Huaqing Cao1,*,

Xiaofen Ji2
1 Zhejiang Sci-Tech University,
College of Textile Science and Engineering,
Hangzhou, 310000, P.R. China,
* e-mail: owencho@163.com
2 Zhejiang Sci-Tech University,
School of International Education,
Hangzhou, 310000, P.R. China
GENERAL PROBLEMS OF THE FIBRE AND TEXTILE INDUSTRIES
Prediction of Garment Production Cycle
Time Based on a Neural Network
DOI: 10.5604/01.3001.0014.5036
Abstract
The process of garment production has always been a black box. The production time of
different clothing is different and has great changes, thus managers cannot make a produc-
tion plan accurately. With the world entering the era of industry 4.0 and the accumulation
of big data, machine learning can provide services for the garment manufacturing industry.
The production cycle time is the key to control the production process. In order to predict
the production cycle time more accurately and master the production process in the garment
manufacturing process, a neural network model of production cycle time prediction is esta-
blished in this paper. Using a trained neural network to predict the production cycle time,
the overall error of 6 groups is within 5%, and that of 3 groups is between 5% and 10%.
Therefore, this neural network can be used to predict the future production cycle time and
predict the overall production time of clothing.

Key words: garment production, big data, cycle time, neural network, prediction.

Introduction and digital production has become a new

The textile and garment industries are
two of the most important in China. Ac-
cording to statistics of the World Trade
Organization, in 2017 global textile and
garment exports were worth 647.89 bil-
lion euros, while China’s exports were
245 billion euros, ranking first in the
international textile and garment trade
pattern, with a market share of more
than 1/3 [1]. In 2018, China’s total im-
port and export value for textiles and
clothing reached 252.79 billion euros, up
3.7% year on year [2]. The contribution
rate of exports was the largest among all
the factors of garment demand, which
means that exports were the most signif-
icant factor in driving the development
of China’s textile industry [3]. The EU is
China’s largest export market. The value
of textile and garment products imported
by the EU is nearly 114.8 billion euros,
mainly from China, Bangladesh, Turkey,
India and Cambodia [4]. With the devel-
opment of production technology, such
lean production methods as 5S manage-
ment, VSM (Value Stream Mapping) and
others are applied to garment produc-
tion, China is also facing competition
from low-cost labor forces in Southeast
Asia and other countries. Thanks to the
close relationship of the textile industry
with agriculture and ancient culture, the
Indian textile industry has the ability to
produce a wide range of products suita-
ble for various market segments, both in
India and around the world [5]. After im-
plementing lean tools in Indian garment
enterprises, researchers observed a re-
duction in work-in-progress inventory,
as well as increased production and ef-
ficiency of the production line [6]. Lean
development direction to improve the ef-
ficiency of garment production.
However, the sewing process is still
a black box. Influenced by factors such
as proficiency, style, fabric, etc., the cy-
cle time of each garment production is
different, which decreases with an in-
crease in the number of work pieces.
The process of garment sewing produc-
tion is a declining learning curve, which
in theory is a smooth curve, as shown in
Figure 1. In fact, workers need time to
adapt to the cycle time after the decline.
The learning curve in actual production
is shown in Figure 2.
The abscissa of the image is the number
of production pieces, and the ordinate is
the cycle time. Wright was the first person
to observe and study the learning curve in
production and operation management
[7]. Through the learning curve, managers
can set more accurate labour standards,
monitor actual production targets [8],
as well as predict the available working
hours of the process [9] and production
volumes [10]. In production and opera-
tion management, the learning curve can
describe employee performance improve-
ment caused by repetition or experience,
which makes it a useful tool for manage-
ment decision [11]. Production big data
combined with regression analysis can
help managers choose the best learning
curve model [12]. Through the proficien-
cy rate, a functional relationship between
different production days and the pipeline
cycle time can be established [13]. How-
ever, there is no study on the cycle time
of garment production using big data and
a neural network.

8 Cao H, Ji X. Prediction of Garment Production Cycle Time Based on a Neural Network.

FIBRES & TEXTILES in Eastern Europe 2021; 29, 1(145): 8-12. DOI: 10.5604/01.3001.0014.5036
 Cycle time, s
Cycle time, s

Production quantity Production quantity

Figure 1. Theory cycle time. Figure 2. Actual cycle time.

Research methodology GST per capita: General sewing time the production cycle time is higher than
(GST) is a pre-set action time system spe- the GST per capita due to the low pro-
An artificial neural network (ANN) cially designed for the clothing industry, ficiency of workers. The initial cycle
is a mathematical model based on the which originated from the production time is the highest point on the produc-
working principle of a biological neural therblig analysis proposed by F.B. Gil- tion learning curve, which determines
network. A neural network is one of the breth. The GST programs the common the starting point of the learning curve.
most basic forms of machine learning. In actions in the sewing process and describe The initial cycle time needs to be de-
the textile and garment industry, a neu- the actions in code form. Each code has termined according to the bottleneck
ral network has been used to deal with a clear time value, and then a standard process on site, which cannot be calcu-
complex non-linear sewing process
problems, suchandas describe the actions in code form. Each code has a clear time value,
time is estimated according to the moving lated before production, and can reflect
prediction of fabric size
andand performance
then a standard time is estimated
distance and actionaccording
difficulty. to the moving distance
In a garment and action
the production difficulty of a garment to
[14], classification of fabric defects [15], factory, industrial engineers account for a certain extent.
difficulty. In a garment factory, industrial engineers account for the GST of each
human body recognition [16], evalua- the GST of each garment. The calculation
tion of the sewn thread consumption
garment. of formula
The calculation formula of GST
of GST perper capita
capita is: is: Building neural network
jean trousers and so on [17]. In this pa-
The structure of a BP neural network for
per, a BP (back propagation) neural net-
clothing production cycle time prediction
work is used to predict the learning curve
of the garment production cycle time. (1) is shown in Figure 3. There are three lay-
which reflects the production beat of the water line in an ideal ers: state.theHowever, in hidden layer and out-
input layer,
The curve is affected by many factors,
which reflects the production beat of the put layer. The input layer is responsible
but the logic relationactual production,
is difficult to ex-the GST per capita can only be used as a reference value. The final
water line in an ideal state. However, in for receiving data, with three variables
press with an accurate function relation.
stable cycle time may actual be larger than thethe
production, GSTGSTperpercapita
capitaduecan to factors
x1, x2 andsuch
x3 as the
representing the production
A BP neural network is a kind of forward
only be used as a reference value. The fi- quantity, GST per capita and initial cycle
tutor learning neural bottleneck
network withprocess
erroror complex style. It may also be smaller than the GST per capita due
nal stable cycle time may be larger than time.
reverse propagation. It does not need to
to factors such as workers the being
GST per more familiar
capita due to with the style
factors such and
as higher efficiency.
determine the mapping relationship be-
the bottleneck process or complex style. The hidden layer is responsible for data
tween the input and output in advance,
It may also be smaller than the GST per decomposition and processing. There are
but only learns someInitial
rules cycle
through its when the first garment of a production order is produced, the
time: capita due to factors such as workers be- 10 neurons, each of which is a processing
own training. Its basic idea is the gradi-
production cycle time is inghigher
more than
familiar
the with
GSTthe perstyle anddue
capita high-
to the unit
low receiving input
proficiency of from the upper layer.
ent descent method, which uses gradient
er efficiency. The activation function, also called the
search technology to workers.
minimiseThe the initial
error cycle time is the highest point on the production learning curve,
mapping function tansig, can transform
mean square deviation of the actual out-
which determines Initial cycle time: when the first gar-
the starting point of the learning curve. The initial the cycleresult of calculation
time needs to in a non-line-
put value and the expected output value
ment of a production order is produced, ar way, so as to improve the expression
of the network. be determined according to the bottleneck process on site, which cannot be calculated
Training parametersbefore production, and can reflect the production difficulty of a garment to a certain
Input layer Hidden layer Output layer
Production quantity: extent.
the larger the pro-
duction quantity is, the more skilled
the workers can be. The stable cycle
time will be greatly Building
reduced compared
neural network
with the initial cycle time, but there is
The structure of a BP neural network for clothing production cycle time prediction is
also a limit. The smaller the production
shown
quantity is, the shorter in Figuretime
the learning 3. There are three layers: the input layer, hidden layer and output layer.
of the workers is. It is possible that the
The input layer is responsible for receiving data, with three variables 、 ,
production has not reached the ultimate 10 160
representing
efficiency and the production hasthe production quantity, GST per capita and initial cycle time.
been
completed. Figure 3. Structure of cycle time prediction BP neural network.

FIBRES & TEXTILES in Eastern Europe 2021, Vol. 29, 1(145) 9

 Figure 4. Fitting of
is extended to 1600 pieces, with the last
165 predicted cycle time.
piece completed, and the interval being
160 10 pieces. For example, the production
of one training group has been completed
155
in 300 pieces, and the cycle time of the
150 last one is 200 seconds; thus the cycle
Cycle time, s

145 time of 300 pieces to 1600 pieces is set

to 200 seconds. It should be noted that
140
For example, For theexample,
For example, production
thedue thetoofproduction
production onesmall
the of training
oneof one
production group
training training
hasgroup
quantity beengroup has completed
of has been
been compleinco3
135 this section, 300 pieces will complete the
and the and
cycleand
thetimetheof
cycle cycle
the last
time time
of the one
of lastthe
is 200 last seconds;
one one
is 200 is 200 thus
seconds; seconds;
thethus cycle thus
thetime theof
cycle cycle30
tim
130 whole order. If the production quantity
ure 3 Structure of cycle time prediction BP neural network 1600 pieces
16001600ispieces
setpieces
tois200setisseconds.
toset200
increases, to 200 It should
seconds.
the seconds.
cycleIttime be Itnoted
should should
may bebe that bedue
noted
further noted to the
that that
duesmall
due
to the t
125
er is responsible for 0data decomposition and250processing. reduced. In the neural network, because
  50 100 150 200 300 350There 400 are 10 quantityquantity
of quantity
this section,
of thisof this300section,
section, pieces 300 will 300
pieces pieces
complete
will will the
complete complete
whole the order.the whole
whole Iforder.
theo
the production quantity is also an input
f which is a processing unit receiving input from the upper layer. The
Production quantity quantityquantity
increases,
quantity the
increases,
increases, cycle thetime
variable, thethecycle
cycle may time be
influence time
may furthermay reduced.
be further
be production
of further Inreduced.
reduced. the neura
In thI
quantity on the cycle time can be taken
ion , also called the mapping function, can transform the result because of thebecause
because production the quantity
the productionproduction is also
quantity quantity an
is also
into account in the training of the neural
input
is also anvariable,
an input
input thevariable,
variable, influence theofin
the influe
non-linear ability
way, soof asthetoneural
improve network and deal ability
the expression experiment.
of the120 groups
neural ofquantity
quantity production
on quantity
the on cycle ontime
the the cycle
network.
cycle can time betime taken
can can
beinto be account
taken takeninto into in account
account the training
in the in train
the
of
with the problem of linear indivisibility. process data of the enterprise from May
eal with thewijproblem of linear
is the hidden layer indivisibility.
weight and bij is is the hidden
to July 2019 were network.
layerobtained. 89network.
network. groups 89 sets of data are used to train the neural
the hidden layer threshold. The tansig were selected for neural network
89 sets 89 89train-
of sets
data sets
ofaredata network
ofuseddata
aretoused
are several
train used
tothe times.
to
train neural theThe decision
train network
the neural
neural several
network co-
network times.
severalsevera Th
tim
is the hidden layer isthreshold.
formula as follows:The formula is ingasand
follows:
10 for a simulation test. The data efficient R2 is stable at 0.95483. The cal-
experiment was carried coefficient
out coefficient
in MATLAB is stableisculation
coefficient atstable
0.95483.
is stable atformulaThe
at 0.95483.
0.95483. calculation
R2 The of is The formula
calculation
calculation
as follows: of formula
formula isof as offois
2
tan sig(n) = -
–     (2) 2017b. Through the training of the neu-
�1 æ l æ ll æ l lö
2
l
2
öl ö
2
1+ e�2n ral network, 10 groups of prediction data ç l
è i=1
å ŷ y
i iç
–
� l å
� ŷl å
�
y � ŷ
y �
ç ii i i ÷ i i y ŷ
� � ŷ
y i÷� y i÷
The output layer receives the output of were obtained, and curve fitting was car- è i=1 i=1 è i=1 i=1 ø i=1 øi=1 ø
r receives the output of the hidden layer and has 160 output , which is R2 = R2 = R2 =
the hidden layer and has 160 output y, ried out for the prediction data. Then the æ l æ æ ll æ öl öæ æ l l æ ö l ö æ öæ löl æ öl öæ l æ ö l2 ö ö 2 ö
2 2 2 2

ç l å ŷi 2 – � åççŷŷlii2å
ç çl å ÷� ÷çŷçiå
2
åŷçi ÷å
l� –çi ÷lçå
yi 2÷ŷ� å÷÷ ççyylii2å
÷� ÷çyiå�yçi ÷å÷yi ÷ ÷
2
which10
cycle time from is divided
pieces to into1600
the cycle
piecestime
in from final prediction
the production processdataof awere obtained ac- ç i=1 ç è i=1 ø ÷øè çèi=1i=1è øi=1÷ø çè øè i=1 ø ÷øè i=1 è øi=1÷ø ø ÷ø
è è i=1è i=1 ø
i=1 è i=1
10 pieces to 1600 pieces in the produc- cording to the fitting curve, and finally
t. Accordingtion to process
experience (4)
of a [18],
certainthe outputAc-
garment. layer compared
activationwith function
the real data.Where,
Where, represents
Where,represents the
represents
truethe value theoftrue
true thevalue
value ofsample,
the of the sample, -sample,
the predicted - the- predic
theval p
cording to experience [18], the output Where, yi represents the true value of the
function, layer
whichactivation
has a faster convergence speed than
function selects the Relu 1) Select 89 groupssample,
andof production
andsample,
sample, isdata,
theand
and number
isi the is number
sample, the
of samples.
number
y^i – of of
The
thesamples. samples.
predicted range The ofThe
value range coefficient
ofrange the of coeffici
of i coefficient is be
the weight of function, which layer,
the output has a faster
and convergence including
is the output layer five categories of dress, skirt,
threshold. sample, and l is the number of samples.
The closer
TheitThe is tocloser
closer 1,
it the
is it
tobetter
is1,tothe 1,thethefitting
better better
thedegree the fitting
fitting is.
degree degree is. is.
speed than sigmoid and tansig. wmn is the shirt, knitwear and coat; the minimum The range of coefficient R2 is between
ion is expressed
weight as of follows:
the output layer, and bmn is the production quantity Theisresults
181 show
The The
resultsresults
that
pieces, showtheshowfitting
that
0-1. that
thedegree
The closer the fitting
fitting of
it the
degreeis degree
totrained
of1, the ofneural
the the
trained
better trained network
neural
the neural is good
networknetw
output layer threshold. The Relu function the maximum production quantity – fitting degree is.
f (x) = max(0, x) of production of production
of productiondata ofdata prediction
data of samples
of prediction prediction aresamples
samples inputtedare are to inputted
inputted the to neural to th
the ne
is expressed as follows: 1823 pieces, the minimum GST per cap-
f (x) = max(0, x).    (3) ita – 45 seconds, thesimulation maximum prediction,
simulation
simulation
GST per The results
and
prediction,
prediction, 10 groups
and 10 show
and of
groups that
10prediction
groups ofthe offitting
cycle
predictionprediction degree
time cycle can
cycle be time
time obtaine
can be ca
capita – 277 seconds, the minimum ini- of the trained neural network is good.
al 2）Due 2） to the2）
Due disturbance
Due
to thetodisturbance
the disturbance
of neural ofnetwork
tial cycle time – 60 seconds, and the max- 10 groups of production data of predic-
of neural
neural prediction
network network data,
prediction prediction the
data, latter
data,
thepredi the
latt
Experimental imum initial cycle may time is 450 seconds. tion samples are inputted towhichtheviolatesneural
, the production data of a women's clothing enterprise in Hangzhou, bemay larger
maybe thanbe larger
larger thethan former
than
the the one,
former former
which one,one, violates
which the violates
actual the produc
the
actual act
In this research, the production data of Because the input data size needs to be network for simulation prediction, and
Therefore, Therefore,
curve fitting
Therefore, curve curve
of thefitting
fitting neural
of the ofnetwork
the neural
neural output
network network data
output output
is needed
data data befor
is needeis n
d to carry outa women’s clothing 120
the experiment. enterprise
groupsin of
Hang- unified
production in the data
process neuralofnetwork, the cycle 10 groups of prediction cycle time can be
zhou, China, was used to carry out the time of different production prediction quantities
results
prediction
prediction results obtained.
are results
obtained.
are are According
obtained. obtained. According According
to the to production
theto the productio
productionquantity,qu
om May to July 2019 were obtained. 89 groups were selected for neural
corresponding
corresponding
prediction
corresponding cycle
prediction
predictiontime,
cyclecycle
such
time,as
time,
the such
such first as
group
as the the of
first first
data.
groupgroup
ofThe
of
dat
g and 10 for a simulation test. The data experiment was carried out in
180 quantity0.15
is 407,
quantity
quantity therefore
is 407,
is 407, select
therefore
thereforea select
prediction
select acycle
prediction
time
a prediction from
cycle
cycle 10time
time to 410,
from from
10 and
to10
41ft
7b. Through the training of the neural network, 10 groups of prediction
predicted cycle time
actual cycle time
in MATLAB inthrough
MATLAB
in MATLAB the through
4th degree
through thepolynomial,
the 4th 4th degree
degree polynomial,
and get and
polynomial, the following
andthe
get getfollowi
thefittin
foll
170 0.1
ned, and curve fitting was carried out for the prediction data. Then the 4
f (x) = 3.495e =f (x)
f (x) � 09 �= x3.495e
3.495e ��09 4
��x09
4.196e x 4 ��x4.196e
��06
4.196e 3 3
+ 06 ��x06
� +� x 3 +
160 0.05
data were obtained according to the fitting curve, and finally compared
Cycle time, s

� x 3 0.001986 xx23 +168.5

� x 3 ��0.4517 � x 2 +168.5
� x 2 +168.5
Cycle time, s

� 0.4517
0.0019860.001986 � 0.4517
ta. 150 0
Figure 4Figure
shows
Figure
4 the
4 fitted
showsshows
theproduction
the fitted
fitted production
learninglearning
production curve.
learning
Takecurve.
41Take
curve. groups
Take of
41
41 gro
oups of production
140 data, including five categories of dress, skirt, shirt,
quantity intoquantity
theinto
quantity
-0.05 formula
into the
to get
formula
the formula the
to final
tothe
get get
41final
the
groups
final
41 of41predicted
groups
groups ofcycle
predicted
of predictedtime.
cycl
oat; the minimum
130 production quantity is 181 pieces, the maximum -0.1

ntity - 1823 pieces,

120 the minimum GST per capita - 45 seconds, the -0.15
0   
50 100 150 200 250 300 350 400 450 0   
50 100 150 200 250 300 350 400 450
per capita - 277 seconds, the minimum initial cycle time - 60 seconds,
Production quantity Production quantity
um initial cycle time is 450 seconds. Because the input data size needs to
Figure 5. Comparison of predicted cycle time and actual cycle time.
e neural network, the cycle time of different production quantities isFigure 6. Error percentage of predicted cycle time.
10 the last piece completed, and the interval being 10 pieces.
0 pieces, with FIBRES & TEXTILES in Eastern Europe 2021, Vol. 29, 1(145)
2) Due to the disturbance of neural net- Table 1. Predicted cycle time errors.
work prediction data, the latter prediction
Predicted Actual
cycle may be larger than the former one, No
Production Cycle Initial cycle
production production Error, %
quantity time, s time, s
which violates the actual production log- time, s time, s
ic. Therefore, curve fitting of the neural 1 407 110 180 5499.6 5400 1.8
network output data is needed before the 2 300 115 230 5360.3 5190 3.3
final prediction results are obtained. Ac- 3 300 140 180 4428.3 4590 3.5
cording to the production quantity, select 4 820 211 390 19520.9 19030 2.6
the corresponding prediction cycle time, 5 525 216 260 11127.7 11820 5.9
such as the first group of data. The pro- 6 410 122 200 6682.2 7045 5.2
duction quantity is 407, therefore select 7 870 164 230 13856.9 14181 2.3
a prediction cycle time from 10 to 410, 8 595 206 220 10767.1 10450 3.0
and fit the curve in MATLAB through 9 445 180 210 8420.5 7520 11.9
the 4th degree polynomial, and get the 10 190 157 180 2786.7 3010 7.4
following fitting formula:

f (x) = 3.495e – 09 × x4 – 4.196e – 06 × x3 +

+ 0.001986 × x3 – 0.4517 × x2 + 168.5
(5)
Figure 4 shows the fitted production
learning curve. Take 41 groups of produc-
tion quantity into the formula to get the
final 41 groups of predicted cycle time.
Results and discussion
Compare the final predicted cycle time
with the real cycle time, as shown in Fig-
ure 5, The error percentage is shown in
Figure 6. It can be seen that the sorting
error is within 10%, and most of the er-
rors are within 5%. Finally, the error of
the complete production time is calculat-
ed. The predicted time is 5499.6 seconds,
the actual time 5400 seconds, the error
99.6 seconds, and the overall error per-
centage is 1.8%. It can be seen that the
trained neural network can predict the
product cycle time and complete produc-
tion time in a certain error range.
Take the above analysis method to pre-
dict the production cycle time of 10 test
groups, the overall error of which is
shown in Table 1.
It can be seen from the prediction results
that among the 10 groups of prediction re-
sults, the overall error of 6 groups is with-
in 5%, and that of 3 groups is between 5%
and 10%. The overall prediction results
are better. The error of prediction results
of group 9 is larger than others. After anal-
ysis, the initial cycle time of this group is
210 seconds, and the error of the predic-
tion cycle time of the first 200 pieces of
the neural network is within 5%. Howev-
er, due to certain reasons, such as simple
style, or high similarity with the previous
one, the production cycle time drops rap-
idly to 150 seconds at 270 pieces, while
the neural network prediction result is
180 seconds, and thus there is an error.
Conclusions
In the production of clothing manufactur-
ers, experience is often used to estimate
the complete production time of orders,
which is subjective and lacks scientific
basis. Although a large amount of data
have been accumulated in production, it
cannot be combined with the results. As
a machine learning model, a neural net-
work can establish the potential relation-
ship between nonlinear variables, and the
larger the amount of data, the stronger
the learning ability. In recent years, with
the development and accumulation of big
data, the neural network has risen again.
As a traditional manufacturing industry,
garment manufacturing has accumulated
a lot of data and has certain rules.
By establishing a neural network model
of garment cycle time prediction, we can
predict the production cycle time. The re-
sult shows that the deviation between the
prediction of the neural network and the
real value fluctuates within 10% in each
cycle time prediction, or is greater or less
than the real value. In the prediction of
the total production time, the error be-
tween the prediction results of the neural
network and the real value is mostly con-
centrated in the range of 5%, and the av-
erage error of 10 groups of the prediction
group is 4.69%. In the actual production,
we can predict the total production time
within a small error range, help the man-
ufacturer to arrange the production plan
reasonably, avoid the imbalance phe-
nomenon of order accumulation or pro-
duction waiting caused by the uncertain
production time, improve the production
efficiency, and realize lean production.
It should be pointed out that in some cas-
es, e prediction of the production cycle
time of the neural network will produce
large deviation: for example, the sam-
ple data are not enough; or the produc-
tion quantity is small, for instance, only
20 pieces; or the production cycle time
is for personalised minority clothing. For
example, in the 10 cases in this study,
one has a large overall error. One possi-
ble reason is that there is no sample with
a small amount of production and a rapid
decline in the cycle time in the sample
data. Compared with the prediction cycle
time of the mathematical model, the neu-
ral network has the function of self-learn-
ing, that is, with an increase in data ac-
cumulation, the prediction accuracy can
be further improved. In practical applica-
tion, the neural network can be trained by
a large number of data, which can avoid
this kind of situation.
Acknowledgements
This study was funded by Zhejiang Soft
Science Foundation Project (Grant No.
2020C35038).
References
 1. World Trade Organization. www.wto.org.
 2. CHINA COTTON ASSOCIATION. www.
china-cotton.org/data/cottonDataList.
 3. Zhang J, Liu Y, Cheng L. Structu-
ral Changes and Growth Factors of
China’s Textile Industry: 1997-2012.
FIBRES & TEXTILES in Eastern Eu-
rope 2018; 26, 2(128): 20-25. DOI:
10.5604/01.3001.0011.5734.
 4. Salerno-Kochan R, Kowalski M. Safety
Management of Textile Products in the
European Union and Estimation of its
Efficiency. Part 1. FIBRES & TEXTILES
in Eastern Europe 2020; 28, 2(140):
8-14. DOI: 10.5604/01.3001.0013.7307.
 5. Antczak A, Greta M, Kopeć A, Otto J.
Characteristics of the Textile Industry
of Two Asian Powers: China and India.
Prospects for their Further Development
on Global Markets. FIBRES & TEXTI-
LES in Eastern Europe 2019; 27, 5(137):
9-14. DOI: 10.5604/01.3001.0013.2895.

FIBRES & TEXTILES in Eastern Europe 2021, Vol. 29, 1(145) 11

 6. Kumar DV, Mohan GM, Mohanasunda-
ram KM. Lean Tool Implementation in
the Garment Industry. FIBRES & TEXTI-
LES in Eastern Europe 2019; 27, 2(134):
19-23. DOI: 10.5604/01.3001.0012.9982.
 7. Wright TP. Factors Affecting the Cost
of Airplanes. Journal of Aeronautical
Sciences 1936; 3 (4): 122-128.
 8. Smunt TL, Watts CA. Improving Ope-
rations Planning with Learning Curves:
Overcoming the Pitfalls of ‘Messy’ Shop
Floor Data. Journal of Operations Mana-
gement 2003; 21 (1): 93-107.
 9. Smunt TL. A Comparison of Learning
Curve Analysis and Moving Average
Ratio Analysis for Detailed Operational
Planning. Decision Sciences 1986; 17
(4): 475-495.
10. Smunt TL. Incorporating Learning Cu-
rve Analysis Into Medium-Term Capa-
city Planning Procedures: A Simulation 2nd International Conference on Cellulose Fibres,
Experiment. Management Science the fastest growing fibre group in textiles, the largest
1986; 32 (9): 1164-1176.
11. Christoph H. Glock. Applications of Le-
investment sector in the bio-based economy
arning Curves in Production and Ope- and the solution for avoiding microplastics.
rations Management: A Systematic Li-
terature Review. Computers & Industrial The conference will cover the entire value chain from
Engineering 2019; 131(5): 422-441.
12. Attia EA, Megahed A, Duquenne P. the lignocellulosic feedstock, dissolving pulp, cellulose
Towards a Learning Curve for Electric fibres – such as rayon, viscose, modal or lyocell
Motors Production under Organizational
Learning via Shop Floor Data. IFAC-Pa-
and new developments, to a wide range of applications,
persOnLine 2016; 49(12): 1086-1091. woven textiles (clothing) and non-wovens (wipes
13. Hu S, Zhang L, Zhang W. Cycle Time
Design Model in Garment Assembly
and technical applications). All these sectors have
Line Based on Skilled Ratio. Journal of significantly gained in dynamics over the last few years.
Textile Research 2015, 36(05): 133-138.
14. Kalkanci M, Kurumer G, Öztürk H, Si- Cellulose fibres are a success story within the textiles
necen M, Kayacan Ö. Artificial Neural
Network System for Prediction of Di-
market with a cumulated annual growth rate (CAGR)
mensional Properties of Cloth in Gar- between 5 and 10% over the last ten years. This makes
ment Manufacturing: Case Study on a
T-Shirt. FIBRES & TEXTILES in Eastern
cellulose fibres the fastest growing fibre group in the
Europe 2017; 25, 4(124): 135-140. DOI: textile industry and also the largest investment sector
10.5604/01.3001.0010.2859. in the bio-based economy worldwide. The high growth
15. Hasnat A, Ghosh A, Khatun A, Halder
S. Pattern Classification of Fabric De- rates are driven by the demand for natural fibres
fects Using a Probabilistic Neural Ne- (and bottlenecks in cotton), the microplastic problem
twork and Its Hardware Implementation
using the Field Programmable Gate and possible bans for plastic fibres. All three drivers
Array System. FIBRES & TEXTILES in will continue to play a significant role in the future
Eastern Europe 2017; 25, 1(121): 42-48.
DOI: 10.5604/01.3001.0010.1709.
development of the sector.
16. Jin J, Yang Y, Zou F. Developing an
Intelligent Model for the Construc-
tion a Hip Shape Recognition System
Based on 3D Body Measurement.
FIBRES & TEXTILES in Eastern Eu-
rope 2016; 24, 5(119): 110-118. DOI:
10.5604/12303666.1215535.
17. Jaouachi B, Khedher F. Evaluation of Se-
wed Thread Consumption of Jean Tro-
users Using Neural Network and Regres-
Organiser Contact
sion Methods. FIBRES & TEXTILES in
Eastern Europe 2015; 23, 3(111): 91-96.
Dominik Vogt
DOI: 10.5604/12303666.1152518. Phone: +49 2233 48-1449
18. Michael P, Stephanie P. Practical e-mail: dominik.vogt@nova-institut.de
MATLAB Deep Learning: A Project-Ba- nova-institute.eu
sed Approach 2020.
nova-Institut GmbH cellulose-fibres.eu
Received 12.04.2020 Reviewed 21.05.2020

12 FIBRES & TEXTILES in Eastern Europe 2021, Vol. 29, 1(145)