has a strong relationship with the ability to
increase the opportunity to reduce pollution, which is potentially able to be
streamlined with the ability to increase the capacity and flexibility in
manufacturing. This is not surprising because through high flexibility,
manufacturers can manage the production operations in a more efficient
manner, including the way to manage pollution that possibly results from the
activities implemented. Furthermore, the ability to reduce the setup times also
correlates with the ability to allocate resources based on the requirements.
This does not only increase the level of operational efficiency, but can also
reduce the operating cost, as well as the marginal cost in handling the
environmental concerns. This indicates that LP always has a strong
relationship with the manufacturing efficiency, primarily in achieving high
sustainability in manufacturing operations.

67

5. CONCLUSION

As a conclusion, this initial study has shown that LP has a strong
relationship in the development of manufacturing sustainability. Coupled
with correlation analysis, it is shown that each of the practices in LP has a
significant positive correlation with each item in the manufacturing
sustainability components. The findings in this study can be used as a basis in
the next stage of a study in developing a more comprehensive strategy in
establishing manufacturing sustainability, primarily in Malaysia’s
manufacturing industry.

REFERENCES

1. S. Vinodh, Improvement of agility and sustainability: A case study in an Indian rotary
switches manufacturing organisation, J. Cleaner Prod. 18 (2010) 1015-1020.

2. M. Despeisse, M.R. Oates, P.D. Ball, Sustainable manufacturing tactics and cross-
functional factory modelling, J. Cleaner Prod. 42 (2013) 31-41.

3. S. Schrettle, A. Hinz, M. Scherrer, T. Friedli, Turning sustainability into action:
Explaining firms’ sustainability efforts and their impact on firm performance, Int. J.
Prod. Econ. 147 (2014) 73-84.

4. M.Z. Yusup, W.H. Wan Mahmood, M.R. Salleh, M.N.H. Mohd Rosdi, The trigger
signal for lean production practices : A review, Global Eng. Technol. Rev. 3 (2013) 23-
32.

5. S. Vinodh, R.J. Girubha, PROMETHEE based sustainable concept selection, Appl.
Math. Modell. 36 (2012) 5301-5308.

6. I. Awudu, J. Zhang, Uncertainties and sustainability concepts in biofuel supply chain
management: A review, Renewable Sustainable Energy Rev. 16 (2012) 1359-1368.

7. Y. Hongbing, Z. Haiyan, Sustainable development strategies of manufacturing in
Jiangsu based on the constraints of resources and environment, Energy Procedia. 5
(2011) 872-878.

8. E.W.T. Ngai, D.C.K. Chau, J.K.L. Poon, C.K.M. To, Energy and utility management
maturity model for sustainable manufacturing process, Int. J. Prod. Econ. 146 (2013)
453-464.

9. S. Hajmohammad, S. Vachon, R.D. Klassen, I. Gavronski, Lean management and
supply management: Their role in green practices and performance, J. Cleaner Prod. 39
(2013) 312-320.

68

APPENDIX

1. Performance of Lean Production Practice (LP)

How you rate your current performance in lean production practices for the past two (2)
years? (On a seven point scale from strongly disagree to strongly agree).

Item Lean Production Performance

LP1 Decrease customer lead time
LP2 Improve layout to reduce unnecessary movement
LP3 Increase knowledge in production management
LP4 Improve the production takt time
LP5 Defect detection ability of the product
LP6 Reduction in the throughput time
LP7 Maximise the operational flexibility
LP8 Minimising handling
LP9 Optimise usage of equipment
LP10 Reorganise of working space
LP11 Reduce changeover & handling time
LP12 Reduce inventories & storage
LP13 Setup time reduction
LP14 Better environmental management & control
LP15 Environmental practice & performance
LP16 Increase quality of products
LP17 Improve working conditions
LP18 Reduce the non-added value activities
LP19 Increase manufacturing capability & flexibility
LP20 Increase the operation efficiency
LP21 Increase the production productivity
LP22 Improve the organization of work environment
LP23 Reduce the production lead time
LP24 Improve the material flow
LP25 Improve the operation procedure

69

2. Sustainability of Environmental Performance (SEP)

What are your current environmental performances from the past two (2) years? (On a seven
point scale from strongly disagree to strongly agree).

Item Sustainability in Environmental Performance

e1 Increase recyclability activities
e2 Use recycle material in product & process
e3 Increased reusable, non-toxic & bio-degradable materials
e4 Increase amount of recycled components
e5 Increase amount of recycling of packaging materials
e6 Adapt reuse & recycling in design
e7 Use eco-friendly material
e8 Increase usage of renewable material
e9 Increase practice, management & performance in environmental
e10 Establish material & energy consumption reduction programmes
e11 Decrease energy consumption in the process
e12 Establish pollution prevention & reduction control
e13 Increase the opportunities of preventing pollution
e14 Reduce emission of substances & control
e15 Reduce the waste of materials
e16 Established waste reduction & energy efficiency programmes
e17 Use easily degradable chemicals
e18 Consider the direct environmental effect on operation
e19 Better environmental management and control

3. Sustainability of Economic Performance (SCP)

What are your current economic/financial performances from the past two (2) years? (On a
seven point scale from strongly disagree to strongly agree).

Item Sustainability in Economic Performance

c1 Minimise overall production cost
c2 Increase business and financial performance
c3 Efficient utilization of equipment and technology
c4 Efficient utilization of resources
c5 Increase the quality of product
c6 Reduce the non-added value activities
c7 Increase added value activities
c8 Reduce the total operation cost
c9 Reduce the total production lead time
c10 Minimize environmental costs in transportation
c11 Consider economic effect on the selection of the system
c12 Consider the risk taken in the investment
c13 Reallocating the resources based on requirement
c14 Increase the innovative in technological improvements
c15 Use renewable energy in production & transportation
c16 Purchase materials, parts & resources based on demand
c17 Product easily disassembly
c18 Establish quantitative, objective of quality in operation
c19 Redefine the competitive environment
c20 Reduce the number of parts in a product
c21 Properly plan the requirements of the material

70

4. Sustainability of Social Competency Performance (SSC)

What are your current social competency performances from the past two (2) years? (On a
seven point scale from strongly disagree to strongly agree).

Item Sustainability in Social Competency Performance

s1 Improve housekeeping & labour safety
s2 Improve process & flexibility
s3 Increase efficiency and competitiveness
s4 Increase product durability
s5 Driving force for improvement & process innovation
s6 Improve the manufacturing capability & flexibility
s7 Improve working conditions
s8 Improve the movement of operation flow
s9 Increase the operation efficiency
s10 Increase the production productivity
s11 Improve the organization of work environment
s12 Reduce the queuing time
s13 Develop standard & establish the consistency in the process
s14 Comply with environmental, regulation & safety issue
s15 Consider technology feasibility & labour safety
s16 Consider the social effect in product development
s17 Consider engaging of the workforce in the process
s18 Collaborate with other parties on environmental issues
s19 Possess environmental policy statements
s20 Possess strategic alliance with organizational strategies
s21 Use long length of life of components
s22 Utilize the electronic communication & documentation
s23 Consider previous & the common experience in operation

71

2019 M Shariff MA et al.

Response Surface Methodology (RSM) Model to
Evaluate Surface Roughness in Machining of

Titanium Alloy (Ti6-Al-4V) using End Milling Process

Mohd Azlim Bin Mat Shariff*, Asmizam**
*Kolej Kemahiran Tinggi MARA Balik Pulau,
**Faculty of Mechanical and Manufacturing Engineering

Universiti Malaysia Pahang
[Email : [email protected]]

Abstract – The inspiring working demand in surface finish product of
manufacturing process will step up the world into the next level. This
situation will drive its effect on product appearance, function and reliability.
The objective of these studies is to improve a better understanding of the
effects of cutting parameters such as speed, feed and axial depth of cut on
the surface roughness and to build up a response surface methodology
(RSM) model. An attempt has been made to achieve finest cutting conditions
with respect to center line average roughness (Rₐ) measured in the current
study with the help of response optimization technique. The design of
experiment (DOE) has been used to carry out the modeling and analysis of
the influence of process variables on that method. Analysis of variance
(ANOVA) has been done to verify the fit and competence of the established
mathematical model. Finally, the result from the effect of speed, feed and
axial depth of cut on the surface roughness of titanium alloys (Ti-6Al-4V)
samples were studied for coated carbide insert of the cutting tool during
machining process.

[Keywords: Surface Finish, Surface Roughness, Response Surface Methodology (RSM),
Speed, Feed, Axial depth of cut and Roughness Aver]

72

1. INTRODUCTION

The challenge of modern machining industries dedicated on the
achievement of high quality, in term of workpiece dimensional accuracy,
surface finish, high production rate, less wear on the cutting tools, an
economy of machining in terms of cost saving and an increase of the
performance of the product with reduced environmental impact.

The ability to control the process for better quality of the final product
is supreme importance. The surface texture is apprehensive with the
geometric irregularities of the surface of a solid material which is well-
defined in terms of surface roughness, waviness, lay and flaws. Surface
roughness (Rₐ) consists of the surface texture, including feed marks produced
by the machining process. The quality of a surface is a pointedly important
factor in assessing the efficiency of machine tool and machined parts.
Therefore, a decent quality machined surface essentially improves fatigue
strength, corrosion resistance and creep life1.

Titanium alloy is one of the most useable materials in the world. It
represents 50 percent of titanium use the world over. Its ease of use lies in its
numerous advantages. Titanium alloys might be warmly treated to expand its
quality. It tends to be utilized in welded development at administration
temperatures of up to 600° F. This alloy offers its high quality at lightweight,
valuable formability and high erosion obstruction 29.

Titanium alloys also ease of use makes it the best application for use
in a few industries, similar to the aviation, restorative, marine, and chemical
processing industries. It very well may be utilized in the formation of such
specialized things as Aircraft turbines Engine, components Aircraft, basic

73

components Aerospace fasteners, high-execution programmed parts Marine
applications and sports equipment 2.

The end milling processing is a standout amongst the most imperative
procedures which is generally used to create the primary parts in numerous
ventures, for example, the form and pass on parts, the aviation parts, and the
car parts 3.

End Milling is a process of generating machined surfaces by
progressively removing a predetermined amount of material from the
workpiece. Axis of tool rotation is perpendicular to feed direction. End
Milling is an interrupted cutting operation. In these operations, the tool is
constantly being heated and reheated 4.

The surface roughness is assuming an essential job to assess the
nature of a workpiece. Surface texture parameters and statistical functions are
superior in characterizing and evaluating surface quality and corresponding
functionality-related performance of machined components, when compared
with the traditional means in which only single valued standard surface
parameters being adopted 5.

Reasonable selection surface texture characterization and statistical
functions could give more specific and complete descriptions of the micro
geometry and functionality related properties for the machined surfaces
having identical values for primary indexes.

Effective correlation of the selective surface texture characterization
parameters and statistical functions with specific functionality-related
properties is implemented. These symbols provide a standard system of
determining and indicating surface finish. The inch unit for surface finish

74

measurement is micro inch (μm), while the metric unit is micrometer (μm) 15.

Roughness is defined as closely spaced, irregular deviation on a scale
smaller than that waviness. It is caused by the cutting tool or the abrasive
grain action and the machine feed. Roughness may be superimposed by
waviness. There are most two types significant of surface roughness as
follow:
i. Roughness Average, Ra

Roughness height is the deviation to the center line in micro inches or
micrometers.
ii. Roughness Depth, Rz

Roughness width is the distance between successive roughness peaks
parallel to the nominal surface in inches or millimetres12.

Figure 1 indicates standard phrasing and symbol to determine value
of surface roughness. The symbol p is the shape of any predetermined area
through a machined surface on a plane that is opposite to the surface.
Roughness width cut off l (i.e., testing length) is incorporated into the
estimation of normal unpleasantness stature. The mean line m of the profile p
is found with the goal that the entirety of the regions over the line (inside the
testing length l) is equivalent to the whole of the regions underneath the
line13.

Figure 1: Surface roughness definition

75

Despite the different surface finish parameters, the roughness average
Ra is the most used international parameter of surface roughness. It is
defined as equation (1).

1 | ( )| (1)
=

2. EXPERIMENTAL SETUP

A. Design of Experiments
Response Surface Method (RSM) is a collection of statistical and

mathematical methods that are useful for the modelling and optimization of
the engineering problems. In this technique, the main objective is to optimize
the responses that are influencing by various parameters. This method also
quantifies the relationship between the controllable parameters and the
obtained response6-7.

This study uses the Box-Behnken design in the optimization of
experiments using RSM to understand the effect of important parameters.
Three levels of cutting parameters are selected to investigate the
machinability of this alloy which is consist of range feed rate, 0.05mm/rev,
0.1mm/rev, 0.15mm/rev, different value of cutting speed, 50m/min,
100m/min and 150m/min and different value of depths of cut such as 0.2mm,
0.5mm and 0.8mm were selected8-11. The values of parameters in conducting
the experiment are shown in Table 1 and design of experimental shown in
Table 2.

76

Table 1: Machine parameter and level

Destination Process parameters Level
-1 0 1
X1 Cutting Speed (m/min) 50 100 150
X2 Feed (f/rev) 0.05 0.1 0.15
X3 0.2 0.5 0.8
Axial Depth (mm)

Table 2: Design values of parameters generated by Minitab 18

Run Speed, Vs Feed, f Axial Depth,
Order (m/min) (mm/rev) ap(mm)
0.2
1 100 0.05 0.2
2 100 0.15 0.8
3 50 0.10 0.5
4 100 0.10 0.5
5 150 0.05 0.8
6 100 0.05 0.5
7 50 0.15 0.2
8 50 0.10 0.8
9 150 0.10 0.2
10 150 0.10 0.8
11 100 0.15 0.5
12 50 0.05 0.5
13 100 0.10 0.5
14 150 0.15 0.5
15 100 0.10

B. Workpiece and Cutting Tool Material

Titanium alloy also has numerous applications in the medical industry
and biocompatibility of titanium alloy is excellent, especially when direct

77

contact with tissue or bone is required. The mechanical properties of titanium
alloy shown in Table 3 and chemical composition are shown in Table 4.

Table 3: Mechanical properties of Titanium Alloy (Ti6-Al-4V)
(Source: ASM Material Data Sheet)

# Mechanical Properties Value
1 Hardness, Vickers 349 HV
2 Tensile Strength, Ultimate 950 MPa
3 Tensile Strength, Yield 880 MPa
4 Modulus of Elasticity 113.8 GPa
5 Poisson’s Ratio 0.342
6 Shear Modulus 44 GPa
7 Shear Strength 550 MPa

The tool holder with an index able insert, examples as coated carbide
is very good for cutting a hard material. Its positive cutting edge removes
metal by slicing through the material, rather than by scraping.

Table 4: Chemical Composition (% by weight) of Titanium
Alloy (Ti6-Al-4V)

(Source: Fine tubes product Material)

Element Min (% by weight) Max (% by weight)
Al 5.5 6.5
C - 0.08
Fe - 0.25
H - 0.015
N - 0.05
Ti Balance
O - 0.2
V 3.5 4.5

78

Figure 2: End Mill with Indexable Toolholder (Source: www.ceratizit.com)

Figure 3: Shape of Coated Carbide Insert (Source: www.ceratizit.com)
The configuration ought to be adequate to fit a quadratic model, that

is, one containing squared terms, results of two variables, straight terms, and
a catch. The proportion of the quantity of trial focuses on the number of
coefficients in the quadratic model ought to be sensible15.

This experiment was conduction in wet machining condition on a
CNC Milling Machine HAAS VF6 by slotting cutting process equipped with
a spindle max of 3000 rpm. The type of cutting tools used is coated carbide
by toolholder diameter 16mm. One pass machining cycle at Y Axis was used
to determine the machining surface condition16.

79

Two common methods of surface roughness measurement in
engineering practice are described, including the instrumentation involved
and brief description of surface roughness requirements in engineering
practice. All the surface roughness measurement data collected. The result of
the surface roughness must be less than 1 μm for better surface finish, if more
than that value; the process of machining by using end milling will be
repeated17.

Then the value of surface roughness was measured by surface
perthometer manufactured by Mahr model (Surf PS1). Roughness average
definitions are shown in Figure 4.

Figure 4: Arithmetical mean height (Ra) (Source: www.keyence.com)
The observation of cutting surface were taken for each cutting process

each sample and were averaged in order to get the significant value of
Roughness average (Ra)18. The type of milling machine is shown in Figure 5
and material set up on machine and cutting tool are shown in Figure 6.

80

Figure 5: CNC Milling HAAS VF6 (Source: www.haascnc.com)

Figure 6: Experiment setup of tooling and material cutting using HAAS
VF6 CNC Milling Machine

C. Response Surface Methodology
The fundamental target is to optimize the response between surfaces

that is affected by different process parameters. RSM is measured the
connection between the information parameters and the acquired reaction
surfaces the second-order polynomial scientific model for surface roughness
is created as equation (2).

81

=+ + + (2)

where Y is the corresponding response (surface roughness, SR) yield by
the various variables and X 1 (1,2,3…n) are coded levels of n quantitative
process variables, the term 0 C, j C, jj C and ij C are the second order
regression coefficients. Equation (2) can be written as equation (3).

Y = -C0 + C1 X1 + C2 X2 + C3 X3 – C11 X1²- C12 X2² + (3)
C13 X1 X2 + C23 X1 X3- C33 X2 X3

where 1, 2 3 X X , X are feed rate (mm/tooth), axial depth (mm) and cutting
speed (m/min) respectively20-24. The equations of the fitted model for SR are
represented in equation (4).

Y = -0.100 + 0.00532 X1 + 0.00 X2 + 0.967 X3 – 0.000037 X1² -

1.053 X2² + 0.0001 X1X2 + 0.00390 X1X3- 3.57 X2X3 (4)

Besides, to make sure the overall values are fit and lack of errors, P-
values were identified. Table 5 shows the corresponding P-values for the data
machining surfaces. Based on Table 5, the P-values show that the
mathematical model is significant and adequate in order to determine the
value of surface roughness. The coefficients generated can be used for
mathematical modeling.

82

Table 5: Analysis of Variance for Roughness Average (Ra)

Source of Degree Sum of Mean of F- P-Value
Variation of square Square Value 0.016
0.077765 8.3 0.003
Freedom 0.699884 0.191594 20.45 0.103
0.574783 0.033321 3.56 0.505
Regression 9 0.099963 0.008379 0.89
0.025138 0.108
Linear 3 0.00937 8.43
0.046849
Square 3 0.014472
0.043416 0.001716
2-Way 3
Interaction 0.003433
0.746733
Residual 5
Error

Lack-of- 3
Fit

Pure Error 2

Total 14

3. RESULT AND DISCUSSION

The analysis of variance is exhibited in Table 5. The sufficiency of
the model is confirmed utilizing using ANOVA. At a confident level of
95%, the model is checked for its sufficiency. In the Table 5, model is
satisfactory because of the way that the P value of lack-of-fit is not
significant. This suggests that the model could fit, and it is sufficient.
Hence, the model is satisfactory and there is some indicator to measure the
viability of the model that worked in the estimation of surface roughness
prediction data25-28. Figure 7 shows that normal probability during
response is average at the value of value of 95%.

83

Figure 7: Normal Probability Plot
The variation of the effect of cutting parameter against cutting speed,
feed and axial depth are represented in Figure 8. From the graph, the value of
surface roughness is increasing as the feed increases.

Figure 8: Main effect of machine parameter
This condition happens due to the increase in heat between the tool
and the workpiece, thus causing the tool to be exposed to damage because of
the high frictional force between the tool and workpiece. Effect on cutting

84

speed and axial depth shows the same pattern. When the value of cutting
speed is low it will give the best condition of surface on machine material.
Means that, feed rate minimum will give a better surface finish.

The effect of feed rate depends on other factors, mainly spindle speed.
This is because feed rate and spindle speed involve movement which then
decides the thermal barrier and frictional forces between tool and workpieces.
According to the graph, surface roughness value becomes more significant as
the cutting speed decreasing. From the graph also, it shows that value of feed,
f gives more significant result to the specimens.

Figure 9: Interaction plot of machine parameter
Variation of surface roughness demonstrates the effect of surface
roughness against the feed rate and cutting speed. It very well may be seen
that the feed has the most predominant impact of surface roughness, trailed
by the axial depth of cut and cutting speed. Littler cutting powers cause less
vibration and give a superior surface completion Figure 10(a). It is obvious

85

from Figure 10(b) that surface at roughness increments with the reduction in
feed rate.

Figure 10(a): 2D contour plot of Surface Roughness
A low feed rate uniforms the external surface accordingly expanding
the surface completion. Another factor to consider is cutting velocity. It is
comprehended that an expansion in cutting pace improves surface quality.
This outcome supports the disagreement that sufficiently high cutting rates
reduce cutting powers together giving a superior surface achievement.
Subsequently, better surface roughness can be acquired by utilizing
high cutting speed, low axial depth of cut, and low feed rate.

86

Figure 10(b): 3D contour plot of Surface Roughness
Figure 11 demonstrates the predicted outcomes closely concur with
the experimental qualities. Thusly, the model of the response surface strategy
can be acknowledged too.

Figure 11: Value of surface roughness between predicted and experimental

87

4. CONCLUSION

This part will conclude the significant values of this investigation.
This study has thought of a few models to the expectation of surface
roughness approval on coated carbides due to machining titanium alloy
(Ti6Al4V). Separately from that, it has been focused on the optimization of
machining parameters and their impact on surface roughness. A few
suggestions and recommendations have been accommodated future research
in this field.

In these investigations, RSM has been utilized to decide the
expectation of surface roughness by machining titanium alloy (Ti6Al4V)
with coated carbide for different input parameters to be specific the feed rate,
axial depth of cut, and cutting speed. The feed rate has the most significant
impact surface roughness, trailed by the feed, axial depth of cut and the
cutting speed.

The higher estimation of feed rate decreases the surface defect of and
may add to surface irregularities, for example, surface flaw and cavities
during cutting process. The RSM model can effectively relate the above
procedure parameters with the reaction surface roughness. Therefore,
advancing the cutting conditions are fundamental so as to improve the
surface roughness in machining.

These studies also likewise have fundamentally utilized reaction
surface technique to create scientific models to enhancement so as to tackle
and to locate the best information and output parameters. Based on these
studies, coated carbides have assumed a significant job in machining titanium
alloy Ti6Al4V.

88

REFERENCES

1. Stark G A and Moon K S 1999 Trans. of the ASME, J. Manuf. Sci. Eng. 121, 251.
2. D. He et al., “Influences of deformation strain, strain rate and cooling rate on the

Burgers orientation relationship and variants morphology during phase transformation
in a near titanium alloy,” vol. 549, pp.
20–29, 2012.
3. S. T. P. Thesniyom and S. Ratanakuakangwan, “Prediction of surface roughness in
ball-end milling process by utilizing dynamic cutting force ratio,” J.
Intell. Manuf., pp. 13–21, 2017.
4. N. Masmiati and A. A. D. Sarhan, “Optimizing cutting parameters in inclined end
milling for minimum surface residual stress – Taguchi approach,” MEASUREMENT,
vol. 60, pp. 267–275, 2015.
5. S. D. Philip, P. Chandramohan, and P. K. Rajesh,
“Prediction of surface roughness in end milling operation of duplex stainless-steel
using response surface methodology,” J. Eng. Sci. Technol., vol. 10, no. 3, pp. 340–
352, 2015.
6. S. L. C. Ferreira et al., “Box-Behnken design : An alternative for the optimization of
analytical methods,” vol. 597, pp. 179–186, 2007.
7. N. Aslan and Y. Cebeci, “Application of Box –
Behnken design and response surface methodology for modeling of some Turkish
coals,” vol. 86, pp. 90–97, 2007.
8. Montgomery, D.C.. 1997. “Design and analysis of experiments”, Wiley, New York.
9. Draper, N.R. and H. Smith, 1981. “Applied Regression Analysis”, Wiley, New York.
10. Box, G.E.P. and N.R. Draper. 1987. “Empirical model-building and response
surfaces”,JohnWiley & Sons, New York.
11. Box, G.E.P. & Behnken, D.W. 1960. “Some new three level designs for the study of
quantitative variables”, Technometrics2, pp. 455-475.
12. N. E. Karkalos, N. I. Galanis, and A. P. Markopoulos,“Surface roughness prediction
for the milling of Ti-6Al-4V ELI alloy with the use of statistical and soft computing
techniques,” Meas. J. Int. Meas. Confed., vol. 90, pp. 25–35, 2016
13. Q. Zeng, Y. Qin, W. Chang, and X. Luo, “International Journal of Mechanical
Sciences Correlating and evaluating the functionality-related properties with surface
texture parameters and specific characteristics of machined components,” Int. J.
Mech. Sci., vol. 149, no. September, pp. 62–72, 2018.
14. A. Moufki, G. Le Coz, and D. Dudzinski, “End-milling of Inconel 718 superalloy -
An analytical modelling,” Procedia CIRP, vol. 58, pp. 358–363, 2017.
15. T. Thepsonthi and T. Özel, “Multi-objective process optimization for micro-end
milling of Ti-6Al-4V titanium alloy,” Int. J. Adv. Manuf. Technol., vol. 63, no. 9–12,
pp. 903–914, 2012.
16. A. Zaretalab, H. S. Haghighi, S. Mansour, and M. S. Sajadieh, “A mathematical
model for the joint optimization of machining conditions and tool replacement policy
with stochastic tool life in the milling process,” Int. J. Adv. Manuf. Technol., vol. 96,
no. 5–8, pp. 2319–2339, 2018.
17. S. A. Abbasi, P. Feng, Y. Ma, J. Zhang, D. Yu, and Z. Wu, “Influence of
microstructure and hardness on machinability of heat-treated titanium alloy Ti-6Al-
4V in end milling with polycrystalline diamond tools,” Int. J. Adv. Manuf. Technol.,
vol. 86, no. 5–8, pp. 1393–1405, 2016.

89

18. M. Chahal, V. Singh, and R. Garg, “Optimum surface roughness evaluation of dies
steel H-11 with CNC milling using RSM with desirability function,” Int. J. Syst.
Assur. Eng. Manag., vol. 8, no. 2, pp. 432–444, 2017.

19. B. Bhardwaj, R. Kumar, and P. K. Singh, “An improved surface roughness prediction
model using Box-Cox transformation with RSM in end milling of EN 353,” J. Mech.
Sci. Technol., vol. 28, no. 12, pp. 5149–5157, 2014.

20. E. Kilickap, A. Yardimeden, and Y. H. Çelik, “Mathematical Modelling and
Optimization of Cutting Force, Tool Wear and Surface Roughness by Using
Artificial Neural Network and Response Surface Methodology in Milling of Ti-
6242S,” Appl. Sci., vol. 7, no. 10, p. 1064, 2017.

21. K. H. Hashmi, G. Zakria, M. B. Raza, and S. Khalil, “Optimization of process
parameters for high speed machining of Ti-6Al-4V using response surface
methodology,” Int. J. Adv. Manuf. Technol., vol. 85, no. 5–8, pp. 1847–1856, 2016.

22. M. T. Hayajneh, M. S. Tahat, and J. Bluhm, “A Study of the Effects of
Machining Parameters on the Surface Roughness in the End-Milling Process,” vol. 1,
no. 1, pp. 1–5, 2007.

23. N. Qehaja, K. Jakupi, A. Bunjaku, M. Bruçi, and H. Osmani, “Effect of Machining
Parameters and Machining Time on Surface Roughness in Dry Turning Process,”
Procedia Eng., vol. 100, pp. 135– 140, 2015.

24. S. Selvakumar and R. Ravikumar, “A Novel Approach for Optimization to Verify
RSM Model by Using Multi-Objective Genetic Algorithm (MOGA),” Mater. Today
Proc., vol. 5, no. 5, pp. 11386–11394, 2018.

25. A. Gupta, R. Shah, H. Dave, and N. Khanna, “Multi-objective optimization of surface
parameters such as concavity, straightness and roughness in milling process,” Mater.
Today Proc., vol. 5, no. 2, pp. 5296– 5302, 2018.

26. C. Bandapalli, B. M. Sutaria, D. V. Bhatt, and K. K. Singh, “Experimental
Investigation and Estimation of Surface Roughness using ANN, GMDH & MRA
models in High Speed Micro End Milling of Titanium Alloy (Grade-5),” Mater. Today
Proc., vol. 4, no. 2, pp. 1019–1028, 2017.

27. H. A. Sonawane and S. S. Joshi, “Modeling of machined surface quality in high-speed
ball-end milling of Inconel-718 thin cantilevers,” Int. J. Adv. Manuf. Technol., vol. 78,
no. 9–12, pp. 1751–1768, 2015.

28. T. Thepsonthi and T.zel, “An integrated toolpath and process parameter optimization
for high-performance micro-milling process of Ti6Al4V titanium alloy,” Int. J. Adv.
Manuf. Technol., vol. 75, no. 1–4, pp. 57–75, 2014.

29. S. A. Sulaiman, A. K. M. N. Amin, and M. D. Arif, “Optimization of Surface
Roughness in End Milling of Titanium Alloy Ti-6Al-4V under the Influence of
Magnetic Field from Permanents Magnets,” Adv. Mater. Res., vol. 576, pp. 51–55,
2012.

90

2019 AA Ghafar et al.

Pre-Forming Inspection System to Detect Deep
Drawing Defect Due to Punch-Die Misalignment using

Image Processing Technique

Alimi Abdul Ghafar*, Prof. Madya Ir. Dr Ahmad Baharuddin Abdullah**
*Department of Plastic Injection Moulding

Kolej Kemahiran Tinggi MARA Balik Pulau,
**School of Mechanical Engineering, Universiti Sains Malaysia

[Email : [email protected]]

Abstract – Deep drawing processes are extensively used in the industry for
the fabrication of a wide range of sheet metal products from household
appliances to automotive parts. In a deep drawing process, the alignment of
the punch and die is extremely important to avoid defects such as variation of
thickness and intolerable thinning condition of the drawn part. In this paper, a
computer vision system is proposed that is able to identify the degree of
misalignment in the punch-die configuration. The apparatus consists of an
image capture and analysis system that can be used as a tool for pre-forming
inspection setup. The image analysis implements the centroid recognition
approach using the Image Processing Toolbox function in MATLAB. The
offset value between centre of the punch and the centre of die indicates the
severity of misalignment between punch and die. The system was
successfully tested for a square profile, with more accurate results obtained
for a colored punch as compared to the punch with the original metallic
surface. Verification of the system was then conducted to align the centroids
of a punch and die, and the results show that a good alignment of the
centroids was achieved within two alignment attempts.

[Keywords: Deep drawing process, misalignment, thinning, image processing technique]

91

1. INTRODUCTION

Deep drawing is defined as the forming of a flat stretched sheet metal
through a die into various shapes by using a compression force. Four basic
components are required to perform a drawing process, namely, the punch,
die, blank and blank holder. The drawing process can be conducted in a
single step, although for greater depths, multiple drawings are usually
needed. The total number of redrawing depends on many factors such as the
material used (the blank), the dimensions required, and the final shape of
parts. As compared to other production methods, the deep drawing process is
able to produce high quality and complex shaped parts with minimal costs1.

Some of the defects that can occur in deep drawing process are earing,
wrinkling (either on the flange or wall of part), tearing and surface scratches2.
Earing defects occur due to variations in properties because of the anisotropy
of the material used as the blank. Anisotropy is a phenomena that is difficult
to avoid when working with sheet metals3. Wrinkling is another type of
defect that usually occurs in a deep drawing process, its severity increases by
insufficient blank holder force. Wrinkling usually occurs on the flange of the
part, although it can also appear on the wall of the part. Tearing defects can
occur due to severe thinning of the blank, caused by incorrect geometrical
dimension of the die such as die radius, punch nose radius and punch-die
clearance. Excessive thinning at the wall of the part is one the major
contributing factor in the failure of the draw process. As mentioned by
Colgan and Monaghan4, a decrease in the die radius would increase the
drawing force. This increase in drawing force creates uneven thickness
distribution at the wall of the part and thinning of the part structure. Severe
thinning would result in tearing of the part.

92