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

net/publication/338160049

Productivity Improvement through Line Balancing by Using Simulation

Modeling (Case study Almeda Garment Factory)

Article · March 2020

DOI: 10.22094/JOIE.2019.567816.1565

CITATIONS READS

11 1,053

4 authors, including:

Aregawi Yemane Misgna Hailemicheal

Mekelle University Bahir Dar University
9 PUBLICATIONS   17 CITATIONS    1 PUBLICATION   11 CITATIONS   

SEE PROFILE SEE PROFILE

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

waste to energy View project

Thesis View project

All content following this page was uploaded by Aregawi Yemane on 25 December 2019.

The user has requested enhancement of the downloaded file.

Journal of Optimization in Industrial Engineering
Vol.13, Issue 1, Winter & Spring 2020, 153-165
DOI:10.22094/JOIE.2019.567816.1565

Productivity Improvement through Line Balancing by Using

Simulation Modeling (Case study Almeda Garment Factory)
Aregawi Yemanea,*, Gebremedhin Gebremicheal a, Teklewold Meraha a , Misgna Hailemicheal a
a
Industrial Engineering, Faculty of mechanical and Industrial Engineering , Bahirdar University ,Bahirdar , Ethiopia
Received 20 June 2018; Revised 24 July 2019; Accepted 02 June 2019

Abstract
The typical problems facing garment manufacturers are long production lead time, bottlenecking, and low productivity. The most critical
phase of garment manufacturing is the sewing phase, as it generally involves a number of operations or for the simple reason that it’s labor
intensive. In assembly line balancing, allocation of jobs to machines is based on the objective of minimizing the workflow among the
operators, reducing the throughput time as well as the work in progress and thus increasing the productivity. Sharing a job of work between
several people is called division of labor. Division of labor should be balanced equally by ensuring the time spent at each station
approximately the same. Each individual step in the assembly of product has to be analyzed carefully, and allocated to stations in a
balanced way over the available workstations. Each operator then carries out operations properly and the work flow is synchronized. In a
detailed work flow, synchronized line includes short distances between stations, low volume of work in process, precise of planning of
production times, and predictable production quantity. This study deals with modeling of assembly line balancing by combining both
manual line balancing techniques with computer simulation to find the optimal solution in the sewing line of Almeda textile plc so as to
improve productivity. In this research arena software, is employed to model and measure the performance of the existing and proposed
sewing line of the federal police trousers sewing line model. For each operation, the researchers have taken 15 sampling observations using
stopwatch and recorded the result. All the collected data are statistically analyzed with arena input analyzer for statistical significance and
determination of expressions to be used to the simulation modeling; SAM is also calculated for these operations to be used to the manual
line balancing. An existing systems simulation model is developed and run for 160 replications by the researchers to measure the current
performance of the system in terms of resource utilization, WIP, and waiting time. The existing systems average utilization is 0.53 with a
line efficiency of 42%. This study has developed a new Sewing assembly line model which has increased the system utilization to 0.69 at a
line efficiency of 58.42% without incurring additional cost.

Keywords: Line Balancing; Productivity; SAM; Simulation; Trouser; WIP

1. Introduction are seeking various effective ways to improve their

industrial productivity through minimization of wastes
Historically, many countries began their process of without hampering the product quality (Dr.Kesavan,
industrialization through focusing on labor-intensive Elanchezhain, & vijaya, 2008). The demand for higher
industries, typically the textile industry. This industry was value at lower price is increasing and to survive, garment
at the forefront of industries leading the Industrial manufacturers must need to improve their operations
Revolution in the United Kingdom from the mid- through producing right first time quality while reducing
eighteenth to the mid-nineteenth centuries. Japan is waste (Khan, 2013). Assembly line is an industrial
another nation that exploited the labor-intensive nature of arrangement of machines, equipment’s and workers for
the textile industry to promote industrialization and the continuous flow of work pieces in mass production
absorption of the country’s abundant labor (fukishini & operation. Manufacturing a product in an assembly line
yamagata, 2014). The case is the same with our country requires partitioning the total amount of work into a set of
Ethiopia; the industry have shown a great growth since elementary operations called tasks (Breginski, Cleto, &
the US preferential trade policy AGOA (the African Junior, 2013).
Growth and Opportunity Act) was signed into law in Therefore sewing process is of critical importance and
2000. It has provided garment Industries in sub-Saharan needs to be planned more carefully. As a consequence,
Africa with duty-free and quota-free access to the US good line balancing with small stocks has to be drawn up
market (otsuka & sonobe, 2011). Due to this many textile to increase the efficiency and quality. So the aim of
industries have been installed in the country to meet the assembly line balancing in sewing line is to assign tasks
extended demand of customers. to the workstations, so that the machines of the
Nowadays, it becomes vital to maximize utilization of the workstation can perform the assigned task with a balanced
resources, working efficiency of the employees, and loading with different labor skill levels.
increase production
*Corresponding capacity
author Email ofaregawi_yemane@yahoo.com
address: the industries to meet the This research project will provide some suggestive
growing demands. For this reason, garment manufacturers remarks to the Almeda garment manufacturers about their

153
 Aregawi Yemane et al./ Productivity Improvement Through Line…

industrial productivity improvement and cost reduction industries like Almeda textile. In product layout machines
along with the implemented tools. are arranged according to the product manufacturing
Out of the four sections of garment department of textile sequences. It is a layout in where workstations or
companies sewing section is mostly prone to various departments are arranged in a linear path. This strategy is
problems since most tasks are complex, labor intensive also known as line flow layout.
and are operated over extended (long) operations; some of Design of the workstation layout widely vary from one
the problems that are being faced in the aforementioned operation to another depending on size of work, number
company are; Operators do not meet their standard target of components to be worked on and type of machine to
due to a large queue of pieces of cloths, Idleness of handle during operation. An efficient layout in plant could
operators, and over stretched (too long) line since most help to reduce the production cycles, work-in-progress,
operations which can be done by one operator and one idle times, number of bottlenecks, material handling times
machine are made to be done by adding another operator and increase the productivity (Dwijayanti, 2010) (Chan,
and machine which eventually causes an over stretched Hui, & Yeung, 1998).
line.
2.1 Productivity improvement Techniques
2. Literature Review
Higher productivity brings higher profit margin in a
Productivity is a measure of the efficiency and business. And increment in productivity level reduces
effectiveness to which organizational resources (inputs) garment manufacturing cost. Hence, factory can make
are utilized for the creation of products or services more profit through productivity improvement. Machine
(outputs) (Bheda, 2014) (Dr.Kesavan, Elanchezhain, & productivity as well as labor productivity increases when
vijaya, 2008). In readymade garments industry, output can a factory produces more pieces by the existing resources
be taken as the number of products manufactured, whilst such as manpower, time and machinery. In garment sector
input is the people, machinery and factory resources productivity improvement is defined as the improvement
required to create those products within a given time of the production time and reduction of the wastage
frame. In fact, in an ideal situation, input should be (Sudarshan & Nageswara, 2014).
controlled and minimized whilst output is maximized.
Productivity can be expressed in many ways but mostly 2.1.1 Work measurement (Time Study)
productivity is measured as labor productivity, machine
productivity or value productivity (Sudarshan & Besides other sectors work study can also be used in
Nageswara, 2014) (Khatun, 2016). In simple words garments sector which includes method study and work
productivity is the quantitative relationship between what measurement. Time study is a work measurement
we produce output and the resources inputs which are technique for recording the times of performing a certain
consumed (s.anilkumar & N.suresh, 2006). Productivity specific job or its elements carried out under specified
gains are vital to the economy; because they allow us to conditions, and for analyzing the data to obtain the time
accomplish more with less. A garment production system necessary for an operator to carry it out at a defined rate
is a way how fabric is being transformed into a garment in of performance. Most common methods of work
a manufacturing system. Production systems are named measurements are stopwatch time study, historical time
according to the various factors such as number of study, predetermined motion time system (PMTS) and
machine used to make a garment, machines layout, total work sampling. Among these time study by stopwatch is
number of operators involved to sew a complete garment considered to be one of the most widely used means of
and number of pieces moving in a line during making a work measurement. Time study leads to the establishment
garment. Among the various production systems of work standard. Development of time standard involves
progressive bundle system and one piece flow system are calculation of three times such as observed time (OT),
most commonly found in the readymade garments normal time (NT) or basic time (BT) and standard time
industries (Chen, Chen, Su, Wu, & Sun, 2012) (Bheda, (ST).
2014) (ashkan, hamid, & hesam, 2014.) (bobby & jenson, Time study concept was originally proposed by Fredrick
2013) . Taylor (1880) and was modified to include a performance
rating adjustment. Time study helps a manufacturing
Facility layout is the most effective physical arrangement company to understand its production, investigate the
of machines, processing equipment and service level of individual skill, planning and production control
departments to have maximum co-ordination and system etc (Kanawaty, 1992). One problem of time study
efficiency of man, machine and material in a plant is the Hawthorne effect where it is found that employees
(Syduzzaman & Golder, 2015) (Labour management in change their behavior when they come to know that they
development , 2001). A good layout scheme would are being measured (Farhatun, 2016)
contribute to the overall efficiency of operations. Layouts
can be classified into four classes such as product layout, 2.1.2 Assembly Line Balancing
process layout, group technology layout and fixed
position layout (Chan, Hui, & Yeung, 1998). Among Line Balancing is leveling the workload across all
those product layout is mostly used in the garments processes in a value stream to remove bottlenecks and

154
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

excess capacity (Saptari, Lai, & Salleh, 2011). The main Application of fishbone analysis in garment industries is
objective of line balancing is to distribute the task evenly essential to identify various problem areas for
over the work station so that idle time of man or machine productivity improvement.
can be minimized (Kumar & Mahto, Productivity
Improvement through Process Analysis for Optimizing 2.1.4 Simulation Modeling
Assembly Line in Packaging Industries, 2013). Assembly
line may be classified as single model assembly line, Simulation modeling is a common paradigm for analyzing
mixed model assembly line and multi model assembly complex systems. In a nutshell, this paradigm creates a
line (Amardeep & Gautham, 2013). simplified representation of a system under study. This
Assembly line balancing in Indian garment industries simulation model then proceeds to experiment with the
improved the productivity by decreasing the total system, guided by a prescribed set of goals, such as
equipment cost and number of work stations (Kumar & improved system design, cost–benefit analysis, sensitivity
Mahto, 2013) (Amen, 2000). To meet the production to design parameters, and so on (davidKelton, 2006).
target, maintaining level work flow in the line is very Modeling is the enterprise of devising a simplified
essential. Line balancing can be classified as initial line representation of a complex system with the goal of
balancing, rebalancing, reactive balancing and late hour providing and Predictions of the system’s performance
balancing (Breginski, Cleto, & Junior, 2013). measures (metrics) of interest.
Line balancing is very effective technique in improving
productivity; for example in Bangladeshi garment Modeling and simulation are potential tools for analyzing
industry labor productivity was increased by 22% with the as well as studying sewing assembly lines in garment
application of line balancing techniques (Shumon, Arif- industries (Bahadir & senem, 2013).
Uz-Zaman, & Rahman, 2010).
3. Methodology
2.1.3 Fishbone Analysis
The methodology as shown in the figure below is to be
The fishbone analysis is a tool to evaluate the business followed to improve the operational performance of the
process and its effectiveness. It is defined as a fishbone production system.
because of its structural outlook and appearance (Mahto
& Kumar, 2008) (Bose, 2012). Because of the function of 3.1 Product Selection for the Study
the Fishbone diagram, it may be referred to as a cause-
and-effect diagram. Fishbone diagram mainly represents a Product selection is critical and crucial step as it provides
diagrammatic model of suggestive presentation for the focus to the project and produce tangible improvements in
correlations between an event (effect) and its multiple
an effective manner. Trying to solve all problems at the
happening causes. A cause-and-effect diagram can help to
identify the reasons why a process goes out of control and same time creates confusion, inefficient use of resources
why it is not able to meet the standard. It helps to identify and delays. Product selection refers to the process of
root causes and ensures a common understanding of the identifying a “product” or “family” of similar products to
causes. be the target of an improvement project or study.
Root-cause identification for quality and productivity In our case we will be doing our project in federal police
related problems are key issues for manufacturing trouser since it is prone to problems due to the length and
processes. Tools that assist groups or individuals in
number of operations.
identifying the root causes of problems are known as root-
cause analysis tools. Every equipment failure happens for
a number of reasons and root-cause Analysis is a 3.2 Time Study
systematic method that leads to the discovery of faults or
root cause. A root-cause analysis (RCA) investigation Time study is a technique used to establish a time
traces the cause and effect trail from the end failure back standard to perform a given assembly operation. It is
to the root cause (Bon, Rahman, Bolhassan, & Nordin, based on the measuring the work content of the selected
2013). Fishbone analysis was practiced to evaluate the sewing assembly line, including any personal allowances
supply chain and business process of a Hospital. The and unavoidable delays. It is the primary step required to
analysis reveals that the problem areas are lack of proper
determine the opportunities that improve assembly
equipment, faulty process, misdirected people, poor
materials management, improper environment, and operations and set production standards (s.anilkumar &
inefficient overall management (BGMEA’s Chittagong N.suresh, 2006).
unit, 2009), (Bose, 2012).

155
 Aregawi Yemane et al./ Productivity Improvement Through Line…

Fig. 1. Methodology chart for productivity improvement

3.3 Simulation modeling

For a successful completion of a simulation project it is a

must to follow an established methodology or guide lines.
This simple guide provides a basic framework for
ensuring success with simulation projects.
Step 1: Develop the Functional Specification Document
Step 2: Identify and Collect Data
Step 3: Build the Model
Step 4: Document the model
Step5: Verification and Validation
Step 6: Analysis
Step 7: Project Deliverables

4. Data Collection and Analysis

Fig. 2. A sample of federal police trousers

Table 1
Estimated distributions for processes
No. Operation name Machine type Operator M/C
number number Expression
1 back patch press IRON 1 1 TRIA(25.2, 27.9, 30)
2 mark bkpkt table2 1 0 UNIF(19.3, 24)
3,4 back patch attach S.N.L.S 2 2 UNIF(56, 59.7)
6 bk+pkt+flapatttach APW 1 1 NORM(23.1, 0.745)
6T pick trouser panel table6 1 0 NORM(8.91, 0.38)
5,7,8 bkpkt+flap top s/t S.N.L.S 3 3 NORM(155, 1.15)
9 back pkt bag close 5 TH 1 1 TRIA(50, 52.3, 53)
10 sew back rise MH-380 1 1 NORM(52.2, 0.562)
11A back flap BH BH 1 1 28 + GAMM(0.759, 2.39)
11B thigh flap B.H BH 1 1 NORM(33.5, 0.949)
12 back pkt flap inner s/t S.N.L.S 1 1 44.3 + 4.75 * BETA(1.37, 1.29)
13 thigh pkt flap inner s/t S.N.L.S 1 1 37 + 5 * BETA(1.21, 0.994)
12T turn bkpkt flap table12 1 0 29 + 5 * BETA(1.26, 1.49)
13T turn thigh pkt flap table13 1 0 27 + 5 * BETA(1.16, 1.19)
14 top s/t bkpkt flap S.N.L.S 1 1 34 + 3.69 * BETA(1.36, 1.72)
15 top s/t thigh pkt flap S.N.L.S 1 1 UNIF(37.2, 40)
16 thigh pkt 2carego edge s/t&tuck S.N.L.S 1 1 79 + 12 * BETA(0.676, 0.732)
17 top s/t 2care go S.N.L.S 1 1 66.1 + 9.92 * BETA(1.28, 0.991)
18 att.thighpkt facing edge s/t S.N.L.S 1 1 54.1 + WEIB(2.61, 1.79)
19 att. Thigh pkt facing S.N.L.S 1 1 44.2 + WEIB(2.15, 1.66)
20 top s/t&3rd carego S.N.L.S 1 1 TRIA(79, 83.7, 89)
21 thigh pkt edge press IRON 1 1 50.3 + WEIB(2.91, 2.18)
22 fly o/l 3 TH 1 1 NORM(23.3, 0.61)
23 front rise 3TH 1 1 29.4 + 6.58 * BETA(1.16, 1.55)

156
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

24 knee patch press IRON 1 1 TRIA(16.8, 17.8, 19)

25,26 knee patch attach D.N.L.S 2 2 TRIA(44.5, 47.2, 49.8)
27 tuck front knee patch S.N.L.S 1 1 26.2 + ERLA(0.537, 4)
28 side pktatt.&top s/t S.N.L.S 1 1 TRIA(77, 79.1, 84)
29 outer side pkt top s/t S.N.L.S 1 1 69.1 + ERLA(1.09, 4)
30 side pkt bag close 5 TH 1 1 TRIA(38, 41.4, 44)
31 buttone hole left fly BH 1 1 NORM(129, 3.06)
32 side pkt side tuck S.N.L.S 1 1 NORM(56.5, 0.613)
33 att.&top s/t left fly S.N.L.S 1 1 141 + 10 * BETA(1.13, 1.63)
34 tuck left fly+J-s/t S.N.L.S 1 1 45 + 4 * BETA(1.74, 1.17)
35 att.&top s/t righ fly S.N.L.S 1 1 TRIA(104, 110, 117)
36 front rise attach D.N.L.S 1 1 TRIA(48, 50.3, 52.5)
37,38T machnumberig table 37,38 2 0 NORM(32.8, 0.842)
37,38 sew side seam 5 TH 2 2 NORM(88.1, 0.665)
39,40 top s/t sew side seam F.O.A 2 2 85 + ERLA(0.589, 2)
39,40T mark thigh pktpostion table 39,40 2 0 TRIA(45, 46.3, 47.5)
40,39T match thigh pkt flaps table 40,39 2 0 27 + 2.4 * BETA(0.794, 1.06)
41,42,44 thigh pkt attach S.N.L.S 3 3 126 + 3.44 * BETA(1.03, 1.29)
43 flap attach D.N.L.S 1 1 53 + LOGN(0.548, 0.292)
45 close inseam 5 TH 1 1 TRIA(45.6, 46.9, 48.3)
46 In seam top s/t F.O.A 1 1 TRIA(42.5, 44, 45.5)
55BLM belt loop mak&cut BLM 1 1 NORM(26.6, 0.508)
47,48 mark belt loops tuck S.N.L.S 2 2 NORM(59.3, 0.659)
48 F wast band press fusing m/c 1 1 28 + GAMM(0.503, 4.34)
49,50 w.band attach S.N.L.S 2 2 NORM(56.5, 0.53)
51,52 w.band edge s/t S.N.L.S 2 2 NORM(68.9, 0.714)
53,54 w.band close & label S.N.L.S 2 2 TRIA(70.2, 71.1, 72)
56,57 belt loop lower tuck S.N.L.S 2 2 73.7 + LOGN(0.975, 0.699)
58 button hole BH 1 1 NORM(38.9, 0.777)
58T insert ropes table 58 1 0 TRIA(27, 29.1, 31)
59,60 bottom hem S.N.L.S 2 2 62.5 + WEIB(1.75, 2.44)
61T mark butten set postion table 61 1 0 TRIA(23, 24.5, 25.6)
61 button attach BA 1 1 TRIA(73, 75.7, 77)
62,63,64 bar tuck BT 3 3 TRIA(157, 160, 161)
65 trimming table 65 6 0 273 + LOGN(1.73, 1.27)
66 inspection table66 2 0 TRIA(57.1, 60.1, 61)
67 ironing big IRON 2 2 TRIA(17.2, 18.6, 20)
68 folding + packaging table 68 3 0 57 + WEIB(1.19, 1.94)

5. Development of Simulation Model for the Existing  450 minutes working time does not include breaks,
Production System  Set-up times are not taken into consideration while
modeling the system, because in a real system the
As mentioned in our methodology, the study of the as is setup process is usually carried out at the end of the
system or just the existing system involves simulation working time,
modeling using arena software. The aim of developing the  The trousers assembly line is never lacking input
model for the existing system is for bottleneck materials from cutting section.
identification and further experimentation using various  Transportation of raw materials (bundle
scenarios. The construction of the model is based on a man/woman) is performed by workers who aren’t
production process flow of the company’s sewing line. used for sewing operations.
This study represents discrete-event modeling and the In this study, among other products in the sewing section
sewing line works for 450 minutes (7.5 hours) in a day. the case under consideration is the production of federal
The following assumptions are taken in to consideration police trousers (line 29 30). The production of federal
while modeling the system; police trousers (line 29 30) consists of a total of 88
operators. The line works for 7.5 hrs per shift. To model
 At the beginning of each order, the production line and run the simulation model, the seven phase procedures
is assumed to begin empty, outlined in the methodology of the study have been used.
 There is no maintenance process performed during
the working period,

157
 Aregawi Yemane et al./ Productivity Improvement Through Line…

Fig. 3. Simulation model of the existing production system

5.1 Model validation and verification found to be idle in the actual system were having lower
utilization in the simulation model.
A model is simplified representation of a real system
which includes the most important system components 5.2 Model verification
and the interaction between them, a model should
represent the real system a model cannot represent the real One way for verification is by reviewing the SIMAN code
system exactly rather it can approximate the system how and check if the model is performing exactly what was
it behaves and interact. This is mainly due to the planned for it to do. But, this way of verification requires
assumptions made while developing the model. The a skill in the SIMAN programming language. So the best
performance measures extracted from a model will only way for us to verify the model is by looking at the
represent the real system if the model is a good animation. In this verification method it’s allowed only a
representation of the system. So Verification and single entity to enter the system and follow that entity to
validation of the simulation model of this study were be sure that the model logic and data are correct. In this
carried out using the daily production statistic validity by simulation model, allowing only a single entity to go
comparing through the system has shown that the entity is going
The actual system and the simulation model results for the through every module as per programmed.
existing federal police trousers model. We made
simulation trial runs under a variety of settings of the run 5.3 Simulation Run Results and Interpretation
parameters, and checked the model output result for its
appropriateness. We have taken the real (actual) data (i.e. Although, there are many possibilities to manipulate the
the plant produces 364 per shift on average) from the developed simulation model, this study has addressed two
production line. The simulation model output is tested to major options. Since the objective of this thesis is to
run for 7.5 hrs and its output is 365which is almost the model the sewing assembly line and balance the activities
same with that of actual data and this is a clear indication that assemble the trousers in the sewing line; we have
that the model is valid (w.davidkelton & p.sadowski, identified the bottleneck and idle operations for further
2000). In addition to this some work stations which are analysis so as to balance the sewing line. A bottleneck is

158
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

one process in the chain of processes, such that its limited From the study of the existing simulation model we have
capacity reduces the found out the idle and bottleneck stations. This will be the
capacity of the whole assembly process or simply the input for carrying out line balancing calculations in the
output of the system is determined by this process; we next sections, but before proceeding to that let’s see the
have identified some variations in process capacity from possible root causes of variation between the actual and
the benchmark target and the lower capacity from the target output of the sewing line.
benchmark target is the bottleneck process; as production
flow would stuck on the bottleneck point; Whereas those 5.4. Root causes of variation between the actual and
which work under capacity are termed as idle processes. target output of the sewing line
Total production has been blocked in these seven work
stations and large WIP has been stuck in these bottleneck A cause-and-effect diagram can help to identify the
processes. The bottlenecks of the existing system are reasons why a process goes out of control. It helps to
summarized as follows; identify root causes and ensures a common understanding
Table 2 of the causes (Bose, 2012). We have interviewed the
Bottleneck operations in the existing production system production managers of the garment section, operators,
R-no; Operations Numbers quality assurance manager, planning manager, and
waiting (Items) supervisors of the lines in the production line and
1 Button Hole left fly 31 151
discussed about the causes and the effects of why the
2 Attach & top stitch left fly 33 44.5
company did not meet the planned output; or in other
3 thigh flap B.H 11B 62.5
words the causes of high variation between actual output
4 Button attach 61 75.4
and target output. In this study, we have summarized the
5 back flap B.H 11A 98.4
main causes as follows in the root cause analysis diagram.
6 Thigh pkt 2 cargo edge stitch & 134
tuck 16
7 Side pkt attach & top stitch 28 91.2

Table 3
The idle stations of the existing production system
R-no; Operations Instantaneous Utilization
1 Fly O/l 22 0.3666
2 Ironing 67 0.3775
3 mark button set position 61T 0.3673
4 Match numbering 37, 38T 0.1725
5 mark thigh pkt postion39, 40T 0.2921
6 Pick trouser panel 6T 0.1344
7 Knee patch press 24 0.2820
8 trimming 65 0.5861
9 Back pktflap attach 6 (APW) 0.3487
10 knee patch attach 25 26 0.3706
11 w.band attach 49 50 0.4183
12 mark belt loops tuck 47 48 0.4143
13 mark bkpkt table 2 0.3409
14 Inspection table 66A 0.3788
15 Inspection table 66B 0.3768
16 insert ropes table 58 0.4510
17 turn thigh pkt flap table 12 0.4303
18 turn bkpkt flap table 13 0.4149
19 bottom hemS.N.L.S 59 60 0.4966
20 w.bandattachS.N.L.S 49 50 0.4183
21 back patch attachS.N.L.S 3 0.4541
22 belt loop mak&cutBLM55 0.4180
23 top s/t sew side seamFOA 39 0.4511
24 top s/t sew side seamFOA 40 0.4503
25 wast band pressFusing machine 1 0.4740
26 top s/t bkpktflapS.N.L.S 14 0.4890
27 tuck front knee patchS.N.L.S 27 0.4451
28 tuck left fly+J-s/tS.N.L.S 34 0.3021
29 back patch attachS.N.L.S 4 0.4541
30 thigh pkt attach S.N.L.S 41 0.4520
31 thigh pkt attach S.N.L.S 42 0.4529
32 thigh pkt attach S.N.L.S 44 0.4497

159
 Aregawi Yemane et al./ Productivity Improvement Through Line…

Fig. 4. Root cause analysis for variation between actual and target output

Theoretical Minimum(TM)= a bench mark or goal for the

6. Balancing of the Sewing Assembly Line smallest number of station possible;

In line balancing of operations, the optimum calculating

method of the workstation number and the minimum There for the researchers needed at minimum 42 work
calculating procedure of the cycle time are studied by stations.
using the concept of unbalanced time. These calculating Now let’s see how operations are balanced by accessing
methods are then applied to a sewing process in practical various alternatives in the sewing line;
clothes making (Nakajima, Uchiyama, & Yoshito Miura,
1980) (Patrick & Frency, 1999). Alternative 1
Balancing method is very essential to make the In order to determine bottlenecks and idle stations in the
production flow smoother compare to the previous layout. reference layout model; a number of the existing system’s
Considering working distance, type of machines and machine utilization, waiting time of jobs, average output,
efficiency, workers who have extra time to work after number busy, number waiting etc..as well as desired cycle
completing their works, have been made to share their time were taken into account. It was observed that process
work to complete the bottleneck processes; in some cases 31,33 11B, 61, 11A 16,28 are busy and process 22,
we have also merged some underutilized (idle) operations 67,61T, (37 38T),(39,40T), 6T, 24, 65,6,(25 26), 12T, (49
based on the desired cycle time. Before balancing the line, 50), (47 48) e.t.c are idle. By this way, we have developed
in appendix, time study sheet is attached showing the the first Alternative by merging, decreasing capacity, and
different types of machine used, number of operators and adding of machines in order to overcome the processes
helpers, basic and SPT and cycle time. Process wise problems. The result of this alternative was obtained by
capacity of each work station has been shown in running the simulation model for 160 numbers of
Appendix where SAM has been calculated by adding SPT replications. The following table summarizes results
of each process. We have done the line balancing based on alternative 1, when six extra machines with their
calculations as follows; operator are added to the busy operations and when idle
Desired cycle time= 1/ output rate operations are merged, in the reference model.
Cycle time= 7.5hrs*60min/382
Cycle time=1.18min/piece

Table 4
Performance measures of alternative 1
R no: Performance measures Value
1 The standard deviation for average number of jobs waiting in queues (WIP) 21.8
2 Average daily output/shift 364
3 Average Resource utilization 0.6545
4 Total number of machines the line requires 62
5 Total number of operators the line requires 75
6 Number of newly added machines 6
7 Number of newly added operators 6
8 Standard minute value (SMV) 47.25
9 Labor productivity 4.85
10 Machine productivity 5.87

160
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

Alternative 2
model for 160 numbers of replications. The sewing line
Taking in to account all the conditions of the existing
works 2 shifts a day 7.5 hrs each; so the only suitable time
(reference) model we have developed the second
to schedule for over time is in the lunch hours between
alternative by merging, decreasing capacity, and by letting
11:00 am to 11:30 am. So, only 30 minutes of working
operators work over time instead of adding machine in
hour is to be added to the existing system schedule for
order to overcome the processes problems. The result of
those operations which are busy.
this Alternative was obtained by running the simulation

Table 5
Performance measures of alternative
Rno: Performance measures Value
1 The standard deviation for average number of jobs waiting in queues (WIP) 28.14
2 Average daily output/shift 379
3 Average Resource utilization 0.653
4 Total number of machines the line requires 57
5 Total number of operators the line requires 70
6 Number of newly added machines 0
7 Number of newly added operators Over time for 30min (8)
8 SMV 50.71
9 Labor productivity 5.34
10 Machine productivity 5.97

Alternative 3
Taking in to account all the conditions of the existing So with this concept of sharing and merging of idle and
(reference) model we have developed the third alternative busy operations the average output of the system is 364
by sharing of operational time between idle and busy per shift. Moreover, other performance measures were
operations as well as merging of idle operations in order also considered, as a result it was also observed that
to overcome the processes problem. The result of this average utilization of machines is increased; besides the
alternative was obtained by running the simulation model average staying of jobs (WIP) in queues is also decreased.
for 160 numbers of replications. In reference to the existing production system the
With this alternative we have accessed the existing system performance measures of alternative 3 are summarized in
problems with the available machines and man power. the following table;

Table 6
Performance measures of alternative 3
R.no; Performance measures Value
1 The standard deviation for average number of jobs waiting in queues (WIP) 25.34

2 Average daily output/shift 364

3 Average Resource utilization 0.663
4 Total number of machines the line requires 56
5 Total number of operators the line requires 69
6 Number of newly added machines 0
7 Number of newly added operators 0
8 SMV 52.7
9 Labor productivity 5.28
10 Machine productivity 6.5

Alternative 4 alternative was obtained by running the simulation model

for 160 numbers of replication.
Taking in to account all the conditions of the existing
Having seen the performance measures of the four line
(reference) model; we have developed the fourth
balancing alternatives, we can select the one which is best
alternative by combining the above three alternatives. In
and which can be applied in the sewing line of the
this alternative, things like; sharing of operational time
existing garment section sewing line system. But before
between idle and busy Operations, merging of idle
the selection process we must identify the performance
operations or capacity decreasing, adding of operators
measures which can clearly tell us how much a line is
working hours by means of over time were used in order
balanced. A well balanced line is characterized by a
to overcome the processes problem. The result of this
balanced distribution of jobs across the assembly line,
efficient resource utilization and higher output. So the

161
 Aregawi Yemane et al./ Productivity Improvement Through Line…

performance measures of the four alternatives are comparison;

summed up in the table that follows for further
Table 7
Performance measures of alternative 4
R.no: Performance measures Value
1 The standard deviation for average number of jobs waiting in queues (WIP) 21.6992
2 Average daily output/shift(7.5hrs) 379
3 Average Resource utilization 0.69
4 Total number of machines the line requires 59
5 Total number of operators the line requires 72
6 Number of newly added machines 3
7 Number of newly added operators 3
8 SMV 49.25
9 Labor productivity 5.26
10 Machine productivity 6.42

Table 8
comparison of all alternatives in terms of various performance measures
Rno; Performance measures Alternative 1 Alternative 2 Alternative 3 Alternative 4

1 Average Resource utilization 0.6545 0.653 0.663 0.69

The standard deviation for average
2 number of jobs waiting in queues 21.8 0 28.140 25.34 21.6992
(WIP)
3 Average daily output/shift (7.5hrs) 364 379 364 379

Here, in the above table we can see that the fourth identified based on waiting time and numbers waiting in
alternative is to way better than the rest of three queue; whereas the idle stations are identified based on
alternatives in terms of these performance measures, resource utilization. In the development of the proposed
therefore this fourth alternative model will be applied in model the first thing to be done is to balance the line
the line balancing of the existing sewing line. The four manually using the time study at hand and then interpret it
alternatives of line balancing models are developed to simulation model. To balance a line manually there are
similarly, but here we have only shown their performance various techniques we use like, sharing of operational
measures for the sake of comparison. So it will be time between idle and busy operations, merging of idle
necessary to show all the steps for the selected alternative operations, capacity decreasing, adding of machines and
model; so that readers can have a clear picture of what the of operators. Now let’s see how the line is balanced by
model development will be like. Now let’s see how this combining these various techniques.
model is developed step by step;
Sharing of operational time between idle and busy
operations;
7. Development of the Proposed Model
The following table summarizes the line balancing of
(Alternative 4) processes done by sharing of smv between idle and busy
The Proposed model is built after the analysis of existing operations; keep in mind that for operations to share
model’s performance. During the analysis, the bottlenecks operational time they must use the same resource.
and idle stations were identified. The bottlenecks are
Table 9
Balanced operating time (SMV/ piece) by sharing
R.no; Idle operations Bottleneck operations

Process code SMV Balanced Time Process code SMV Balanced

Time
1 14 0.59 1 16 1.41 1
Remark: Process#14 can work 0.59 min and share work with process#16 for the last 0.41 min
2 15 0.64 0.92 17 1.19 0.92
Remark: Process#15 can work 0.64 min & share work with process#17 for the last 0.27 min
3 19 0.77 1.09 20 1.4 1.09
Remark: Process#19 can work 0.77 min & share work with process#20 for the last 0.31 min
4 27 0.47 0.9 28 1.33 0.9
Remark: Process#27 can work 0.47 min & share work with process#28 for the last 0.43 min

162
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

The researchers have shown how idle and busy operations of trousers panel are done with SMV 0.39 and 0.15
can be balanced by sharing of smv time now let’s see how respectively. So since both are working under capacity,
to balance a line by merging of operations. we can remove 6T and let the work be done by operator 6.
So the new SMV= 0.39+0.15
Merging of operations;
SMV=0.54min/piece
Process3,4 (Eliminate one process) This operation is The rest of the processes are balanced using the same
done by 2 operators who use the same machine (SNLS) techniques; either by decreasing machine (eliminating
with 0.48 min smv, but we can see that they are to way process) or merging of operations.
idle since the cycle time is 1.18min. so we can merge Now we will be developing computer simulation model
these operations to be done by one operator and one for the purpose of experimentation with the system. It will
machine by removing operator 4 along with SNLS4. be developed based on the findings of the manual line
So the new SMV= 2*0.48 balancing which were carried out in the above sections.
SMV=0.96min/piece So before Proceeding to the development of the proposed
simulation model, we must specify how the line balancing
Process 65A,65B,65C,65D,65E&65F: techniques are correlated with arena simulation.
This operation is done by 6 operators who use the same
resource (table) with 0.76 min SMV, but we can see that Table 10
they are to way idle since the cycle time is 1.18min. so we Relating manual line balancing technique to simulation model
can Manual line balancing Technique In simulation model
Remove one operator, so that it will done by 5 operators 1.Sharing of work from idle to busy 1.adding time to standard
operations time ST of busy operation
and 5 table by removing operator 65F along with table from idle
65F. 2.merging different operation have 2.in each process module use
So the new SMV is 6*0.76/5 similar M/C similar resource
SMV= 0.9 min/piece 3.merging one or more similar 3.Decrease capacity in the
process having similar M/C process module
Process 6,6T (Merge): These 6, 6T are side by side
(near) operations where back pkt flap attach and picking

Fig. 5. The newly developed simulation model of the Assembly line

163
 Journal of Optimization in Industrial Engineering Vol.13, Issue 1, Winter & Spring 2020, 149-161

7.1 Proposed Simulation model run results and company’s building layout one sewing line of garment
interpretation section can only be extended up to 36.5meters to the
The aim of this proposed simulation is to show the maximum. For this reason operations such as mark
improved sewing line layout and balancing activities that button set position (table61),button attach (61), bar tuck
assemble the federal police trousers parts and analyze (62,63,64), trimming and inspection are transferred to the
their performance. Therefore, to alleviate the problems the next line code named 29 which is 6.2 meter length to
proposed simulation model for the line balancing of complete the assembly process.
Federal police trousers production, provides the planning
Calculation of length for the new assembly line layout;
manager with a simulation based optimization tool that
helps to gain information without disturbing the actual One work station has 0.9meter length, the gap between
system, and improve system Performance to increase subassembly lines is 1.5 meters, trimming table 2 meter
productivity of the company. and inspection table 1.5 meter; this is according to the
case company actual workstation space and machineries
From this generated output of the proposed simulation
measurement. The researchers have minimized 8
model it can be clearly seen that the utilization of machines and 2 tables; total of 10 work stations.
resources like SNLS 3, SNLS 41 42, SNLS 27, 12T, 37T, Therefore we have minimized; 10*0.9=9 meter, then the
39T and other resources which were idle (underutilized) proposed line length = existing line length (42.7meter)-
in the existing system have dramatically increased minimized length (9meter) =33.7meter. Unlike the
existing assembly line the new assembly line will stretch
7.2 Layout of the existing and proposed model over a one straight 33.7meter line.

The layout model of the existing assembly line is too long Comparison of the existing and proposed model in
having a length 42. 7 meter, but according to the terms of various performance measures

Table 11
Comparison of the existing and proposed Sewing assembly line
Rno: Performance measures Existing Simulation model Proposed Simulation model

1 The standard deviation for average number of jobs waiting in queues 29.98 21.7
(WIP)
2 Average daily output/shift(7.5hrs) 364 trousers 379 trousers
3 Average Resource utilization 0.53 0.69
4 Total number of machines 66 59
5 Total number of operators 88 72
6 Labor productivity 4.14 5.26
7 Machine productivity 5.52 6.42
8 Line Efficiency 42% 58.42%
9 Line length 42.7m 33.7m

proposed model for the trousers sewing line was

8. Conclusion developed by combining both manual line balancing
techniques with computer simulation model. Four
This study addresses the development of a line balancing alternatives were developed and compared with the
model for assembly line-balancing problem; in order to performance of the existing system as well as with each
improve the line balance of federal police trouser’s other using various performance measures and the fourth
sewing line. Initially a simulation model was developed alternatives were found to be optimal. The proposed line
for the existing system since it’s essential to know the balancing model has decreased the number of operators
current situation of a system. Based on the results of the from 88 to 72, the number of machines from 66 to 59 at
existing simulation modeling; Button Hole left fly 31, the same time the daily output per shift is increased from
Attach & top stitch left fly 33, thigh flap B.H 11B, Button 365 to 379 trousers per line. The performance comparison
attach 61, back flap B.H 11A, Thigh pkt 2 cargo edge demonstrates that the proposed line balancing model out
stitch & tuck 16, Side pkt attach & top stitch 28 were performs the industry practice; by increasing the assembly
found to be bottleneck (busy) operations and Fly O/l 22, line efficiency from 42% to 58.42% without incurring
Ironing 67, mark button set position 61T, Match additional costs for machines and operators. Finally this
numbering 37 38T, mark thigh pkt postion39 40T, Pick line balancing model can be applied to the federal police
trouser panel 6T, Knee patch press 24, tuck front knee trousers sewing lines and save 40284 birr which were
patch S.N.L.S 27, Inspection table 66A, Inspection table allocated to machines and 20800birr which were allocated
66B, tuck left fly+J-s/t S.N.L.S 34, mark bkpkt table 2 to the wages of operators over monthly bases.
idle operations. Therefore, to alleviate the problems the

164
 Aregawi Yemane et al./ Productivity Improvement Through Line…

References Dwijayanti. (2010). A proposed study on facility planning

and design in manufacturing proces. International
Amardeep, J. & Gautham, J. (2013). Line Balancing of multiconference of engineers and computer scientists
Single Model Assembly Line. International Journal 2010, , (p. vol III).
of Innovative Research in Science engineering and Elanchezhain, K. & Vijaya, R. (2008). Production
technolegy , pp 23-29. planning and control. chennai: Anuradha
Amen, M. (2000). Heuristic methods for cost-oriented Publications.
assembly line balancing. International Journal of Farhatun, N. (2016). Improving sewing section efficiency
Production Economics , Vol.68, pp. 1-14. through utilizing worker capacity. Scientific and
Ashkan, H., Hamid, K., & Hesam, R. (2014.). Line academic publishing.
balancing the production line by the simulation and Fukishini, T., & yamagata, T. (2014). The garment
statistics technique. industry in low income countries. Houndmills:
Bahadir, & Senem, K. (2013). Assembly line balancing in palgrave macmilan.
garment production by simulation. Kanawaty, G. (1992). Motion and Time study. Geneva:
BGMEA’s Chittagong unit. (2009). Quality and International labour organization.
Productivity Management in perspective to Global Kelton, D. (2006). Simulation with arena. Boston: Mc
Economic Situation. Bheda, D. R. ( 2014). Graw Hill.
Productivity in Apparel Manufacturing Back to Khan, I. (2013). Minimization of Reworks in Quality and
Basics, December 2014 by Dr. Rajesh Bheda. Productivity Improvement in the Apparel Industry.
Bobby, J., & Jenson, J. (2013). analysis and simulation of International Journal of Engineering and Applied
factory layout using arena. Sciences , 148-164.
Bon, A. T., Rahman, A. A., Bolhassan, E., & Nordin, M. Khatun. (2016). Application of I.E in the industry for
A. ( 2013). Optimize Assembly Production Line better productivity.
using Line Balancing. The 2nd International Kumar, N., & Mahto, D. (2013). Assembly Line
Conference on Global Optimization and Its Balancing: A Review of Developments and Trends
Applications. melaka, malaysia. in Approach to Industrial Application. Global
Bose, T. K. (2012). Application of Fishbone Analysis for Journal of Researches in Engineering Industrial
Evaluating Supply Chain and Business process. Engineering, , 29-50.
International Journal of Managing Value and Supply Kumar, N., & Mahto, D. (2013). Productivity
Chains , 17-24. Improvement through Process Analysis for
Breginski, R., Cleto, M., & Junior, J. S. (2013). Assembly Optimizing Assembly Line in Packaging Industries.
Line Balancing Using Eight Heuristics. 22nd Global Journal of Researches in Engineering
International Conference on Production Research. Industrial Engineering , Volume 13-.
Chan, G., Hui, K., & Yeung, K. (1998). Handling the Labour management in development . (2001). Labour
assembly line balancing problem in the clothing management in development journal , vol 2.
industry using a genetic algorithm. International Mahto, D., & Kumar, A. (2008). Application of Root
Journal of Clothing Science and Technology , Cause Analysis in Improvement of Product Quality
Vol.10, pp. 21-37. and Productivity . Journal of Industrial Engineering
Chen, J. C., Chen, C.-C., Su, L.-H., Wu, H.-B., & Sun, C.- and Management , 16-53.
J. (2012). Assembly line balancing in garment
industry. Expert Systems with Applications .

This article can be cited: Yemane, A., Gebremicheal, G., Meraha, T. & Hailemicheal, M. (2020).
Productivity Improvement through Line Balancing by Using Simulation Modeling
(Case study Almeda Garment Factory)
Journal of Optimization in Industrial Engineering. 13 (1), 153-165.

http://www.qjie.ir/article_667920.html
DOI: 10.22094/JOIE.2019.567816.1565

165

View publication stats