This article was downloaded by: [University of Newcastle (Australia)]

On: 01 September 2014, At: 11:19

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Materials and Manufacturing Processes

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lmmp20

Investigating the Effects of Injection Molding

Parameters on the Mechanical Properties of Recycled
Plastic Parts Using the Taguchi Method
a b
Nik Mizamzul Mehat & Shahrul Kamaruddin
a
Department of Mould Technology , Kolej Kemahiran Tinggi MARA , Balik Pulau, Malaysia
b
School of Mechanical Engineering , University Sains Malaysia , Pulau Pinang, Malaysia
Published online: 24 Mar 2011.

To cite this article: Nik Mizamzul Mehat & Shahrul Kamaruddin (2011) Investigating the Effects of Injection Molding
Parameters on the Mechanical Properties of Recycled Plastic Parts Using the Taguchi Method, Materials and Manufacturing
Processes, 26:2, 202-209, DOI: 10.1080/10426914.2010.529587

To link to this article: http://dx.doi.org/10.1080/10426914.2010.529587

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Materials and Manufacturing Processes, 26: 202–209, 2011
Copyright © Taylor & Francis Group, LLC
ISSN: 1042-6914 print/1532-2475 online
DOI: 10.1080/10426914.2010.529587

Investigating the Effects of Injection Molding

Parameters on the Mechanical Properties of Recycled
Plastic Parts Using the Taguchi Method
Nik Mizamzul Mehat1 and Shahrul Kamaruddin2
1
Department of Mould Technology, Kolej Kemahiran Tinggi MARA, Balik Pulau, Malaysia
2
School of Mechanical Engineering, University Sains Malaysia, Pulau Pinang, Malaysia
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014

The growing amount of plastic parts produced nowadays makes the search for alternatives in recycling and the further use of these
nonbiodegradable materials imperative. The degradation of the mechanical properties of recycled plastic products poses the primary limitation for
the usage of recycled plastic. One of the foremost causes of mechanical property degradation is variation in processing parameters. An appropriate
optimization method that effectively controls all influential processing parameters during manufacturing is therefore critical. This study investigates
the effects of injection molding parameters on the mechanical properties of recycled plastic parts. The preliminary experiment is conducted by
using Moldflow Plastic Insight (MPI) integrated with the L18 Taguchi orthogonal array (OA). The significant processing parameters obtained from
the preliminary experiment were used to conduct the principal experiment. By adopting L9 Taguchi OA, the parts made from recycled plastic
were produced by injection molding. ANOVA confirms that the most significant factor for flexural modulus of a recycled toolbox tray is injection
time (∼ 40.49% percentage contribution). For stress at yield, the most significant factor is melt temperature with percentage contribution of
about 43.34%.
Keywords Flexural modulus; Injection molding; Recycled plastic; Simulation; Stress at yield; Taguchi method.

Introduction processing parameters for the setup and start-up of the

Processing parameters critically influence the quality
of injection-molded parts. The plastic injection moulding
(PIM) process is complex, involving many processing
parameters such as pressure, temperature, and time. These
parameters are dependent on each other and, by changing
any one of these parameters, one or all will be affected.
These processing parameters need to be optimized to
improve part quality and maximize production capacity. The
advanced development of injection molding flow analysis
simulations has assisted designers/engineers to predict and
overcome problems in selecting the optimum processing
parameters. Thus, relying on experience, intuition, or
trial and error in obtaining information regarding the
processing parameters can be avoided [1, 2]. However,
single-handedly implementing these simulation package
enhancements requires the optimization of processing
parameters to adopt a trial-and-error procedure. In fact, to
determine the optimal processing parameters for producing
a good quality injection-molded plastic part, the simulation
model must be carried out in a number of replications
if the experiment is single-handedly performed. As a
result, the experimental procedures are cumbersome and
chaotic [3]. Therefore, to reduce this acrimonious condition,
optimization methodologies are normally coupled with
simulation to assist engineers in identifying the optimal
Received July 22, 2010; Accepted September 28, 2010
Address correspondence to Shahrul Kamaruddin, University Sains
Malaysia, School of Mechanical Engineering, Nibong Tebal, Pulau
Pinang, Malaysia; E-mail: meshah@eng.usm.my
202
injection molding process. One of these efficient and simple
optimization methods is the Taguchi method [4–6].
The Taguchi method, also known as Robust Design, was
developed by Dr. Genichi Taguchi [7] and is extensively
used to optimize the performance characteristics of a
plastic part produced through the setting of the design
parameters. This method is also essential for designing
high-quality systems and determining the optimum settings
for the controllable parameters to make the product or
process insensitive to noise factors [8, 9]. The Taguchi
method introduces an integrated approach that is simple and
efficient for finding the best range of designs for quality,
performance, and computational cost [10, 11]. There are
three stages applied in this method: (1) system design, (2)
parameter design, and (3) tolerance design [12]. Of all the
three, parameter design is often identified as the most crucial
and dominant stage for process optimization [13, 14]. In
general, parameter design of the Taguchi method utilizes
Orthogonal Array (OA), signal-to-noise (S/N) ratios, and
ANOVA. Taguchi OA is a highly fractional orthogonal
design used to estimate the main effects by only a few
experimental runs. Thus, adopting OA can reduce research
and development costs by simultaneously studying a large
number of parameters instead of studying one parameter at
a time as done in traditional methods [15, 16]. Roy [17]
further discussed the Taguchi concept.
In recent years, the Taguchi Method has become a
widely accepted technique for improving productivity and
optimizing processes. There is a number of published
studies discussing the effect of processing parameters on
the quality of injection-molded parts. Most studies focus on
 EFFECTS OF INJECTION MOLDING PARAMETERS
minimizing the problems related to dimensional stability of
the produced part by integrating the Taguchi method either
in simulation or in real manufacturing process. Focusing
on the effects of processing parameters on warpage,
Liao et al. [15], Ozcelik and Sonat [16], and Huang
and Tai [18], used the experimental design of Taguchi
method to determine injection molding parameters. The
most influential factor on warpage has been found to be
the packing pressure. Further work on the warpage problem
was reported by Tang et al. [19] and Chen et al. [20].
The study reveals that the most influential factor on the
warpage of thin plate parts is the melt temperature. Warpage
can also be associated with the variation in shrinkage. By
conducting the three-level L27 and L9 in the Moldflow
packages, Oktem et al. [21] reduced the warpage problems
related to a variation in processing parameters dependent
on shrinkage. Packing pressure has been found to be the
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014
most important parameter affected by shrinkage in the part.
Another work focusing on shrinkage was done by Chang
and Faison [22] and Altan [23]. The results show that
mold and melt temperature along with packing pressure are
the most significant factors affecting shrinkage behavior.
Chen et al. [24] improved the silver streaks problem that
appeared on the surface of polycarbonate/poly(butylene
terephthalate) (PBT) automobile bumpers by applying the
Taguchi method. The main factors directly related to the
generation of silver streaks are mold temperature, fill time,
fill/post fill switch over control, and injection rate. For
the weld line that occurs in injection-molded parts, the
melt temperature and mold temperature are the important
parameters affecting weld line strengths. The experiments
were conducted by Liu and Yang [25] and Xie and
Ziegmann [26].
Fung and Kang [27] investigated the influence of injection
molding processing parameters on the friction properties
of 30 wt% PBT by integrating the C-MOLD package
and L9 Taguchi OA. The ANOVA results showed that
among all four controllable factors, the melt temperature
and ram speed are the most influential to friction properties.
A work by Kuo and Jeng [28] discussed the effect of
various processing parameters on the tensile strength of the
ultrahigh molecular weight polyethylene injection-molded
part, and six processing parameters were studied: melt
temperature, mold temperature, injection velocity, packing
pressure, packing time, and cooling time. Yang et al. [29]
analyzed contour distortions, wear mass losses, and tensile
properties of polypropylene (PP) components applied to
the interior coffer of automobiles. The nine selected
controlling factors with two levels for each are melt
temperatures, injection speeds, and injection pressures
through three computer-controlled progressive strokes.
Other studies related to the mechanical properties of the
part produced were completed by Fung and Tien [30]. They
investigated the ultimate stress and fracture toughness of
injection-molded 30 wt% glass fibre-reinforced (PBT) using
the Taguchi method and Grey relational analysis.
Most studies based on the Taguchi method have
been concerned with optimizing processing parameters
to minimize problems related to dimensional stability or
mechanical properties of the part produced using virgin
materials only. With the rapid rise of serious pollution
203
problems due to the large volume of plastic disposal in
landfills, issues of recycling and reusing plastic wastes
are gaining significance. PIM, one of the most prevalent
plastics processing techniques, facilitates the substitution of
virgin plastic for recycling in producing parts. However,
the deterioration in mechanical properties of the part
made from recycled plastic is a major limitation of
recycled plastic usage. One of the foremost causes for
this deterioration is variation in processing parameters. In
practice, using these recycled plastics, particularly through
injection molding, involves the consideration of intricate
and complex processing parameters. Despite the extensive
literature, no reports have been found on the optimization of
injection molding parameters by the Taguchi method related
to the mechanical properties of the plastic part produced
using recycled plastic.
The main goal of this paper is to investigate the effects
of variation in injection molding processing parameters
on the mechanical properties of parts produced using
recycled plastic. To achieve this goal, a simulation using the
Moldflow Plastic Insight (MPI) integrated with L18 Taguchi
OA was used to conduct the preliminary experiment.
Compared with other studies that use the simulation for
minimizing shrinkage and warpage in injection-molded
parts, MPI is utilized in this article to identify the crucial
processing parameters affecting the mechanical properties
of the part before real parts are produced through injection
molding in the principal experiment. By incorporating
the significant processing parameters obtained from the
preliminary experiment, the parts made of recycled plastic
are injected using the L9 Taguchi OA as the design of the
experiment. The specimen tests are prepared for mechanical
testing. Using the S/N ratio and ANOVA, the optimal
processing parameters/levels and the effects of each on the
mechanical performance of the recycled plastic part are
further evaluated.
Preliminary experiment
The preliminary experiment is conducted by the
commercial simulation package MPI version 6.1
(Moldflow) [31]. The purpose of conducting the preliminary
experiment is to identify and select the significant
processing parameters affecting the mechanical properties
of the recycled plastic part. In this experiment, Taguchi OA
is adopted to perform all simulations.
Simulation Model and Material
For the Moldflow analysis, the part known as the toolbox
tray with dimensions of 221 mm length × 111 mm width ×
40 mm height is used. The part designed using SolidWorks
2007 is shown in Fig. 1. In this study, the simulation model
is meshed with the total number of elements (8115). There
is no such information on recycled plastic available in
MPI; therefore, the crystalline thermoplastic PP is specified
for the meshed model. The PP manufacturer is Idemitsu
Petrochemical Co. Ltd.; the trade name is Polypro J2000G.
Parameter Design of the Preliminary Experiment
The aim of this study is to investigate the effects
of processing parameters on the mechanical performance
 204 N. M. MEHAT AND S. KAMARUDDIN

Table 2.—L18 OA used in the preliminary experiment.

1 2 3 4 5 6 7 8 Results

Residual
Factors/ stresses
trials no. A B C D E F (MPa)

1 20 220 0.2 70 10 10 27.77

2 20 243 0.6 80 20 20 26.79
3 20 260 1.0 85 30 30 25.40
4 40 220 0.2 80 20 30 28.36
5 40 243 0.6 85 30 10 27.23
6 40 260 1.0 70 10 20 24.22
7 60 220 0.6 70 30 20 27.68
Figure 1.—Tool box tray. 8 60 243 1.0 80 10 30 25.66
9 60 260 0.2 85 20 10 23.95
10 20 220 1.0 85 20 20 28.40
11 20 243 0.2 70 30 30 25.25
of recycled plastic parts. Therefore, to characterize the 12 20 260 0.6 80 10 10 25.79
simulation model in terms of mechanical properties, residual
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014

13 40 220 0.6 85 10 30 28.67

stress in the cavity (available in MPI) is selected as the 14 40 243 1.0 70 20 10 25.54
quality characteristic for study. The residual stress is a 15 40 260 0.2 80 30 20 24.74
16 60 220 1.0 80 30 10 27.97
process-induced stress frozen in a molded part; here, it 17 60 243 0.2 85 10 20 26.41
is obtained from the preliminary experiment. The quality 18 60 260 0.6 70 20 30 24.74
characteristic represents the mechanical properties of the
simulation model because it affects a part in the same way
as externally applied stresses do. If the quality characteristic
is strong enough to overcome the structural integrity of the preliminary experiment. The results representing the
the part, the part will warp upon ejection or crack when residual stress values are recorded in the final column.
the external service load is applied. In this preliminary All the simulations are run as OA presented in Table 2
experiment, the smaller and better quality characteristic accordingly. In Trial 1, the simulation is conducted by a set
is selected for studying the effects of processing parameters of processing parameters including mold temperature (A)
on mechanical properties. The minimum residual stress is at 20 C, melt temperature (B) at 220 C, injection time (C)
preferable in producing the part with enhanced mechanical of 0.2 s, packing pressure (D) of 70%, packing time (E) of
properties. 10 s, and cooling time (F) of 10 s, respectively.
For the MPI simulation, six processing parameters
are selected according to previous investigations: mold Principal experiment
temperature, melt temperature, injection time, packing In the principal experiment, three levels of L9 of Taguchi
pressure, packing time, and cooling time. The processing OA are adopted to produce the toolbox trays. The trays
parameters and level values are shown in Table 1. In this are produced by various processing parameters selected
study, L18 (21 , 37 ) OA is selected to represent the design from the ANOVA results of the preliminary experiment.
of the experiment in the preliminary experiment. OA is Mechanical testing is performed to investigate the effect
presented in Table 2. of the selected processing parameters on mechanical
In Table 2, the trial numbers indicate the number properties. Finally, a series of confirmation test is conducted
of conducted simulations. In this study, 18 simulations to verify the effectiveness of this approach. Details of
are conducted and assigned as Trials 1–18. Factors the principal experiment will be explained in detail in the
A–F represent the processing parameters: A for mold subsequent sections.
temperature, B for melt temperature, C for injection time,
D for packing pressure, E for packing time, and F for Principal Experiment Material
cooling time (Table 1). All these factors are assigned to The material used in the principal experiment is
Columns 2–7. The two remaining columns (i.e., Columns post-industrial recycled PP (rPP) with a melt flow index of
1 and 8) are eliminated and will not be used when running 15.80 g/min. The rPPs (100% of recycled PP) are used as
raw material for the toolbox trays. The toolbox trays are
injected by the Battenfeld TM750/210 injection molding
Table 1.—Factors and their respective levels studied in the machine.
preliminary experiment.

Column Factors Level 1 Level 2 Level 3 Parameter Design of the Principal Experiment
A Mold temperature ( C) 20 40 60
In the simulation experiment, residual stress, which gives
B Melt temperature ( C) 220 243 260 an effect similar to that of externally applied stresses,
C Injection time (s) 0.2 0.6 1.0 is selected as the quality characteristic for study. The
D Packing pressure (%) 70 80 85 characteristic can warp the part upon ejection. The part
E Packing time (s) 10 20 30 cracks when external service load is applied if the residual
F Cooling time (s) 10 20 30
stress is strong enough to overcome the structural integrity
 EFFECTS OF INJECTION MOLDING PARAMETERS 205

Table 3.—Factors and their respective levels studied in the principal

experiment.

Column Factors Level 1 Level 2 Level 3


A Melt temperature ( C) 220 243 260
B Packing pressure (%) 70 80 85
C Injection time (s) 0.2 0.6 1.0
D Packing time (s) 10 20 30

of the part. Therefore, the tendency for the part to bend

and the ability to resist deformation under load are the Figure 2.—Test specimen.
two mechanical characteristics of concern in the principal
experiment. The mechanical performance of the toolbox
tray made of rPP will be evaluated in terms of flexural room temperature (23 C) and 50% humidity. The test span
modulus and stress at yield. The two quality characteristics distance of 100 mm at a crosshead speed of 6.29 mm/min,
characterize the part with the application of external sample rate of 10 pts/s, and full scale load range of
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014

stresses. 30 kN are specified in the test. The test specimens are

The selection of processing parameters for the principal
experiment is determined from the ANOVA result of
the preliminary experiment. After an intensive analysis
of the preliminary experimental results, four processing
parameters are identified to have a major influence on the
mechanical properties of the part. All four (i.e., melting
temperature, packing pressure, injection time, and packing
time) are varied to obtain the optimum levels of quality.
Other processing parameters, such as mold temperature
(32 C), injection pressure (40 bar), cooling time (10 s), and
stroke distance (140 mm), are kept constant during the
experiment. The selected processing parameters along with
their levels are given in Table 3. In this principal experiment,
L9 OA with nine trials is selected to lay out the experiment
(Table 4).
As shown in Table 4, factor A (melt temperature) is
assigned to Column 1, factor B (packing pressure) to
Column 2, factor C (injection time) to Column 3, and
factor D (packing time) to Column 4. After nine trials with
five repetitions are produced using L9 Taguchi OA, the
specimens tests are prepared for mechanical testing.
Mechanical testing
The three-point-bending flexural test is used to measure
the flexural modulus and stress at yield in accordance
with ASTM-D790. The test is performed on an INSTRON
Table Mounted Universal Testing Machine (model 3367) at
Table 4.—L9 OA used in the principal experiment.
1 2 3 4 Results
prepared with dimensions of 170 mm length × 20 mm width
× 2.65 mm thickness as shown in Fig. 2. The flexural
modulus and stress at yield obtained are then analyzed
by adopting the S/N ratios and ANOVA to determine the
optimal processing parameters/levels and the impact of each
processing parameters on the quality characteristics being
studied.
Results and discussion
Results of the preliminary experiment (Table 2) identify
which processing parameters significantly affect the selected
quality characteristic through ANOVA. The summary of
the degrees of freedom (DOF), sums of squares (S),
variance (V), F -ratio (F), and percentage contribution (P)
for the preliminary experiment are presented in Table 5.
The F-ratio or variance ratio is the ratio of the mean square
due to the factor and error mean square. A large value
of F means that the effect of a factor is large compared
with the error variance. The larger the value of F, the more
important it will be in influencing the quality characteristic.
This results in high P as well.
As the DOF of the numerator is 2, and the DOF of the
denominator is 18 in the preliminary experiment, the value
of F05 (f1, f2) is identified as 3.55 [17]. Therefore, based on
the result of ANOVA, the values of F for melt temperature,
packing pressure, and injection time appear to be more
than 3.55 and are thus identified as significant factors. The
ANOVA result of the simulation experiment shows the melt
temperature (factor B) to have a great influence on residual
stress. The percentage contribution of this factor is 84.40%.

Factors/ Flexural Stress at Table 5.—ANOVA summary for the simulation experiment.
trials no. A B C D modulus (dB) yield (dB)
Column Factors DOF S V F P
1 220 70 0.2 10 63.47 31.35

2 220 80 0.6 20 63.50 31.33 A Mold temperature ( C) 2 0
83 0
42 1
75 2
07
3 220 85 1.0 30 63.59 31.47 B Melt temperature ( C) 2 33
80 16
90 70
42 84
40
4 243 70 0.6 30 63.62 30.16 C Injection time (s) 2 1
88 0
94 3
92 4
69
5 243 80 1.0 10 63.33 31.21 D Packing pressure (%) 2 2
28 1
14 4
75 5
70
6 243 85 0.2 20 63.22 31.19 E Packing time (s) 2 0
05 0
03 0
13 0
12
7 260 70 1.0 20 63.49 31.47 F Cooling time (s) 2 0
003 0
00 0
01 0
01
8 260 80 0.2 30 63.35 31.30 All other/error 5 1
21 0
24
9 260 85 0.6 10 63.48 31.43 Total 17 40
05
 206
Packing pressure (factor D) is the second most significant
factor at 5.70%, followed by injection time (factor C)
at 4.69%. For mold temperature (factor A) and packing
time (factor E), the percentage contribution is 2.07% and
0.12%, respectively. Cooling time (factor F) makes the least
contribution at only 0.01%. This implies that the cooling
time does not substantially affect the quality characteristic
selected. Based on the ANOVA results, melt temperature,
packing pressure, and injection time are selected as the main
processing parameters. Due to machine limitations, packing
time is selected as the secondary processing parameter. All
four parameters are used to produce the recycled toolbox
tray in the principal experiment.
As this work focuses on the impact of processing
parameters on the mechanical performance of the toolbox
tray made of recycled plastic, emphasis is given on the
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014
four processing parameters. All the results obtained from
the mechanical testing determine the flexural modulus and
stress at yield (in MPa) and are analyzed and discussed
statistically based on the Taguchi optimization method
concept. Instead of using the average value to analyze
the flexural modulus and stress at yield, this work uses
the S/N ratios to convert the experimental results as a
single response. The S/N ratio is the statistical quantity
representing the power of a response signal divided by the
power of the variation in the signal due to noise. Using
the S/N ratio data (in dB), the average performance or
main effects for each factor are computed. The changes in
flexural modulus and stress at yield due to the variations in
injection molding processing parameters can be evaluated
through the main effect graph, and the optimal sets of
processing parameters/levels can be predicted further. In
predicting the optimal sets, the S/N ratio can be divided
into three types: the smaller the better, the nominal the
best, and the bigger the better. In this work, the bigger
the better quality characteristic is chosen to improve the
mechanical properties of the part produced through the
optimal levels of processing parameters. ANOVA is utilized
to present the influence of processing parameters on the
flexural modulus and stress at yield of the part. The details
of the optimal processing parameters and their levels as
well as the significance of the factors determined from the
ANOVA analyses are discussed in the following section.
Optimal Processing Parameters and Their Levels
PIM involves complex processes including plastication,
injection, packing, cooling, ejection, and process/part
quality control applications [32]. Having an initial
knowledge about the process is necessary because of
the current competitive conditions, which are concerned
with faster and cheaper production with higher product
quality. Although PIM is one of the most widely used
processes in plastic manufacturing, the characteristics
of PIM products are easily affected by variations in
processing parameters. Effectively controlling all the
influential processing parameters during the manufacturing
process through an appropriate optimization method is
critical. The determination of the desired combination of
processing parameters from experience and intuition of
operators is a time-consuming process. Thus, Taguchi
N. M. MEHAT AND S. KAMARUDDIN
methods have been used widely in engineering analysis
to optimize the performance characteristics by setting the
design parameters. To determine the optimal processing
parameters for the selected quality characteristics in this
work, the main effects of the results presented in Table 4
are computed. Figure 3 shows the main effect graph for
flexural modulus (denoted as S/N1FM ) and stress at yield
(denoted as S/N2FS ).
Figure 3 shows that the flexural modulus decreases with
the increase in melt temperature (factor A) and packing
pressure (factor B). However, higher melt temperature
and packing pressure, for example, 260 C (A3) and 85%
(B3), demonstrate higher flexural modulus for the recycled
toolbox tray. The melt temperature is the temperature of the
material as it leaves the injection molding machine (through
the nozzle) and enters the mold through the feed system
(i.e., sprue, runner, and gate). Due to contamination in the
recycled material, higher melt temperature is required to
give higher mobility to the molten material. More molten
material being packed into the cavity requires a higher
packing pressure. The flexural modulus of the recycled
toolbox tray increases with the injection time (Factor C).
However, a longer melt material injected into the cavity
decreases the flexural modulus of the part produced. But,
for packing time (factor D), a longer melt material packed
into the cavity generally increases the flexural modulus of
the part produced.
For stress at yield (denoted as S/N2FS ), the similar effect
is shown by the melt temperature (Factor A). Increasing the
melt temperature initially gives a lower stress at yield for the
recycled toolbox tray. However, the increment of the melt
temperature eventually increases the stress at yield. This
effect is observed at a melt temperature of 260 C (Fig. 3). A
similar result is also demonstrated by injection time (Factor
C). But, for packing pressure (Factor B), the stress at yield
increases with the increment of packing pressure causing
more melt material to be packed into the cavity and higher
stress at yield. For packing time (Factor D), a longer time
for the melt material to be packed into the cavity decreases
the stress at yield of the recycled toolbox tray.
This study focuses on the effects of processing parameters
on the mechanical properties of the recycled toolbox tray;
thus, higher values of the flexural modulus and stress at yield
are most desirable. Therefore, for the principal experiment,
the larger the better quality characteristic is chosen. The
values of S/N ratio are intended to be large. A high value
of S/N ratio implies that the signal is much higher than the
random effect of noise factors. The processing parameters
with the highest S/N always yield the optimum quality
with minimum variance [33]. In this principal experiment,
the processing parameters with the highest S/N ratio result
in higher value of flexural modulus and stress at yield.
As a result, A1B1C2D3 and A3B3C3D2 are identified as
the optimal processing parameters for the flexural modulus
and stress at yield, respectively. All the optimal processing
parameters are encircled (Fig. 3). For the flexural modulus,
the set of optimum conditions is recognized as the melt
temperature of 220 C, packing pressure of 70%, injection
time of 0.6 s, and packing time of 30 s. For stress at yield,
the set of optimum conditions is identified as the melt
 EFFECTS OF INJECTION MOLDING PARAMETERS 207
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014

Figure 3.—Main effects graph of the principal experiment.

temperature of 260 C, packing pressure of 85%, injection Table 7.—ANOVA summary for the stress at yield (S/N2FS ).
time of 1.0 s, and packing time of 20 s.
Column Factors DOF S V F P

Significance of the Factors A 

Melt temperature ( C) 2 0.57 0.29 – 43
34
To investigate which parameters significantly affect B Packing pressure (%) 2 0.23 0.11 – 17
13
C Injection time (s) 2 0.27 0.13 – 20
51
the quality characteristics and determine the percentage D Packing time (s) 2 0.25 0.13 – 19
02
contribution of each factor, ANOVA is conducted. The All other/error 0 0.00 0
00
summary of the DOF, S, V, F, and P for the modulus and Total 8 1.32 100
00
stress at yield is presented in Tables 6 and 7, respectively.
Referring to Tables 6 and 7, all the processing parameters
selected in the principal experiment show P values greater packing pressure show 19.02% and 17.13% contribution,
than 5%. As a level of confidence of 95% is used in this respectively.
study, all the processing parameters significantly affect the
quality characteristics. In Table 6, the most significant factor Estimation of the Maximum Flexural Modulus
for the flexural modulus of the recycled toolbox tray is and Stress at Yield
injection time (Factor C), which is about 40.49%. From the
set of optimal processing parameters/levels, the injection The estimation of the maximum flexural modulus and
time is 0.6 s. This value is recommended by the Moldflow stress at yield is determined according to the optimized
analysis in the simulation experiment. Packing pressure processing parameters/levels for this recycled toolbox tray.
(Factor B) is the second most influential factor, followed by The optimum processing parameters/levels are A1B1C2D3
melt temperature (Factor A) and packing time (Factor D). and A3B3C3D2 for the flexural modulus and stress at yield,
However, for stress at yield, the most significant factor is respectively. As all values of the flexural modulus and stress
melt temperature, with a percentage contribution of about at yield in the principal experiment are measured in S/N
43.34% (260 C at level 3 as stated in Table 3). This ratios, the predicted S/N ratios of both quality characteristics
melt temperature determines the main effects (Fig. 3) and are as follows:
demonstrates higher stress at yield with the increment of
melt temperature. The second most influential factor is 1) For flexural modulus,
injection time with a percentage contribution of 20.51%
(1.0 s from C3). The longer the time the melt is injected A1B1C2D3 = m + A1 − m  + B1 − m 
into the cavity, the more the stress at yield of the part + C2 − m  + D3 − m  (1)
produced is increased, as the factors of packing time and
2) For stress at yield,
Table 6.—ANOVA summary for the flexural modulus (S/N1FM ).
A3B3C3D2 = m + A3 − m  + B3 − m 
Column Factors DOF S V F P
+ C3 − m  + D2 − m  (2)
A Melt temperature ( C) 2 0.03 0.01 – 20
67
B Packing pressure (%) 2 0.03 0.01 – 20
93
C Injection time (s) 2 0.05 0.03 – 40
49 where m is the mean S/N ratio for the nine trials in the
D Packing time (s) 2 0.02 0.01 – 17
91 principal experiment, and Fn is the S/N ratio for factor F at
All other/error 0 0.00 0
00 level n. Based on these equations, the predicted S/N ratios
Total 8 0.13 100
00
of the flexural modulus and stress at yield for the optimized
 208
processing parameters/levels A1B1C2D3 and A3B3C3D2
are 63.74 and 31.84 dB, respectively. These predicted values
are certainly higher than those obtained in the principal
experiment (Table 4).
Confirmation Test
The confirmation test is performed to ensure that
the experimental performance relates to the predicted
performance. Thus, the accuracy analysis of the results
obtained can be observed. In this study, the confirmation
test is performed using the optimum processing
parameters/levels for both flexural modulus and stress
at yield (i.e., A1B1C2D3 and A3B3C3D2, respectively).
Based on these optimum processing parameters/levels, the
parts are produced, and the test specimens are prepared
and mechanically tested using the method discussed earlier.
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014
Results of the confirmation test are summarized in Table 8.
Using recycled plastic (100% rPP), the part produced has
1498.78 MPa and 37.32 MPa flexural modulus and stress at
yield, respectively. These are the highest values that can be
produced from the optimum processing parameters/levels
obtained from the Taguchi optimization method. Selecting
the processing parameter setting with the highest S/N
ratio always yields the optimum quality. By producing
the part with the optimum processing parameters/levels of
A1B1C2D3 and A3B3C3D2 for flexural modulus and stress
at yield respectively, the errors compared with the estimated
values are 2.6% and 4.7%, respectively, both of which are
less than 5%.
Among the optimum processing parameters/levels of melt
temperature at 220 C (A1), packing pressure of 70% (B1),
injection time of 0.6 s (C2), and packing time of 30 s (D3),
the most influential factor on flexural modulus is injection
time. The optimum time for the molten material to fill
the cavity is 0.6 s. Packing pressure is the second most
significant factor affecting the flexural modulus. Packing
pressure of 70% is appropriate for the molten material to
be packed into the cavity and provide enough melt volume
in the curing stage. For the stress at yield, among the
melt temperature of 260 C (A3), packing pressure of 85%
(B3), injection time of 1.0 s (C3), and packing time of 20 s
(D2), the melt temperature is the most important processing
parameter. The melt temperature affects the flow rate and
curing time of the melt material. Due to contamination in
the recycled plastics, a higher temperature is required to
melt the material and give higher mobility. Thus, this factor
is crucial to the injection molding process especially during
the filling stage. Injection time is the second most significant
factor affecting stress at yield. The optimum time for the
melt material to fill the cavity is 1.0 s.
Table 8.—Results of the estimation and confirmation tests.
Principal experiment
Flexural modulus (MPa) Stress at yield (MPa)
Estimated value 1538.15
Experimental value 1498.78
Error (%) 2.6
N. M. MEHAT AND S. KAMARUDDIN
By identifying the significant processing parameters
affecting the mechanical properties of the recycled toolbox
tray and their optimum values, better quality of the recycled
toolbox tray can be achieved. When mechanical properties
of the recycled plastic toolbox tray are optimized, usage of
the part as a container for goods such as spanners, screw
drivers, and many others will endure.
Conclusion
PIM is a complex process involving many processing
parameters for consideration. In the actual operations,
processing parameters are often selected from references
or handbooks and are subsequently adjusted by trial
and error. This is a common practice that has been
adopted for economic reasons. Despite the merits of this
approach, it consumes and wastes many materials. It is
also costly, time consuming, and highly dependent on the
experience and intuition of the operators. Integrating the
Taguchi optimization method for conducting simulation
experiments in the preliminary stage can support plastic
manufacturers to predict and overcome problems especially
in selecting the appropriate processing parameters. In PIM,
incoherent processing parameters during the manufacturing
process can either lead to significant improvement of the
injection-molded plastic parts or to the degradation of
accuracy, shape, and surface finish of the final parts, as well
as other part characteristics related to mechanical properties.
Utilizing the simulation packages in the preliminary stage
can facilitate the selection of the appropriate materials and
improve part and mold design. Therefore, the performance
of the plastic part can be predicted before the real
process manufacturing takes place. Integrated simulation
packages using the Taguchi method reduces the number
of replications in conducting the experiment and reduces
acrimonious conditions during optimization. Implementing
both simulation and the Taguchi method in the earliest
stage of manufacturing of plastic parts not only reduces
time consumption and manufacturing cost but also provides
engineers with visual and numerical feedback on the
behavior that will eventually reduce defects on the molded
product, particularly when dealing with recycled materials.
Acknowledgment
The authors acknowledge the Short-term Research Grant
provided by University Sains Malaysia, Pulau Pinang, for
funding the study that resulted in this article.
References
1. Huang, M.S.; Lin, T.Y. Simulation of a regression-model and
PCA based searching method developed for setting the robust
injection molding parameters of multi-quality characteristics.
International Journal of Heat and Mass Transfer 2008, 51, 5828–
5837.
2. Cardozo, D. Three models of the 3D filling simulation for
injection Molding: A Brief Review. Journal of Reinforced Plastics
and Composites 2008, 27, 1963.
3. Henz, B.J.; Mohan, R.V.; Shires, D.R. A hybrid global–local
39.08 approach for optimization of injection gate locations in liquid
37.32 composite molding process simulations. Composites Part A 2007,
4.7
38, 1932–1946.
 EFFECTS OF INJECTION MOLDING PARAMETERS
4. Tsai, K.M.; Hsieh, C.Y.; Lo, W.C. A study of the effects of
process parameters for injection molding on surface quality of
optical lenses. Journal of Materials Processing Technology 2009,
209, 3469–3477.
5. Packianatheri, M.S.; Drakei, P.R.; Rowlands, H. Optimizing the
parameters of multilayered feed forward neural networks through
taguchi design of experiments. Quality Reliability Engineering
International 2000, 16, 461–473.
6. Anthony, J.; Perry, D.; Wang, C.; Kumar, M. An application of
Taguchi method of experimental design for new product design
and development process. Assembly Automation 2006, 26 (1),
18–24.
7. Nellian, S. Integration of Taguchi Design of Experiments
and Finite Element Method for Robust Design Msc.
Thesis, Department of Mechanical Engineering, Massachusetts
University: USA, 1996.
Downloaded by [University of Newcastle (Australia)] at 11:19 01 September 2014
8. Pan, L.K.; Wang, C.C.; Wei, S.L.; Sher, H.F. Optimizing
multiple quality characteristics via Taguchi method-based Grey
analysis. Journal of Materials Processing Technology 2007, 182,
107–116.
9. Gaitonde, V.N.; Karnik, S.R.; Mata, F.; Davim, J.P. Taguchi
approach for achieving better machinability in unreinforced
and reinforced polyamides. Journal of Reinforced Plastics and
Composites 2008, 27, 909–924.
10. Taguchi, G. Introduction to Quality Engineering; Mc Graw-Hill:
New York, 1990.
11. Peace, G.S. Taguchi Methods; Addison-Wesley Publishing
Company: Taipei, 1993; pp. 281–291.
12. Yong, W.H.; Tarng, Y.S. Design optimization of cutting
parameters for turning operations based on Taguchi method.
Journal Material Process Technology 1998, 84, 122–9.
13. Kim, Y.J.; Cho, B.R. Economic consideration on parameter
design. Quality and Reliability Engineering International 2000,
16, 501–514.
14. Antony, J. Simultaneous optimisation of multiple quality
characteristics in manufacturing processes using Taguchi’s
quality loss function. International Journal of Advanced
Manufacturing Technology 2001, 17, 134–138.
15. Liao, S.J.; Chang, D.Y.; Chen, H.J.; Tsou, S.J.; Ho, J.R.; Yau,
H.T.; Hsieh, W.H.; Wang, H.T.; Su, Y.C. Optimal process
conditions of shrinkage and warpage of thin-wall parts. Polymer
Engineering Science 2004, 44 (5), 917–928.
16. Ozcelik, B.; Sonat, I. Warpage and structural analysis of thin
shell plastic in the plastic injection molding. Materials and Design
2009, 30, 367–375.
17. Roy, R.K. A Primer on the Taguchi Method; Van Nostrand
Reinhold: Hanser, New York, 1990.
18. Huang, M.C.; Tai, C.C. The effective factors in the warpage
problem of an injection-molded part with a thin shell feature.
Journal of Materials Processing Technology 2001, 110, 1–9.
19. Tang, S.H.; Tan, Y.J.; Sapuan, S.M.; Sulaiman, S.; Ismail, N.;
Samin, R. The use of Taguchi method in the design of plastic
209
injection mould for reducing warpage. Journal of Materials
Processing Technology 2007, 182, 418–426.
20. Chen, C.P.; Chuang, M.T.; Hsiao, Y.H.; Yang, Y.K.; Tsai,
C.H. Simulation and experimental study in determining injection
molding process parameters for thin-shell plastic parts via design
of experiments analysis. Expert Systems with Applications 2009,
36, 10752–10759.
21. Oktem, H.; Erzurumlu, T.; Uzman, I. Application of Taguchi
optimization technique in determining plastic injection molding
process parameters for a thin-shell part. Materials and Design
2007, 28, 1271–1278.
22. Chang, T.C.; Faison, E. Shrinkage behavior and optimization of
injection molded parts studied by the Taguchi method. Polymer
Engineering Science 2001, 41, 703–710.
23. Altan, M. Reducing shrinkage in injection moldings via the
Taguchi, ANOVA and neural network methods. Materials and
Design 2010, 31, 599–604.
24. Chen, R.S.; Lee, H.H.; Yu, C.Y. Application of Taguchi’s
method on the optimal process design of an injection molded
PC/PBT automobile bumper. Composite Structures 1997, 39 (3–
4), 209–214.
25. Liu, S.J.; Yang, C.Y. Application of statistic estimation to
weldline strength of injection molded termoplastic composites.
Journal of Reinforced Plastics and Composites 2004, 23, 1383.
26. Xie, L.; Ziegmann, G. Influence of processing parameters on
micro injection molded. Microsystem Technology 2009, 15,
1427–1435.
27. Fung, C.P.; Kang, C.P. Multi-response optimization in friction
properties of PBT composites using Taguchi method and
principle component analysis. Journal of Materials Processing
Technology 2005, 170, 602–610.
28. Kuo, H.C.; Jeng, M.C. Effects of part geometry and injection
molding conditions on the tensile properties of ultra-high
molecular weight polyethylene polymer. Materials and design
2010, 31 (2), 884–893.
29. Yang, Y.K.; Shie, J.R.; Liao, H.T.; Wen, J.L.; Yang, R.T.
A study of Taguchi and design of experiments method in
injection molding process for polypropylene components. Journal
of Reinforced Plastics and Composites 2008, 27, 819.
30. Fung, C.P.; Tien, Y.F. Study of multiresponse optimization
for fiber-reinforced poly(Butylene Terephthalate). Journal of
Reinforced Plastics and Composites 2005, 24, 923.
31. MoldFlow Corporation. MoldFlow Plastic Insight Release 6.1;
MoldFlow: Framingham, MA, 2008.
32. Ozcelik, B.; Erzurumlu, T. Comparison of the warpage
optimization in the plastic injection molding using ANOVA,
neural network model and genetic algorithm. Journal of Materials
Processing Technology 2006, 171, 437–445.
33. Teng, S.Y.; Hwang, S.J. Predicting the process induced warpage
of electronic packages using the P–V–T–C equation and
the Taguchi method. Microelectronics Reliability 2007, 47,
2231–2241.