EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Optimizing Twill Fabric Structure’s Tensile Strength: A Multi-Objective Approach with Artificial Neural Networks as Prediction Models

Zulfiqar Ali, Zubair Khaliq, Ali Afzal, Muhammad Bilal Qadir, Fayyaz Ahmad, Badar Munir Zaidi, Amir Shahzad, Ammara Phool

Mehran University Research Journal of Engineering and Technology · 01.07.2026 · 45 (3) · 01-24

DOI: 10.22581/muet1982.0369 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mehran University Research Journal of Engineering and Technology, online erschienen am 01.07.2026. Verfasst von Zulfiqar Ali und 7 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study investigates the prediction and multi-objective optimization of tensile strength in woven twill fabrics using artificial neural networks (ANNs). A dataset of 135 fabric samples was developed using yarn tensile strength in the warp and weft directions, ends per inch, and picks per inch as predictors, with fabric tensile strength in the warp and weft directions as responses. Feed-forward ANN models with multiple hidden layers were trained using backpropagation and Bayesian regularization. The optimized models showed strong agreement with laboratory measurements. For warp-direction tensile strength, the correlation coefficients were 0.9998 for training, 0.9991 for testing, and 0.9997 for the complete dataset. For weft-direction tensile strength, the corresponding values were 1.0000, 0.99907, and 0.99994. The maximum and mean absolute errors were 8.4093 and 1.7588, respectively, for the warp direction, and 6.4159 and 0.37005 for the weft direction. The maximum percentage errors were 2.8719% and 4.28048%, respectively. A minimax-based multi-objective optimization identified the optimum manufacturing conditions as warp- and weft-yarn strengths of 495.36, 81 ends/in, and 81.9 picks/in. Experimental validation under these conditions produced tensile strengths of 585 in the warp direction and 660 in the weft direction. The results demonstrate that Bayesian-regularized ANN models can accurately predict twill-fabric tensile strength and provide practical manufacturing parameters for quality improvement and structural optimization.

Zitieren

APA

Ali, Z., Khaliq, Z., Afzal, A., Qadir, M. B., Ahmad, F., Zaidi, B. M., Shahzad, A., & Phool, A. (2026). Optimizing Twill Fabric Structure’s Tensile Strength: A Multi-Objective Approach with Artificial Neural Networks as Prediction Models. Mehran University Research Journal of Engineering and Technology, 45(3), 01-24. https://doi.org/10.22581/muet1982.0369

Vancouver

Ali Z, Khaliq Z, Afzal A, Qadir MB, Ahmad F, Zaidi BM, et al. Optimizing Twill Fabric Structure’s Tensile Strength: A Multi-Objective Approach with Artificial Neural Networks as Prediction Models. Mehran University Research Journal of Engineering and Technology. 2026;45(3):01-24. doi:10.22581/muet1982.0369

BibTeX
@article{Ali2026,
  author = {Ali, Zulfiqar and Khaliq, Zubair and Afzal, Ali and Qadir, Muhammad Bilal and Ahmad, Fayyaz and Zaidi, Badar Munir and Shahzad, Amir and Phool, Ammara},
  title = {Optimizing Twill Fabric Structure’s Tensile Strength: A Multi-Objective Approach with Artificial Neural Networks as Prediction Models},
  journal = {Mehran University Research Journal of Engineering and Technology},
  year = {2026},
  volume = {45},
  number = {3},
  pages = {01--24},
  doi = {10.22581/muet1982.0369}
}
Kostenlos starten →