<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>Recently, Industry 4.0 introduced a breakthrough in the textile industry to meet customer demands. This study aimed to accurately estimate the production rate of a knitting machine through an online monitoring system using the Internet of Things (IoT) and machine learning (ML) concepts. Experimentally, a double knitting machine was attached with sensors for gathering data of the machine speed, yarn feeder speed and stitch length while other production variables remained constant. Two prediction models were introduced since correlation results revealed multicollinearity issues among the parameters measured. The second model achieved a prediction accuracy of 100 %. Thus, it presents a novel formula of production calculation.</p>
</abstract>

Recently, Industry 4.0 introduced a breakthrough in the textile industry to meet customer demands. This study aimed to accurately estimate the production rate of a knitting machine through an online monitoring system using the Internet of Things (IoT) and machine learning (ML) concepts. Experimentally, a double knitting machine was attached with sensors for gathering data of the machine speed, yarn feeder speed and stitch length while other production variables remained constant. Two prediction models were introduced since correlation results revealed multicollinearity issues among the parameters measured. The second model achieved a prediction accuracy of 100 %. Thus, it presents a novel formula of production calculation.

Online Monitoring-Based Prediction Model of Knitting Machine Productivity

Fibres & Textiles in Eastern Europe

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

{"article-title":"Online Monitoring-Based Prediction Model of Knitting Machine Productivity"}

Recently, Industry 4.0 introduced a breakthrough in the textile industry to meet customer demands. This study aimed to accurately estimate the production rate...

	n₁(rpm)	n₂(rpm)	L(mm)	P(kg/h)
Mean	18.00	1174.63	2.11	6.26
Std	2.82	182.78	0.05	0.97
Min	1.00	21.00	0.68	0.11
25%	19.00	1237.00	2.10	6.59
50%	19.00	1238.00	2.10	6.59
75%	19.00	1238.00	2.13	6.59
Max	21.00	1362.00	2.68	7.26

Independent Parameter	n₁&n₂	L&n₁	L&n₂
R_i²	0.980	0.005	0.000
Tolerance	0.020	0.995	1.000
VIF	50.251	1.005	1.000

Online Monitoring-Based Prediction Model of Knitting Machine Productivity

Article Category: Research Article

Publicado en línea: 18 oct 2023

Páginas: 46 - 52

DOI: https://doi.org/10.2478/ftee-2023-0035

Palabras clave
Online monitoring, internet of things, machine learning, multicollinearity, knitting machine productivity, prediction accuracy

© 2023 Łukasiewicz Research Network-Łódź Institute of Technology, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Statistical data of measured parameters and the production rate

Statistical measure of degrees of multicollinearity

Results of prediction precision using (n1 and L)

Results of prediction precision using (n2 and L)

Online Monitoring-Based Prediction Model of Knitting Machine Productivity

Article Category: Research Article

Publicado en línea: 18 oct 2023

Páginas: 46 - 52

DOI: https://doi.org/10.2478/ftee-2023-0035

Palabras claveOnline monitoring, internet of things, machine learning, multicollinearity, knitting machine productivity, prediction accuracy

© 2023 Łukasiewicz Research Network-Łódź Institute of Technology, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Statistical data of measured parameters and the production rate

Statistical measure of degrees of multicollinearity

Results of prediction precision using (n1 and L)

Results of prediction precision using (n2 and L)

Palabras clave
Online monitoring, internet of things, machine learning, multicollinearity, knitting machine productivity, prediction accuracy