Asia Pacific Journal of Emerging Markets, Vol. 4, No.

1, 2020 ISSN: 2706-5634 79

Role of Computer Programming in Processing and

Management of Textile Industry- A Review

Saba Naseem*
Faculty of Computer Science,
National University of Computer and Emerging Sciences (FAST),
Faisalabad, Pakistan
*Corresponding author

Lyba Kashif
Department of Computer Science,
National Textile University, Faisalabad, Pakistan

Misbah.
Department of Computer Science,
Beijing Institute of Technology, Beijing, China

Dr. Sobia Mahmood

Assistant Professor
Preston University, Islamabad, Pakistan

Abstract
Computer science plays a crucial role in every fields of life it makes
the work easier as it produces more work in less time. As move towards
industries, computer programming appear as a remarkable change in
fields of textile industry. Involvement of computer science in terms of
Artificial Intelligence (AI), Data mining, Machine Learning, Neural
Network, image processing etc. plays a vital role in different phases of
textile industry like in processing, spinning, knitting, marketing,
management, designing and many [Link] this review paper, working
of computer programming in processing and management of textile
industry is reviewed. After reviewing the papers, it concludes that by
using computer programming in textile industry, results will be
efficient, reliable and more productive in less time.
 Role of Computer Programming in Processing and Management of Textile
Industry 80
Keywords: Marketing; Management; Computer Science; Textile
industry; Artificial Intelligence (AI); Data mining; Machine Learning;
Processing; Spinning.

Reference to this paper should be made as follows: Naseem, S.,

Kashif, l., Misbah., Mahmood, S. (2020) ‘Role of Computer
Programming in Processing and Management of Textile Industry- A
Review’, Asia Pacific Journal of Emerging Markets, Vol. 4, No. 1, pp.
79–92.

Biographical notes: Saba Naseem is serving as a Faculty Member in

National University of Computer and Emerging Sciences- FAST
university CFD Campus Faisalabad. Dynamically, she’s been having
multiple experiences handling graduate students throughout her career.
She has done her MS in Computer Science (MSCS) from National
Textile University, Faisalabad. By her academic contributions, she has
always served her field by means of proposing unique adaptations for
a better tomorrow.

Lyba Kashif is working as a full-time Freelancer. After her graduation

in 2018, she joined Tecbeck Pvt. Ltd. as a Business Analyst. She also
did the role of AI/ML Team Lead in her company. She has knowledge
about many emerging tools and technologies regarding Web and
Mobile app development. Lyba is passionate about learning new
experiences and technologies. She has done her MS in Computer
Science (MSCS) from National Textile University, Faisalabad. She
has always served her company with new ideas of better management
and planning for adaptive and reliable systems.

Misbah.N is doing PHD in Bioinformatics and she has done her

master’s in computer science from Beijing Institute of technology,
Beijing, China. Her area of interest is Computational Bioinformatics
in which she is working in designing aptamers computationally. By
her academic contributions, she has always served her field by means
of proposing unique adaptations for a better tomorrow.

Dr Sobia Mahmood is an Assistant Professor at Preston University,

Islamabad, Pakistan. Co-supervising PhD thesis and a member of the
Doctoral Review Committee, reviewing PhD thesis. Published several
articles and one book. Have diverse experience of teaching at post-
81 S, Naseem. L, Kashif. Misbah. S, Mahmood.

graduate and graduate level in Business & Management Sciences.

Enhanced the skills and capabilities of students through guiding,
directing and mentoring. In both academic & professional life, she has
been consistently praised as adaptable by professors & pears. Ensuring
a quality outcome by adopting innovative and sophisticated methods
of work. Obtained PhD from Bahria University, Islamabad, Pakistan,
MS from SZABIST, Islamabad, Pakistan and MBA from the
University of Arid Agriculture, Rawalpindi, Pakistan.

1. Introduction

Just like in a series of industries, Textile industry proves itself very

advantageous, and it’s a need of time in human life. There are different
characteristics and components in the textile industry, including yarns,
production methods, quality detection and finishing processes. All these
components are related to each other to produce a finishing fabric. Various
chemicals, dyes, and sizing materials are used in the production of
 Role of Computer Programming in Processing and Management of Textile
Industry 82
fabrics(Madhav et al., 2018). Methods for textile processing are composed of
fiber processing, yarn processing, fabric pretreatment, dyeing, printing, coating,
finishing treatments, and managing (transport, sales, and retails). Figure 1
illustrates the flow diagram of the processing steps.

Figure 1:(Lee & Lin, 2018) A flow diagram for steps involved in the processing
of textile industry
Management of textile industry plays vital role in overall operations of textile
industry like quality management, supply and demand management, distributing
and marketing management, consumer behavior and analysis, retailing and
sustainability etc. Likewise, CS works well in processing, CS shown a
remarkable performance in field of textile management in order to enhance its
selling, production, efficiency/quality. By using CS techniques in management
of textile industry, it shows an extraordinary innovation in order to increase in
production while consuming less time and effort. The development of computer
technology in recent years has contributed significantly to industrial
advancement, particularly to industrial automation. These technological
advancements have led to the automation of various processes such as design,
production, and inspection in the textile industry.
Other than chemicals and sizing materials are used, computer programming is
involved. This article will review about two basic fields of textile industry by
using Computer Programming techniques in these fields. First, Computer
Programming applies in not all steps but some phases of processing like Software
tools used in printing or dyeing, artificial intelligence used at some level, and
many more. This review paper reviewed how knowledge from the computer
programming area is implemented for the advancement of significant steps of
textile processing. It includes knowledge domains like artificial intelligence,
machine learning, neural networking, data mining, and some other modern
programming tools that are being used for dying, printing, designing, and yarn
processing. Second, CS widely covered almost all aspects of textile management,
but this review paper highlights the involvement of computer programming in
quality detect defects, and supply chain management.
Therefore, this review explores and analyzes the conventional tools and methods
available for the use of processing and management phases in the textile industry.
Using computer programming in processing methods (dyeing, printing, yarn,
83 S, Naseem. L, Kashif. Misbah. S, Mahmood.

designing, quality checking, and finishing process) and management phases

(quality management and supply chain management),the processing and
management work in the textile industry will more accurate, less time
consuming, more reliable, and efficient.
This review paper is structured as follows: Section 2 discusses the
implementations of computer science in yarn processing. Section 3 describes
how Computer Science is involved in the dyeing of textiles fabrics. Section 4
discusses the printing method of textile processing, which requires some
programming tools. Section 5 discusses about programming method used in
management fields of textile industry. Section 6 concludes all discussion.

Section 2: Computer Science in yarn processing

(Vitor Carvalho et al., 2013) have worked to evaluate the quality of yarns, and
to do this researcher provides a system or technique which mainly depends on
image processing. Both yarn’s production characteristics and mass
parameterized are done by using image processing. To implement the proposed
system, a web camera which depends on less expensive solution, less expensive
optics microscope and IMAQ Vision from LabVIEW are used. According to the
result’s analysis, the proposed system of yarn parameterization provides a better
solution to assess the quality of yarn by providing reduced hardware, weight,
volume, easy maintenance, and low cost.
For the process of industrial production, (Sette & Boullart, 2000) aim of this
method (GBML) is the pre-programmed development of a ruleset. Fuzzy
efficiency based classifier system (FECS) was effectively implemented to the
case study, “to predict the spinnability for the process of fiber to yarn
production,” in which rules are generated to predict the spinnability to produce
the process of fiber-to-yarn. As an experimental analysis, ruleset of 123
classifiers was taken and produced, which allows to predict 94% accuracy. (Roy
et al., 2014) have also worked on yarn parameterization and production
characteristics to evaluate the yarn quality. To determine the yarn parameters in
the form of uniformity of yarns and yarn hairiness, researchers used
Segmentation, Image Acquisition, Image analysis, and Image thresholding
algorithms of image processing.
Many tools and devices are used to determine yarn appearances. By using image
processing and artificial intelligence, a new computer vision technique is
 Role of Computer Programming in Processing and Management of Textile
Industry 84
implemented for grading of yarn appearance as detecting and classifying of yarn
faults which are majorly based on yarn images of ASTM standard. The proposed
method is used to grade any yarn because it is independent of yarn’s structure
and raw materials. (Semnani et al., 2006)
The automated method for the analysis of the structure and extraction of Yarn
Positional Information of textile fabrics is desired for obtaining accurate and
rapid results. To determine the yarn positional information, 3D images that get
from X-ray CT images, there is a newly implemented, which is used to take out
the positional information of yarn for analysis of the textile fabric structure. The
information on the yarn is centered line in the form of a sequence of the points
that are referred to as “positional information of yarn.” So, the sequence is then
taken out by detecting the yarn and by using the direction of yarn, which got
through evaluating its filaments direction, and to average the evaluated filament
direction, the yarn is being traced. The filament direction is determined by
correlating the filament part with the 3D filament model in the 3D CT images.
The implemented method is applied and useful for both the 3D CT images of
woven fabric and plain knitted fabric. (Shinohara et al., 2010)
Using artificial intelligence and image processing, there’s a novel idea presented
which preprocessed the quantity of production of yarn properties (direction of
cable and fibers, cable numbers and snarl length), yarn mass factors (mass,
diameter, and hairiness) and prediction of yarn linked fabrics. (Vítor Carvalho et
al., 2009) given method which restricts the limitations of previous commercial
users which is used for measurement of yarn quality factorization because of its
less weight, low cost, low volume, having higher efficiency and resolution, more
stability of technology, lessen maintained and lessen the complexity of hardware,
for the control of manufactured process it presents the probability of online use.
So, the results of this implemented approach will be given in which, among all
other advantages for the industry of textile, a new stage of factorization is
introduced, which allows the increasing of products, quality, and higher
efficiency, which has a massive contribution in recovery of economy.
Section 3: Computer Science in the Dyeing process
A lot of discussion and research is being carried out in the field of Artificial
intelligence and textile processing currently. Researchers work on finding the
applications for AI to be applied in the textile processing and color industry.
Expert systems and neural networks have contributed a lot and then explores
different AI subdomains in further sections. Expert systems contributions
85 S, Naseem. L, Kashif. Misbah. S, Mahmood.

include systems like fastness performance prediction, recommendations like

dyeing methods, suitable dyes, color matching, pad steam dyeing, polyester
exhaust dyeing, complete dyeing control system, and fluorescent whiteners
selector on the initial stage. It also includes many other examples from other
stages of textile processing like knitting, weaving, default detection, etc.
(Shamey & Hussain, 2003) explains the fuzzy logic, fuzzy expert system in detail
how it is implemented in the textile processing industry. As an example, a fuzzy
inference system (FIS) related to color grading is mentioned in the paper, which
is used to reduce machine-classer disagreements in cotton color classification.
Other applications include fabric defect identifying system, controlling the
concentration of caustic soda on fabric during processing, controlling the speed
of looms.
Water can be saved by optimizing production scheduling in the textile dyeing
industry states. (Zhou et al., 2017) proposed a methodology using a genetic
algorithm. The objective is to save water so that serious water environmental
problems can be avoided. The methodology used for such purpose was a genetic
algorithm to optimize dyeing production scheduling. To achieve this, a
scheduling system having a DB and MATLAB program implementing a genetic
algorithm was developed. The results depicted that 18.4 – 21% of freshwater can
be saved using this method after rescheduling 50 – 70 orders.
Recipe prediction for getting the required color and shade on a given substrate is
also a hot topic. (Jawahar et al., 2015) stated that the Kubelka model, which was
used before for the prediction, often failed under many conditions so, he
introduced a neural network-based model to predict color depending on three
values known as tristimulus values (x,y,z). The methodology includes training
of neural networks with 300 pairs of known input vectors using the
backpropagation algorithm. Three neurons, including the input layer, three
neurons for output layer containing x,y,z values, and five neurons embedded in
a hidden layer having an implementation of the log-sigmoid transfer function.
Concluded that the neural network can effectively provide better predictions
about color matching.
COMIC was a Colorant Mixture Computer that was used for matching colors
with given input colors. Different researchers have worked on analyzing the
working of Comic and comparing it with other similar systems that were used to
do the same work but efficiently and affordably. COMIC was the first computer
that could successfully find the matching color with the given input color. It was
 Role of Computer Programming in Processing and Management of Textile
Industry 86
an analog computer, but later it was modified into digital computers. It was used
in the textile colorant industry, and it had some limitations as well then on
industry switched to digital computation gradually. COMIC was developed to
solve major color matching problems. These problems included finding the
mixture of dyes to match input color, in different lighting, the colors may not
match with the sample. In contrast, it was matched in some other light like
daylight, which results in major problems when applying the colors on different
materials like cotton and wool. The early prototype of COMIC used a multipole
stepping, which rotated continuously to display on an Eico oscilloscope, the
actual production, whereas replaced the stepper switch with digital vacuum tube
circuits. Those circuits generated sequences of pulses to provided signals to the
HP oscilloscope that was built into the computer. Many companies used it for
years, but limitations of COMIC were the main reasons that people started to
switch to IBM 1130, claiming that COMIC had less storage and was not enough
and costly as well. The working of COMIC internally was also too much that
required table lookups of data and manual entry on dials, and the user of the
system would select a set of colorants and is the results were not matched that is
required to make manual selections again. COMIC was successful due to its
digitalization, but then other computers that were economically affordable for
people were used later on but not on a wide scale. (Hemmendinger, 2014)
(Wu & Chang, 2008) described how textile-dye manufacturing takes place,
keeping in mind the impacts of pollution charges, resource conservation fees,
different other costs, and limitations of production capacity as well. It not only
focused on conventional or external costs but also hidden costs. It involved a
multi-objective evaluation methodology. The objective functions were proposed
for optimal production. These objective functions were to minimize total
production cost and environmental cost, maximize total production capacity, and
minimize inventory cost. Fifteen techniques were evaluated to form an
evaluation matrix. The results at the end indicated that enterprises should
implement production assessments to increase their productivity and
competition.
A methodology was proposed to optimize the cost of enzyme washing for indigo-
dyed cotton by (Xu et al., 2020). The methodology involved combing of Kriging
surrogate and differential evolution algorithm. It is carried out in three steps.
First, the temperature and concentration are focused on keeping the processing
time as an input parameter, and this experiment produces different outputs such
87 S, Naseem. L, Kashif. Misbah. S, Mahmood.

as washing performance, stiffness, etc. The second step involves identifying the
relationships between input and output by applying the kriging model. Then at
step 3, the analysis is performed on washing performance, and then the model is
illustrated. A combination of the kriging and differential evolution models can
have a significant impact on the optimized cost of the enzyme's washing process
for indigo dyed cotton.
(J. Su et al., 2015) proposed the tie-dyeing simulation of art, pattern, and feature.
It explains the fractal theory imitative transformation, and some functions with
an experiment of patterns and parametrizations are also discussed. The
increasing need for a wide variety of textiles with different patterns and dyes
urged the concerned people to focus on digital printing done through
CAD/CAM. Now a day’s graphic designing and pattern design is developing
quickly and widely used in textile designs. To design patterns and combinations,
experts use a combination of different tools to achieve their goals, such as
Photoshop, illustrator, processing software, etc. The results conclude that putting
programming methods can also be used to obtain different patterns in the modern
textile industry.
Section 4: Computer Programming in Printing and Designing
Traditional methods of design, printing and dyeing may not fit the needs of
society, the emergence of dyeing and digital printing technology has
significantly altered the traditional textile industry production method, which has
changed the way people use and thoughts for textiles. (J. X. Su, 2015) have also
worked on Software Methods for Digital Textile Pattern. Based on the analyzing
of digital textile picture, pattern, and feature, Digital textile computer simulation
is proposed in this paper. They are created by experiments of classic patterns and
parameters or combination of several software like photoshop, illustrator,
Processing software and Kaleidoscope software is used to the result shows that
the methods of making digital textile pattern are useful and available. Or by
combine several software’s to obtain patterns, the application of digital
technology in the modem textile pattern design, create a new pattern style.
CAD/CAM technology is becoming increasingly apparent in the textile
industries. (Collier & Collier, 1990) have worked to CAD/CAM in the Textile
Industry. Adoption of voluntary standards of electronic data usage is central to
such integration. Additionally, the linkages between textile production and
distribution should be emphasized, and the facilitation of these linkages is
suggested as an important research focus The textile industry may already be
 Role of Computer Programming in Processing and Management of Textile
Industry 88
effective in other industries through CAD / CAM processes, but further progress
requires the development of more specialized processes and tools/equipment.
Section 5: Computer Programming in management of textile industry
Machine learning and computer vision techniques are used to detect the defects
automatically. In order to detect defects most used systems need textiles to be
spread flat. (Syed & Nazir, 2017) described method is simple yet efficient, novel
method which is used for non-uniform and spacious shape’s textiles. In order to
detect various kind of textile problems, classification and normalization
techniques are merge in a decision tree model. As an analysis of result, proposed
method is much accurate by using key points of convolutional neural networks.
Rule based classification system with local binary features by using pattern
texture is used to detect the problems in textiles. By the use of rough set-based
method, rules are learned automatically from the samples of textile. (Lizarraga-
Morales et al., 2019) proposed method utilizes local binary features to analyze
the quality and appearance of fabrics, due to this characteristic proposed method
is highly differentiated. This approach is divided into two steps i.e. training and
[Link] training step, binary features are obtained and used to generate a group
of rules which is based on rough set from both defect free and defective [Link]
testing phase, fabrics with various samples are submitted and then they are
organized as defect free or defective samples. In experimental analysis, results
of proposed method show that this method has more accuracy than the results
obtained from state-of-the-art method.
There is a need of an efficient models and management systems in textile
industry to manage the market fluctuation which make the supply chain
management more reliable. For this, (Hwang & Seruga, 2011) proposed a system
in supply chain management of textile industry “collaboration network model”.
This system is developed to enhance information sharing, lessen the product life
cycle time (PLC), to boost the delivery time and customer services. By using this
collaboration network model, this lessens the gap of the competitiveness beyond
the supply chain of textile industry and to improve the collaboration in order to
boost the whole competitiveness of textile industry. Selection of supplier is a
basic operation in phase of supply chain management of textile industry. (Burney
& Ali, 2019) proposed a simple yet efficient method for supplier selection
operation in textile industry “fuzzy analytic hierarchy process (F-AHP).it is a
decision support system based on Fuzzy multi criteria to select the supplier in
supply chain management of textile industry. Proposed method applies fuzzy
89 S, Naseem. L, Kashif. Misbah. S, Mahmood.

technique of soft computing in analysis of AHP in order to allow the decision

maker to manage with imprecision, linguistic chaos and vagueness while doing
pair-wise comparison.

5. Conclusion

In this review role of computer programming has been discussed and its
usage/involvement in processing and managing fields of textile industry. Textile
processing is a step wise procedure and this paper has reviewed the textile
processing steps using Computer science in detail also the management phases
of textile industry. Starting from yarn processing, image processing is used to
check the quality of yarn different algorithms have been implemented to predict
spinnability to produce Fibre to yarn and evaluate the yarn. Artificial
intelligence, neural network and data mining fields are used in different
processing phases of entire textile industry. In the business environment,
information technology (IT) plays an important role for Textile industrial
performance. It provides information flow of management which makes the
supply chain more robust and resilient without undermining its efficiency. Defect
detection is also being performed on textile fabrics to detect defects using image
processing techniques. Similarly, computer science is involved in dyeing of
textiles in many ways. The most important thing in dyeing is the color matching
and CS has really played its part well in analyzing the colors for colorant
industry. A fuzzy inference system (FIS) related to color grading is mentioned
in the paper, which is used to reduce machine-classer disagreements in cotton
color classification. Similarly, many other applications are used carried out in
dyeing process. As computer programming techniques are involved in
processing and managing phase of textile industry so, in the future computer
programming and tools will used in many other fields of textile industry like
knitting and spinning etc. to reduce the time, cost ,effort and make this process
much effective than before..
 Role of Computer Programming in Processing and Management of Textile
Industry 90

References
Burney, S. M. A., & Ali, S. M. (2019). Fuzzy Multi-Criteria Based Decision
Support System for Supplier Selection in Textile Industry. IJCSNS
International Journal of Computer Science and Network Security, 19(1),
239–244.
Carvalho, Vitor, Gonçalves, N., Soares, F., Vasconcelos, R., & Belsley, M.
(2013). Yarn Parameterization and Fabrics Prediction Using Image
Processing. Textiles and Light Industrial Science and Technology, 2(1).
Carvalho, Vítor, Soares, F., & Vasconcelos, R. (2009). Artificial intelligence and
image processing based techniques: A tool for yarns parameterization and
fabrics prediction. ETFA 2009 - 2009 IEEE Conference on Emerging
Technologies and Factory Automation.
[Link]
Collier, B. J., & Collier, J. R. (1990). CAD/CAM in the Textile and Apparel
Industry. Clothing and Textiles Research Journal, 8(3), 7–13.
[Link]
Hemmendinger, D. (2014). COMIC: An analog computer in the colorant
industry. IEEE Annals of the History of Computing, 36(3), 4–18.
[Link]
Hwang, H., & Seruga, J. (2011). An intelligent supply chain management system
to enhance collaboration in textile industry. International Journal of U-
and e- Service, Science and Technology, 4(4), 47.
Jawahar, M., Narasimhan Kannan, C. B., & Kondamudi Manobhai, M. (2015).
Artificial neural networks for colour prediction in leather dyeing on the
basis of a tristimulus system. Coloration Technology, 131(1), 48–57.
[Link]
Lee, C., & Lin, J. (2018). Improve Quality and Efficiency of Textile Process
using Data-driven Machine Learning in Industry 4.0. International
Journal of Technology and Engineering Studies, 4(2), 64–76.
[Link]
91 S, Naseem. L, Kashif. Misbah. S, Mahmood.

Lizarraga-Morales, R. A., Correa-Tome, F. E., Sanchez-Yanez, R. E., & Cepeda-

Negrete, J. (2019). On the Use of Binary Features in a Rule-Based
Approach for Defect Detection on Patterned Textiles. IEEE Access, 7,
18042–18049. [Link]
Madhav, S., Ahamad, A., Singh, P., & Mishra, P. K. (2018). A review of textile
industry: Wet processing, environmental impacts, and effluent treatment
methods. Environmental Quality Management, 27(3), 31–41.
[Link]
Roy, S., Sengupta, A., & Sengupta, S. (2014). Yarn hairiness evaluation using
image processing. International Conference on Control, Instrumentation,
Energy and Communication, CIEC 2014, 11(5), 588–592.
[Link]
Semnani, D., Latifi, M., Tehran, M. A., Pourdeyhimi, B., & Merati, A. A. (2006).
Grading of Yarn Appearance Using Image Analysis and an Artificial
Intelligence Technique. Textile Research Journal, 76(3), 187–196.
[Link]
Sette, S., & Boullart, L. (2000). Implementation of genetic algorithms for rule
based machine learning. Engineering Applications of Artificial
Intelligence, 13(4), 381–390. [Link]
1976(00)00020-8
Shamey, R., & Hussain, T. (2003). Artificial intelligence in the colour and textile
industry. Review of Progress in Coloration and Related Topics, 33, 33–
45. [Link]
Shinohara, T., Takayama, J. ya, Ohyama, S., & Kobayashi, A. (2010). Extraction
of Yarn Positional Information from a Three-dimensional CT Image of
Textile Fabric using Yarn Tracing with a Filament Model for Structure
Analysis. Textile Research Journal, 80(7), 623–630.
[Link]
Su, J., Li, Z., & Sun, X. (2015). The Computer Aided Methods for Tie Dyeing
Pattern. Iiicec, 620–622. [Link]
Su, J. X. (2015). The Software Methods for Digital Textile Pattern. Applied
Mechanics and Materials, 738–739, 1328–1331.
[Link]
 Role of Computer Programming in Processing and Management of Textile
Industry 92
Syed, A. Z., & Nazir, A. (2017). Pattern Recognition - 9th Mexican Conference,
{MCPR} 2017, Huatulco, Mexico, June 21-24, 2017, Proceedings.
10267, 244–253. [Link]
Wu, C. C., & Chang, N. Bin. (2008). Evaluation of environmentally benign
production program in the textile-dyeing industry (II): A multi-objective
programming approach. Civil Engineering and Environmental Systems,
25(1), 1–28. [Link]
Xu, J., He, Z., Li, S., & Ke, W. (2020). Production cost optimization of enzyme
washing for indigo dyed cotton denim by combining Kriging surrogate
with differential evolution algorithm. Textile Research Journal.
[Link]
Zhou, L., Xu, K., Cheng, X., Xu, Y., & Jia, Q. (2017). Study on optimizing
production scheduling for water-saving in textile dyeing industry. Journal
of Cleaner Production, 141, 721–727.
[Link]