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

net/publication/268524453

Implementing Lean Manufacturing in the Sewing Industry

Conference Paper · May 2012

CITATIONS READS
5 4,810

3 authors:

Mohammed Said Obeidat Z.J. Pei

Jordan University of Science and Technology Texas A&M University
36 PUBLICATIONS   115 CITATIONS    309 PUBLICATIONS   7,827 CITATIONS   

SEE PROFILE SEE PROFILE

Raid Al-Aomar
Abu Dhabi University
95 PUBLICATIONS   1,136 CITATIONS   

SEE PROFILE

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

TSP heuristic application in pharmaceutical industry View project

Additive manufacturing of soft materials and hybrid structure View project

All content following this page was uploaded by Mohammed Said Obeidat on 29 November 2014.

The user has requested enhancement of the downloaded file.

Proceedings of the 2012 Industrial and Systems Engineering Research Conference
G. Lim and J.W. Herrmann, eds.

Implementing Lean Manufacturing in the Sewing Industry

Mohammed S. Obeidat, Z.J. Pei
Industrial and Manufacturing Systems Engineering Department
Kansas State University
Manhattan, KS 66502, U.S.A.

Raid Al-Aomar
College of Engineering & Computer Science
Abu Dhabi University
Abu Dhabi, U.A.E.

Abstract

Lean manufacturing has been adopted in a number of industries including electronics, automotive, and consumer
products manufacturing. Several industries are questioning the applicability of lean manufacturing principles. One
such industry is sewing industry. This research is focused on extending the scope of lean application by
implementing lean manufacturing in sewing industry. A sewing factory is selected for a case study. Characteristics
of sewing are first identified. Value Stream Mapping is developed to identify wastes. Five types of wastes are
identified and analyzed in one sewing production line: defects, inventory, overproduction, transportation, and
waiting time. Applied lean techniques include line balancing, layout redesign, and quality-at-the-source. The
impacts of these techniques on production are analyzed and an improved future state VSM is developed. Results
show a reduction of 96% in production wastes and 43% in lead time.

Keywords: Lean manufacturing, value stream mapping, sewing industry, waste reduction

1. Introduction
In a highly competitive global marketplace characterized by decreasing profit margins, waste reduction has become
an essential element in companies’ effort to thrive or in some cases to just survive. Muda is a Japanese word for
waste, “specifically any human activity that absorbs resources but creates no value” [1]. Wastes include defects,
inventory, overproduction, transportation, waiting time, overprocessing, and excessive motion [2]. The cost of such
wastes contributes to the total production cost, reduces profits, and decreases competitiveness.

One of the initiatives that companies have been trying to implement to maintain competitive in the global market is
lean manufacturing. The main focus of lean manufacturing is on cost reduction by eliminating or reducing wastes.
Lean manufacturing has been spanned in many sectors including electronics, automotive, and consumer products
manufacturing. Examples can be found in [3-5]. Sewing industry is one important section of textile industry. It
includes a wide range of operations such as cutting fabric, using different machines to assemble fabric pieces,
attaching different accessories such as buttons, checking, and packaging. Details on sewing can be found in [6, 7].

This paper reports implementation of lean in sewing industry. A case-based approach is used to identify the different
key wastes in sewing industry and to prescribe suitable lean techniques to reduce wastes, and result in better overall
financial and operational performance. A sewing plant in Jordan referred to as AB sewing throughout the paper is
used as a case study. The study is focused on one production line that produces one product. Value stream map
(VSM) is firstly used to map the current state of the line and to identify different types of wastes in sewing process.
Formulas are developed to quantify each type of waste and guide the improvement. Relevant lean techniques are
applied to reduce wastes. Future state VSM is then developed to illustrate the improvement achieved. Actual results
are used to verify lean benefits.
 Obeidat, Pei, and Al-Aomar

2. Case Study Description

One production line in AB sewing factory is selected, it’s the line that produces V436 product (V436 line). The
monthly demand for this product is 37500 pieces. Because time can be either value added (VA) or non-value added
(NVA), it is important to differentiate between three time phenomena in our study. The first one is cycle time: which
has the same definition of time standard: “the time required to produce a product at a work station with the
following three conditions: (1) a qualified, well-trained operator, (2) working at normal pace, and (3) doing a
specific task” [8]. The second one is lead time: the amount of time between ordering a product and receiving it, it
includes both the VA and the NVA times. The last one is the takt time: the available time for production divided by
the customer demand [8]. Each sewing operation in this line has a specific cycle time (OCT) that was initially
measured according to motion and time studies (MTS). Figure 1 shows V436 line details. To implement lean
techniques in sewing process, it is essential to define some important process measures and quantify cost of
processing. Sewing is a labor intensive process that highly laded on the skills of manual workers [9]. Line
productivity is related to the number of line workers (LW), and to product cycle time (PCT). To increase
productivity of workers, a line target (LT) is set to each production line in the sewing plant. Line supervisor is
responsible to meet the line target. LT represents the number of units required during an 8 hours shift per day.
Consequently, LT can be calculated using Equation 1.


  

 (1)


Figure 1: V436 line details (process, machine type, and cycle time)

Similarly, operation target (OT) can be calculated based on the number of workers in the operation (OW) and
operation cycle time (OCT). OT is calculated for a shift of 8 hours as expressed in Equation 2. Worker produced
quantity (WPQ) which was used to calculate workers efficiency (WE) is calculated based on Equation 3.


  
 (2)

 Obeidat, Pei, and Al-Aomar

!
    100% (3)


The company costing system expresses production cost in term of time spent by products in sewing processes. Each
hour spent on product processing or handling cost an amount (Ch) which is $3.54/ hour. Wastes are also converted
into time. The processing cost (PC) for a quantity (Q) for any product on a sewing line is determined by Equation 4.
!(
%&& '&
%' (4)


3. VSM: Current State Map

Value stream is the combination of all actions that are required to fulfill a customer order [10]. Value stream map
(VSM) includes main processes starting from receiving order(s), coordinating with the required supplier(s), passing
through processes of production and inspection, and ending with shipping to customer [9, 11]. The main goal of
VSM is to highlight the different sources of all types of wastes, and to develop solutions to eliminate or reduce them
[11]. In constructing a VSM, a particular product or product family with a high production volume must be selected
for process improvement [10]. The next step is to draw the current state VSM that represents the current
performance [10]. This can be achieved by following all processes related to the selected product aiming to find
different types of wastes. The final step is to develop and design the future state VSM, which represents the
improved picture of processes after implementing lean techniques [10]. Readers may refer to [11] for more details
about VSM.

All data for the current state VSM for V436 line were collected, starting from receiving and analyzing customer
orders, and ending with shipping. The overall collected data include ordering amounts and their frequency, materials
flow, information flow, inventory level, and shipping time and capacity. Specific data at each production operation
include cycle time, number of workers in each operation, daily available time, and inventory levels between
operations. Data is collected through direct observation, from production manager, line supervisor, and workers.

Figure 2 represents the current state VSM for V436 line. Production operations are mapped using a data box for
each operation. The number at the top of each data box represents operation number based on Figure 1. Inventory
triangles represent the level of materials between operations. Arrows in the map represent the flow of information
between the plant sections. The timeline at the bottom of the current state VSM has two components; one for delay
(in days) and the other is the operation cycle time (in minutes). Inventory is converted into NVA time by dividing
inventory level in front of each operation by the daily demand. For example, in Figure 2, the inventory level in front
of operation 2 is 827 pieces, and the daily demand from V436 product is 1500 pieces, so the NVA time is 0.551 day.
Another source of NVA time is the quality inspection time. In AB sewing there are two final quality inspections, one
for product quality (inspecting samples from product lots), and the other is the packaging quality (inspecting 100%
product packaging). The total NVA time for V436 line is 5.31 days, VA time is 5.48 minutes, and the total lead time
is 5.322 days.

3.1 Sewing Process Wastes in the Current VSM

Plant observation combined with a detailed analysis of the current state VSM had led to detecting five types of
wastes (defects, inventory, overproduction, transportation, and waiting time). They were observed over a period of
one month (30 days) of production to characterize their nature and quantify their impact using lean measures.
Formulas are developed to translate these wastes into cost values, and to guide the development of future state VSM.
Following is a description and assessment for them.

3.1.1 Defects
Defects can be defined as “frequent errors in paperwork or materials, product quality problems resulting in scrap or
rework, as well as poor delivery performance” [12]. Defects in AB sewing plant are measured using two daily
measures: the quality level percent (QL%) and the rejection percent (R%). Equations 5 and 6 describe how both
measures are obtained, respectively. QL% is used in the plant to give management an idea about quality level in the
company. According to Equation 5, for each product, the size of all samples being inspected during a day are added
together, the number of defects found in these samples is recorded and totaled, and at the end of the day QL% is
obtained by dividing the total number of defects by the total size of samples. In AB sewing plant, sample size is
determined by management according to the quantity lot size that needs to be checked. For example, if 5 samples for
the same product were inspected with a size of 150 pieces each, the total sample size will be 750 pieces, and if 15
 Obeidat, Pei, and Al-Aomar

I I I I I I I I

Figure 2: Current state VSM for V436 line

 Obeidat, Pei, and Al-Aomar

defective pieces from those five samples were found, the QL% will be 2% by dividing the total number of defects
(15) by the total size of samples (750). Based on Equation 6, for each product, the sizes of all lots being inspected in
a day are added together, the sizes of all lots being rejected (if any) is added together, and R% is obtained by
dividing the total rejected lots by the total lots being inspected in the same day. For example, if 4 lots from V436
product with 750 pieces each were inspected, (3000) in total, and if one of those lots was rejected by final quality
inspection, then R% will be the result of dividing the rejected lot size (750) by the total size of the 4 lots (3000) to be
25%. Defects cost (DC) is calculated based on Equation 7. These lots need to be repaired by operations 10 and 12
with a 0.85 minute cycle time in total. Table 1 shows the current defects waste and their cost.

∑0
123 ./
)*+
+,+ % )
% ∑0
 100% (5)
123 .

∑0
123 6!
45
 % 4% ∑0
 100% (6)
123 !

6!689
7
'&
7'  ': (7)


Where SD is sample defective pieces, S is the sample size, RQ is the size of the rejected lot, LQ is the lot size, and
RCT is the total operations cycle time involved in reworking.

Table 1: Quality measures and defects waste

Rejected quantity Quality level % Rejection % Defect cost ($)
3053 4.93 13.32 153.11

3.1.2 Inventory
Inventory is defined by “more materials, parts or products on hand than the customer needs” [5]. The actual
inventory level during the period of studying for V436 line is shown in the current state VSM in Figure 2. These
amounts are converted into NVA time by dividing each inventory quantity (IQ) by the customer daily demand (D),
and then converted into cost using Equation 8. Table 2 shows the NVA time cost due to inventory during the study
period.
<!
;,
 &
;'  8  ': (8)
/

Table 2: Inventory waste cost

Inventory quantity Daily demand NVA time (days) Inventory cost ($)
7404 1500 4.936 139.78

3.1.3 Overproduction
Overproduction is producing items that are no more demanded [1]. In quantifying overproduction waste during the
study period, the daily operations production (Q) was monitored and recorded. Overproduction quantities can be
calculated by subtracting the daily operation’s target (OT) that being calculated based on Equation 2 from the daily
operation’s production. A Positive result means there is an overproduction. Operations that had overproduction
during the study period were 4 and 9 with overproduction quantities of 10330 and 11310, respectively. The
overproduction cost is calculated using Equation 9. Table 3 shows the overproduction cost.

!?
,>*
 &
'  ': (9)


Table 3: Overproduction waste cost

Overproduction (hour) Cost ($)
183.21 648.72

3.1.4 Transportation
Transportation waste is defined by “excessive movement of people, information or goods, resulted in wasted time
and cost” [12]. Figure 3 shows the current layout of V436 line in the study period, where numbers inside boxes are
 Obeidat, Pei, and Al-Aomar

related to operations number in Figure 1. To quantify transportation waste, movement number in the line can be
assumed if we know worker number in the first operation (Ni-1) and in the next one (Ni) as in Equation 10.

 @,@
& *@A 
BC ∑F
DGE CD?E  CD (10)

Based on Equation 10, the total workers movements to produce one batch from V436 product was 37. This was also
calculated referring to Figure 1. To measure the waste associated with transportation, the predetermined time
standard system was used to determine the movement time. During the study period, we noticed that the walking
distance was around 4.5 meters per batch. We considered worker’s walking distance only and neglected other
components of movement such as reaching, and grasping. In the predetermined time standard system, walking 3.048
meters requires 0.04 minute [8]. As a result, 0.0591 minute is required for walking 4.5 meters. The cost of moving
one batch can be obtained using Equation 11. The batch size was 25 pieces. Equation 12 explains the process of
calculating transportation cost. AB sewing management assumes 50% of transportation in lines as waste, because
workers went to pick a batch with free hands and come back loaded. Table 4 shows the transportation cost.
JFK/LMN
H
I
&

 &
H'  ': (11)


'&

&

 ' H'  CH (12)

Where; AWD is the average walking distance by worker, TAWD is the time during the AWD in minutes, and NB is
the number of transported batches.

Figure 3: V436 line current layout

Table 4: Transportation waste cost

Production Number of batches Batch transportation cost ($) Transportation cost ($) Waste ($)
13350 534 0.581 310.25 155.13

3.1.5 Waiting Time

Waiting time waste is defined by “long period of inactivity for people, information or goods, resulting in a poor flow
and longer lead-times” [12]. During the study period in the sewing line, buffers were used to absorb fluctuation in
production stream throughout the line. To quantify this waste, worker produced quantity (WPQ) at each operation is
recorded and compared by the operation target (OT). Waiting time cost is computed for each operation using
Equation 13 where i refer to the operation number. Table 5 shows the cost of waiting time waste.

!?
W

@ &
' ∑EO
DGE  ': (13)


Table 5: Waiting time waste

Waiting time (hour) Cost ($)
2230 7897

4. Applying Lean Techniques

Lean techniques are focused on reducing different types of wastes. A set of relevant lean techniques that can suit to
the nature of sewing industry is implemented to develop the future state map of V436 line. Observing the actual line
behavior during the 30 days had showed a huge glare for the key issues of the sewing plant such as unbalanced line
operations, ineffective process layout, frequent defective units, flow delays, and lack of an effective monitoring
scheme. Process observation resulted in selecting three important lean techniques to treat plant issues. These lean
techniques were line balancing, layout redesign, and quality-at-the-source. Following is a brief summary on
implementing these techniques in sewing plant.
 Obeidat, Pei, and Al-Aomar

4.1 Line Balancing

Table 6 shows operations cycle times, operation target per day if only one worker used based on Equation 2, and the
required worker(s) to fulfill the daily demand (1500 pieces/ day). Number of workers is calculated based on
Equation 14 that was modified to calculate for each operation. The total required number of workers to produce the
daily demand based on the daily available time and product cycle time is 17.14 workers. Decimal means workers
can be shared in more than one operation. According to the required number of workers as shown in Table 6, no
bottleneck in V436 line will be found because of selecting only the required number of workers to produce the
required daily demand and this will result in reducing overproduction, Inventory, and waiting time wastes.
/

 P *@A 
C (14)


Table 6: Line balancing details

Operation 1 2 3 4 5 6 7 8 9 10 11 12
Cycle time (minute) 0.43 0.40 0.80 0.67 0.26 0.36 0.72 0.18 0.36 0.60 0.45 0.25
Operation target (1 worker) 1116 1200 600 716 1846 1333 666 2666 1333 800 1066 1920
Required worker(s) 1.34 1.25 2.50 2.10 0.81 1.13 2.25 0.56 1.13 1.88 1.41 0.78

4.2 Layout Redesign

Based on the line balancing results, the actual production flow shown in Figure 1, and our study period information;
a new line layout was developed as shown in Figure 4. Each operation was located in a specific place to make a
serial production flow and to enhance cellular manufacturing. Product flow, transportation and cross linking
between operations were considered and minimized in the implemented layout. In the current state, for each 3-4
sewing lines, a packaging line located in the side of the sewing hall for packaging these lines products. For reducing
transportation, the packaging line was divided into operations distributed into sewing lines in the hall. This enhanced
the implementation of the cellular manufacturing, and reduces workers movements to the minimum. Transportation
waste was also reduced by discarding large buffers between operations, and installing part bins instead to each
machine in the operation for storing materials and work in process (WIP). This allowed sewing workers, especially
in serial operations, to be loaded with work while they are setting in their operations.

Figure 4: V436 improved layout

4.3 Quality-at-the-Source
A new quality policy was implemented aiming to increase quality level by focusing on products quality in the
sewing line rather than relying on quality procedures after packaging. To this end, workers were trained to inspect
one piece out of five at their operations; this procedure was easy for them because each one is an expert in his or her
operation. Moreover, workers were convinced to call their supervisor or quality monitor (the one responsible on
selecting samples from all operations hourly) if there is any defective piece passed to them from any operation
before. In addition, line supervisor was asked not only to meet production level, but also to select samples from all
workers randomly from all operations. Furthermore, we installed a new final quality inspection inside the line; the
function of this is to select a sample from the finished products before sending them to packaging, which had a good
effect in reducing reworking time for rejected lots, and increasing the control on quality.

5. Lean VSM: Future State Maps

The process of designing future state VSM starts when developing the current state VSM, where critical areas that
required improvement start to shine [10]. While considering the current state map for V436 line, several issues were
standing out: large WIP, and the difference between lead time and VA time. Based on lean improvement discussed
earlier, we constructed the new future state VSM for V436 line as shown in Figure 5.
 Obeidat, Pei, and Al-Aomar

Figure 5: Line V436 future state VSM

 Obeidat, Pei, and Al-Aomar

5.1 Sewing Process Wastes in the Future VSM

After implementing lean techniques to the line in AB sewing, we selected another month (30 days) to measure
results. Results of lean implementing were quantified by reassessing the five identified wastes in the line based on
the future state VSM. Following is a summary of this assessment.

5.1.1 Defects
This waste was reduced by implementing quality-at-the-source approach, where quality became everyone’s
responsibility in the plant. During the improvement period, both quality level percent and rejection percent had been
reduced. Table 7 shows the quality measures and the cost of quality during the period of study after improvement;
reworking time includes operations 10 and 12 times (0.85 minute) as shown in Figure 1.

Table 7: Quality measures and defects waste

Rejected quantity Quality level % Rejection % Defect cost ($)
630 3.66 3.26 32.19

5.1.2 Inventory
The applied lean techniques contributed together to reducing inventory waste. Lead time reduction was the main
benefit. Implemented lean techniques assisted in the reduction of WIP as well. Table 8 shows the NVA time due to
inventory and its cost based on Equation 8.

Table 8: Inventory waste cost

Inventory quantity Daily demand NVA time (days) Inventory cost ($ per month)
3941 1500 2.627 74.41

5.1.3 Overproduction
Line balancing and layout redesign played a great role in reducing overproduction cost. The monthly overproduction
quantities were identified in operations 1, 3, 5, 6, 9, and 11 with quantities of 1004, 13, 470, 678, 685, and 137,
respectively. According to Equation 9, the new overproduction cost is shown in Table 9.

Table 9: Overproduction waste cost

Overproduction (hour) Cost ($)
18.61 65.88

5.1.4 Transportation
Transportation waste was reduced by implementing line balancing and layout redesign lean techniques. Line layout
was improved by sequencing sewing operations according to the result of line balancing and the flow of product
shown in Figure 1. In the improvement period, workers walked about 1 meter on average to pick a batch, and
according to the predetermined time standard system, 0.013 minute is required to walk this distance [8]. We used the
management information mentioned before about plant transportation, the average batch size that used in the plant
was 25, and the least number of movements remains 37. Table 10 shows the transportation waste cost based on
Equation 11.

Table 10: Transportation waste cost

Production Number of batches Batch transportation cost ($) Transportation cost ($) Waste ($)
45013 1800 0.018 32.4 16.2

5.1.5 Waiting Time

Line balancing and layout redesign resulted in reducing waiting time waste. Furthermore, inventory reduction
contributed to waiting time reduction. Waiting time cost is recalculated based on Equation 13 as shown in Table 11.

Table 11: Waiting time waste

Waiting time (hour) Cost ($)
29.56 104.64
 Obeidat, Pei, and Al-Aomar

6. Conclusions
This study shows a great evidence of the applicability of lean manufacturing implementation in sewing industry. In
this study, lean manufacturing implementation resulted in a 96% saving in the wastes cost comparing to before
implementing lean techniques, as shown in Table 12. Moreover, Lean manufacturing implementation in sewing
resulted in reducing process lead time, 43% reduction in process lead time was obtained after lean manufacturing
implementation, as shown in Table 13.

The most effective lean techniques in sewing are layout redesign, line balancing, and quality-at-the-source. These
techniques might work together to reduce five types of wastes in sewing: defects, inventory, overproduction,
transportation, and waiting time.

Table 12: Waste comparison before and after lean implementation

Waste Cost before ($) Cost after ($) Saving (before-after) ($) Saving %
Defect 153.11 32.19 120.92 79
Inventory 139.78 74.41 65.37 47
Overproduction 648.72 65.88 582.84 90
Transportation 155.13 16.20 138.93 90
Waiting time 7897.00 104.64 7792.36 99
Total 8993.74 293.32 8700.42 96

Table 13: Lead time comparison before and after lean implementation
Lead time (days) before Lead time (days) after Reduction %
5.322 3.04 43

References
1. Womack, J., and Jones, D., 2003, Lean Thinking: Banish Waste and Create Wealth in Your Corporation,
Free Press, New York.
2. Svensson, G., 2001, “Just-In-Time: The Reincarnation of Past Theory and Practice,” Management
Decision, 39(10), pp. 866-879.
3. Rajenthirakumar, D., and Thyla, P., 2011, “Transformation to Lean Manufacturing by an Automative
Component Manufacturing Company,” International Journal of Lean Thinking (in press).
4. Holweg, M., 2007, “The Genealogy of Lean Production,” Journal of Operations Management, 25, 420-437.
5. Taj, S., and Berro, L., 2006, “Application of Constrained Management and Lean Manufacturing in
Developing Best Practices for Productivity Improvement in an Auto-Assembly Plant,” International Journal
of Productivity and Performance Management, 55(3/4), 332-345.
6. Shaeffer, C., 2000, Sewing for the Apparel Industry, Prentice Hall, New Jersy.
7. Rosenau, J., and Wislson, D., 2001, Apparel Merchandising: The Line Starts Here, FairChild Publications,
Inc., New York.
8. Meyers, F., and Stewart, J., 2002, Motion and Time Study for Lean Manufacturing, 3rd Edition, Prentice
Hall, New Jersey.
9. Obeidat, M., and Al-Aomar, R., 2011, “Adopting Lean Techniques in Textile Industry: A Sewing Plant
Case Study,” ALHOSN University Journal of Engineering and Applied Sciences, 4(1), 33-53.
10. Abdulmalek, F., and Rajgopal, J., 2007, “Analyzing the Benifits of Lean Manufacturing and Value Stream
Mapping Via Simulation: A Process Sector Case Study,” International Journal of Production Economics,
107, 223-236.
11. Rother, M. and Shook, J., 1999, Learning to See: Value Stream Mapping to Add Value and Eliminate
MUDA, The Lean Enterprise Institute, Inc., Brookline, MA.
12. Sullivan, W., McDonald, T., and Van Aken, E., 2002, “Equipment Replacement Decisions and Lean
Manufacturing,” Robotics and Computer-Integrated Manufacturing, 18(3-4), 255-265.

View publication stats