Simulation analysis and optimization of garment manufacturing process based on Em-Plant software

Abstract

In recent years, the way of competition has changed in terms of garment manufacturing. Quality and scale competition gradually turn into speed competition. Garment manufacturers need to shorten the production cycle to win development opportunities. This paper takes a garment manufacturing factory as the research object, and innovatively uses the MOST method and Em-Plant simulation to improve the efficiency of its production line. According to each activity, operation time and worker assignment of the garment manufacturing process, the corresponding simulation model based on Em-Plant software is created to reveal the current problems existing in the manufacturing process. By analyzing these results, the bottlenecks in the process are found out. Applying theoretical knowledge such as production line balancing, lean production, motion study and time analysis, the improvement two plans are put forward, and the corresponding simulation results are given subsequently. Finally, a relatively better improvement plan is determined, which can better achieve the goal of balancing production line, improving efficiency, and reducing cost.

Keywords

Production line balancing system simulation MOST (Maynard Operation Sequence Technique)

1. Introduction

Clothing is a necessity in people’s life, and the manufacturing process is changing with times. Garment manufacturers must continuously improve their own production efficiency in order to gain competitive advantage [1].

System simulation technology has obvious effects in improving production efficiency. Through simulation, bottlenecks in the production line can be found, the effect of the optimization plan can be presented. Motion study and time analysis methods in the field of Industrial Engineering can also effectively reduce work time and improve efficiency by streamlining motions and setting standard work time.

This paper uses a garment manufacturer in Wenzhou as an example to illustrate the process of using the above methods to improve the efficiency of its production line.

2. Background introduction of a garment manufacturing factory

This garment manufacturing factory mainly produces man’s clothes and is located in Wenzhou City, Zhejiang Province. Its building area is 40 $\times$ 10 m ${}^{2}$ . It has two floors and owns 23 manufacturing workers. The garment manufacturing factory has some garment processing equipment such as cutting machines, sewing machines, buttoning machines, and ironing machines, etc. This paper takes one of the production lines as the research object, analyzes the imbalance of workers’ efficiency, and provides two optimization plans.

The overall garment manufacturing process is as follows:

(1)
Cloth inspecting: inspect the cloth that has just entered the factory – if it is not qualified, marked as waste;
(2)
Cutting: transport the cloth to the cutting area – cut the cloth into pieces – store the pieces;
(3)
Sewing: sew collar, cuffs, pockets, main parts – label – splice – button – trim;
(4)
Ironing: iron the finished clothes and wait for inspection and package;
(5)
Inspecting: inspect the finished clothes’ quality;
(6)
Returning: return the product that failed in the quality inspection to the sewing process and re-sewn;
(7)
Packaging: pack qualified clothes – storage.

3. Current simulation analysis of the garment manufacturing process

The work content, work time, worker assignment and the corresponding Em-Plant simulation object parameters are shown in Tables 1 and 2.

Table 1
Basic information in the garment manufacturing process

Work content	Time (minute)	Worker (person)	Object name
Semi-finished clothes			Cloth
Cloth entry		1	Entry
Temporarily stored clothes			Buffer
Inspection and diversion	1	1	Inspection1
Waste disposal			Wasted_cloth
Waste cloth elimination			Elimination
Cut cloth	2	2	Cutting
Sew cloth	11.5	13	Sewing
Iron clothes	2	2	Ironing
Quality inspection	1	1	Inspection2
Qualified clothes			Quality_OK
Return	1	1	Quality_not_OK
Pack finished clothes	1	1	Packing
Temporary storage of re-sewn clothes			Buffer1
Storage of finished clothes		1	Storage
Work efficiency display			Chart
Total	19.5	23

Table 2

Objects in the sewing process

Work content	Time (minute)	Worker (person)	Object name
Sew collar	0.8	2	Collar
Label	0.5	2	Label
Sew cuffs	2	2	Cuffs
Sew pockets	1	2	Pockets
Sew main parts	2.5	2	Main parts
Splice	3	1	Splicing
Nail buttons	1.2	1	Buttons
Trim	0.5	1	Trimming
Total	11.5	13

According to the actual process and some other information, the Em-Plant simulation model of current garment production line is established as Fig. 1.

Figure 1.

Current garment production line simulation model.

Run the initial model for 8 hours. It can be seen from the result that sewing process takes the longest time in the entire garment manufacturing process, so there is a problem of unbalanced working time and strength in different stations, which leads to large waste [2].

Figure 2.

Current sewing process simulation model.

Observe the final output of the sewing process simulation model (Fig. 2). Em-Plant software can automatically count the operating state (Working, Waiting, Blocked and so on) of each station. It is clear that the splicing station has the largest amount of work, followed by sewing main parts, sewing cuffs, nailing buttons and other stations (shown in Fig. 3) [3].

Figure 3.

Current work intensity of each station in the sewing process.

The final output of clothes is 184 pieces (Fig. 4).

Figure 4.

Current output of clothes.

4. Improvement of the garment manufacturing process

Based on the MOST method and Em-Plant simulation, optimized plans can be put forward.

4.1 Introduction of the MOST method

Motion study is a science method of eliminating wastefulness resulting from inefficient motions [4]. The main aim of motion study is to find the scheme of least effective motion and time. The MOST (Maynard Operation Sequence Technique) method is a very practical and economical method of motion analysis [5]. It is easy to operate, and can be reasonably used in many fields such as assembly, sewing, packaging, banking service, etc.

MOST describes all manual work in three sequences:

(1)
General Move Sequence (GM): The object follows an unrestricted path under manual control. GM describes get $\backslash$ place $\backslash$ return, and the sequence is:

A $=$ Action Distance

B $=$ Body Motion

G $=$ Gain Control

P $=$ Place Motion
(2)
Controlled Move Sequence (CM): Manual displace an object over a “controlled” path. CM describes get $\backslash$ move or actuate $\backslash$ return, and the sequence is:

M $=$ Move Controlled

X $=$ Process Time

I $=$ Align
(3)
Tool Use Sequence (TU): Describes the activities performed with tools. TU describes get tool $\backslash$ locate tool $\backslash$ use tool $\backslash$ aside tool $\backslash$ return, and the sequence is:

U $=$ Tool Use Parameter (Fastening, Record, Loosening, Measuring, Cut, Read/Inspect, Surface Treat, etc.)

Different activity has different index value. Table 3 is the index value of general move.

Table 3
The index value of general move

General move

Index A B G P

0 $<$ 2 inches $<$ 5 cm Hold, toss, strike

1 Within reach Light object grip Lay aside, place, loose fit

3 1–2 steps Bend and arise 50% Heavy or bulky, blind or obstructed, interlocked, collect Adjustments, light pressure, joggle

6 3–4 steps Bend and arise Care or precision, heavy pressure, blind or obstructed, search, select

10 5–7 steps Sit or stand

16 8–10 steps Through door, climb on or off

TMU is used as a time measurement unit for MOST analysis.

1 TMU $=$ 0.036 SECOND.

The example of standard time calculation is as follows:

Sequence: A ${}_{6}$ B ${}_{6}$ G ${}_{1}$ A ${}_{6}$ B ${}_{0}$ P ${}_{1}$ A ${}_{0}$ Sum of indices $=$ 6 $+$ 6 $+$ 1 $+$ 6 $+$ 0 $+$ 1 $+$ 0 $=$ 20 TMU $=$ 20 $\times$ 10 $=$ 200 Seconds $=$ 200 $\times$ 0.036 $=$ 7.2 SECOND

4.2 Application of MOST

General move
0	$<$ 2 inches $<$ 5 cm			Hold, toss, strike
1	Within reach		Light object grip	Lay aside, place, loose fit
3	1–2 steps	Bend and arise 50%	Heavy or bulky, blind or obstructed, interlocked, collect	Adjustments, light pressure, joggle
6	3–4 steps	Bend and arise		Care or precision, heavy pressure, blind or obstructed, search, select
10	5–7 steps	Sit or stand
16	8–10 steps	Through door, climb on or off

Considering the low output efficiency and the unnecessary waste of costs resulting from the invalid motions of workers and nonstandard operating time of each activity, it is decided to use MOST to analyze each motion, and to specify standard motions and standard operating time [6].

Taking the production of a man’s shirt as a research object. In the sewing process, the cut cloth is placed on the workbench according to different sewing parts, then the collar, the cuffs, the pocket and the main parts are sewn by sewing machines. After labeling, the finished parts are transported to splicing, which is to sew the already finished collar and cuffs on the main parts of the shirt. The buttoning activity is to align the button to the buttonhole, and then use the machine to sew it on the clothes. The trimming activity is the last step, which is, to pick up scissors and subtract the excess thread.

For the sewing collar activity, it can be divided into four motions. First, put the cut cloth on the workbench. Because the range of this motion is small, according to the MOST method, the standard time calculated is 40 TMU; Second, align cloth to the needle of the sewing machine, and sew the collar. The standard time required is 390 TMU; After sewing the collar, pick up the scissors and cut off the excess thread, and then put the scissors back. The time need for this motion is 100 TMU; The last motion is to put the sewn collar into the collection basket on the side of the workbench. Its standard time is 30 TMU. The total time required is 560 TMU, which is 20.16 seconds. Each standard motion and time in sewing collar activity based on MOST are shown in Table 4.

Table 4
Standard motion and time analysis of sewing collar

Activity	Sewing collar
Seq No.	Sequence description	SQM	Sequence model											RF	Total TMU
1	Put the cut cloth on the	GM	A	B	G	A	B	P	A						40
	workbench		1	0	1	1	0	1	0					1
2	Align to the needle, and	TU	A	B	G	A	B	P	U	A	B	P	A		390
	sew the collar	U	0	0	0	1	0	1	35	1	0	1	0	1
3	Pick up the scissors, cut	TU	A	B	G	A	B	P	U	A	B	P	A		100
	off the excess thread, and then put the scissors back	U	1	0	1	1	0	1	3	1	0	1	1	1
4	Put the sewn collar into	GM	A	B	G	A	B	P	A						30
	basket		0	0	0	1	0	1	1					1
Seconds	20.160														560
Minutes	0.336														179

The standard time of other activities can be calculated in sequence as above, which is shown in Table 5.

Table 5

The standard time of each activity in the sewing process

Activity	MOST time (second)	Time ratio (%)
1. Sewing collar	20.160	5.77%
2. Labeling	7.560	2.16%
3. Sewing cuffs	62.280	17.84%
4. Sewing pockets	18.360	5.26%
5. Sewing main parts and splicing pockets	61.560	17.63%
6. Splicing other parts	85.680	24.54%
7. Nailling buttons and cutting thread	93.600	26.80%
Total	349.200	100.00%

4.3 Optimized plan NO.1

According to the ratio of the MOST motion time of each sewing activity, the number of workers needed can be calculated. The specific worker assignment is shown in Table 6.

Table 6
The worker assignment of sewing process (Optimized plan NO.1)

Activity	MOST time (second)	Worker assignment (person)
1. Sewing collar	5.77%	1
2. Labeling	2.16%
3. Sewing cuffs	17.84%	2
4. Sewing pockets	5.26%	1
5. Sewing main parts and splicing pockets	17.63%	2
6. Splicing other parts	24.54%	3
7. Nailling buttons and cutting thread	26.80%	3
Total	100.00%	12

According to the actual situation, the worker assignment can be adjusted, sewing collar and labeling can be combined together in order to reduce the imbalance of the production line. That means one worker can finish the two activities. The total number of workers in the sewing process can now be adjusted to 12 persons.

This optimized model is shown in Fig. 5.

Figure 5.

Optimized simulation model NO.1 of sewing process.

The simulation results are shown in Fig. 6.

Figure 6.

Work intensity of each station in optimized plan NO. 1.

It can be seen from Fig. 6 that the working efficiency of each station in the sewing process is more balanced. The final number of shirts in optimized plan NO. 1 adjusted by the MOST method is 222 pieces. Compared with the initial simulation model, the result has increased by 20.7%. The improvement effect is obvious.

The total number of workers in this plan is 12 persons. One person has been reduced and two sewing machines has been reduced compared to the initial model.

Facts show that using the standard operation time and reasonable worker assignment, optimized plan NO. 1 can improve the work efficiency, reduce unnecessary time waste and simplify the number of workers [7].

4.4 Optimized plan NO. 2

This plan focuses on ways to rearrange or combine activities, redistribute the number of workers to reduce production costs, increase efficiency and achieve more economic income [8].

According to the standard time of each activity calculated by the MOST method, the activity with short time can be combined to improve the working efficiency of the production line.

In this plan, the sewing collar, labeling and sewing pockets are combined into one operation, the workers of sewing cuffs and sewing main parts are reduced from two to one separately, and the workers of splicing and nailing buttons are reduced from three to two separately. Reduce the time and cost of storage and delivery at the same time. The total number of workers in the sewing process is reduced from 12 persons to 8 persons.

The specific model is shown in Fig. 7.

Figure 7.

Optimized simulation model NO. 2 of sewing process.

The results of the changed plan are shown in Fig. 8. The final output is 241 pieces, which is increased by 30.9% on the basis of 184 pieces. This improvement basically meets the demand and the effect is more obvious.

Figure 8.

Work intensity of each station in optimized plan NO. 2.

Both of the above plans have achieved the more balance of the production line, but the efficiency of the second plan is higher, and the production cost is lower. Therefore, the second plan is considered as the final improvement plan of the sewing process.

4.5 Improvement of the overall manufacturing process

Figure 8 shows that the overall working efficiency of sewing process is about 50%. The reason is that in the overall manufacturing process, the cutting process and the ironing process take a lot of time, the workload and the number of workers is seriously mismatched. So, the overall performance of the production line is at a lower level [9]. Therefore, it is recommended to adjust the worker assignment of the overall garment production line on the basis of the second optimized plan, so as to balance the time of each process as much as possible. Adding two more workers to the cutting process and ironing process separately would be a good and reasonable idea, so that the operation time of each process is close to balance. The total number of workers is 22, which is one less than the initial model, and five sewing machines are reduced.

The adjusted overall production line model is shown in Fig. 9.

Figure 9.

Adjusted overall production line model.

The results are shown in Figs 10 and 11.

Figure 10.

Work intensity of each station in the final production line model.

Figure 11.

Final output of clothes.

It can be seen from the chart that the work efficiency has been significantly improved. The current average level is maintained at around 74%. The adjusted final output reaches 274 pieces, with the growth rate of 48.9%. It is a satisfactory result.

4.6 Optimization effect evaluation

Take Cycle Time, Line Balance Rate, Smoothness Index of Production Line as the evaluation basis to quantitatively analyze the effect before and after optimization.

The calculation results of the initial situation are as follows:

1) Cycle Time

$\displaystyle\text{CT}=\frac{{T}_{w}}{Q}=\frac{8\times 60\times 60}{184}=156.5% ∼{}\text{s}$ (1)

CT – cycle time, $T_{w}$ – total work time, $Q$ – total output.

2) Line Balance Rate

In general, the higher the Line Balance Rate, the better the production line balance.

$\displaystyle{P}=\frac{\sum{T}_{i}}{T_{\text{max}}\times{n}}=\frac{690}{8% \times 180}=47.9\%$ (2)

$P$ – line balance rate, $T_{i}$ – time of activity $i$ , $T_{max}$ – the longest operation time in the production line, $n$ – number of stations.

3) Smoothness Index of Production Line

The smaller the Smoothness Index, the smaller the time dispersion deviation of the activities, and the better the time balance of the production line.

$\displaystyle\text{SI}=\sqrt{\frac{\mathop{\sum}\nolimits_{i=1}^{n}\left({% \text{CT}-T_{i}}\right)^{2}}{n}}=\sqrt{7769.5}\approx 88.14$ (3)

SI – smoothness index of production line, For comparison, the calculation results of the improvement plan are as follows:

1) Cycle Time

$\displaystyle\text{CT}=\frac{{T}_{w}}{Q}=\frac{8\times 60\times 60}{274}=105.1% ∼{}ss$ (4)

2) Line Balance Rate

$\displaystyle P=\frac{\sum T_{i}}{T_{\text{max}}\times n}=\frac{349.2}{7\times 9% 3.6}=53.3\%$ (5)

3) Smoothness Index of Production Line

$\displaystyle\text{SI}=\sqrt{\frac{\mathop{\sum}\nolimits_{i=1}^{n}\left({% \text{CT}-T_{i}}\right)^{2}}{n}}=\sqrt{4070.2}\approx 63.8$ (6)

In summary, it can be seen that the balance optimization of the production line has achieved good results.

5. Conclusions

This paper takes the manufacturing process of a garment manufacturing factory as the study object, uses Em-Plant simulation software to simulate the real production line situation, analyzes the result data, and determines the bottleneck process in the production line, especially in the sewing process. The MOST method is used to analyze the motion and standard time of each activity in the sewing process, and the number of workers in each station is redistributed. Two improvement plans have been put forward. The results show that the effect of the second plan is more obvious, which better achieves the goal of balancing production line, improving efficiency, and reducing cost.

The research ideas and methods of this paper can provide valuable reference for related research or practical application. They are also suitable for solving other problems of the same type and can be applied to the analysis and improvement of different scenarios in different industries, such as logistics optimization, production line analysis, operation process optimization or environmental improvement.

Footnotes

Acknowledgments

The study is supported by Research plan of visiting scholars by Beijing Union University in 2017. The authors wish to thank teachers in Beijing Union University for their helpful comments and suggestions.

References

Mrugalska

and Wyrwicka

M.K.

, Towards lean production in industry 4.0, Procedia Engineering 182 (2017), 466–473.

Fisel

Exner

Stricker

and Lanza

, Changeability and flexibility of assembly line balancing as a multi-objective optimization problem, Journal of Manufacturing Systems 53 (2019), 150–158.

Goyal

Garg

and Mishra

V.N.

, New composition of graphs and their Wiener Indices, Applied Mathematics and Nonlinear Sciences 4(1) (2019), 163–168.

Tuan

S.T.

Karim

A.N.M.

Kays

H.M.E.

and Amin

A.N.

, Improvement of workflow and productivity through application of Maynard Operation Sequence Technique (MOST), Proceedings of the 2014 International Conference on Industrial Engineering and Operations Managemen, Bali, Indonesia, 2014.

Ashokbhai

C.K.

Puranik

P.S.

and Shah

J.B.

, Maynard Operation and Sequence Technique (MOST) intervention for productivity improvement of manufacturing process of mini tractor, Trends in Mechanical Engineering & Technology 8 (2019), 76–81.

Tatarnikova

L.I.

, Deployment of Kaizen in the framework of the implementation of lean production in passenger transportation company, Proceedings of the Internation Conference on “Humanities and Social Sciences: Novations, Problems, Prospects” (HSSNPP 2019), Novosibirsk, Russia, 2019.

Shen

and Wei

, Does project feature affect perceived attribute and crowdfunding intention in different project types? An empirical study from China, Advances in Industrial Engineering and Management 9(2) (2020), 11–19.

Das

Nag

Pyne

and Roy

D.S.S.

, Low cost energy efficient remotely controlled smart home security system, Advances in Industrial Engineering and Management 6(1) (2017), 53–56.

Eliseo

V.P.

, Interactive dynamic-system simulation, The Journal of the Operational Research Society 65(1) (2017), 158–159.

General move
Index	A	B	G	P
0	$<$ 2 inches $<$ 5 cm			Hold, toss, strike
1	Within reach		Light object grip	Lay aside, place, loose fit
3	1–2 steps	Bend and arise 50%	Heavy or bulky, blind or obstructed, interlocked, collect	Adjustments, light pressure, joggle
6	3–4 steps	Bend and arise		Care or precision, heavy pressure, blind or obstructed, search, select
10	5–7 steps	Sit or stand
16	8–10 steps	Through door, climb on or off