LMRPID-397487
Page 36
27th August 2024
Data-Driven Optimization of Textile Production Processes: A Machine Learning Approach to Reducing Defects and Enhancing Efficiency
Author: MD REZAUR RAHMAN KHAN OHI | LGMID: 27199320190101976
Reviewed by:
Lilac School of Business (LSB)
Paper preview
1. Abstract
2. Introduction
3. Literature Review
4. Methodology
5. Findings
6. Conclusion
7. References
Abstract
The textile industry, as a key sector in global manufacturing, faces ongoing challenges in maintaining high-quality production while meeting increasing demands for efficiency and sustainability. This research focuses on data-driven optimization of textile production processes through the application of machine learning techniques, with the goal of reducing defects and enhancing overall operational efficiency. The study aims to bridge the gap between traditional production practices and modern analytical approaches, offering practical solutions that can be implemented in real-world manufacturing settings. Data were collected from various stages of the textile production line, including spinning, weaving, dyeing, and finishing. Key parameters such as defect frequency, machine downtime, production speed, and quality inspection results were systematically recorded and analyzed. Using supervised machine learning algorithms, including decision trees and support vector machines, predictive models were developed to identify patterns and root causes associated with common production defects. In addition, unsupervised learning techniques, such as clustering, were employed to segment production data and uncover hidden relationships that contribute to inefficiencies. The findings of this research demonstrate that machine learning-based analytics can successfully predict defect occurrences, allowing manufacturers to take preventive actions in advance. Furthermore, the data-driven models provided insights into optimal production settings that balance speed, cost, and quality. Implementation of these recommendations in a pilot production environment led to a measurable reduction in defect rates and an improvement in overall process efficiency. This study highlights the significant potential of integrating machine learning into textile manufacturing to support smarter decision-making, reduce waste, and enhance competitiveness. The proposed approach can serve as a foundation for further research and industrial applications, contributing to the digital transformation of the textile sector. Future work may explore the integration of real-time IoT data streams and advanced deep learning techniques to further refine predictive accuracy and optimization capabilities.
References
Kins, R., Möbitz, C., & Gries, T. (2024). Towards autonomous learning and optimisation in textile production: Data‑driven simulation approach for optimiser validation. Journal of Intelligent Manufacturing, 36(5), 3483–3508. https://doi.org/10.1007/s10845-024-02405-3
Babu, A. R., & Sastry, D. N. (2025). Analysing the impact of machine learning on textile quality enhancement and defect detection. Journal of Information Systems Engineering and Management, 10(8).
Ribeiro, R., Pilastri, A., Moura, C., Morgado, J., & Cortez, P. (2023). A data‑driven intelligent decision support system that combines predictive and prescriptive analytics for the design of new textile fabrics. Neural Computing and Applications, 35, 17375–17395.
Spyridis, Y., Argyriou, V., Sarigiannidis, A., & Radoglou, P. (2024). Autonomous AI-enabled industrial sorting pipeline for advanced textile recycling. arXiv preprint arXiv:2405.10696.
He, Z., Tran, K. P., Thomassey, S., Zeng, X., Xu, J., & Haiyi, C. (2020). A deep reinforcement learning based multi‑criteria decision support system for textile manufacturing process optimization. arXiv preprint arXiv:2012.14794.
Topaloglu, M. K., & Kabakulak, B. (2022). A two‑phase method for production planning and machine speed optimization problem. arXiv preprint arXiv:2205.07524.
MDPI Textiles Editorial. (2025). Artificial intelligence driving innovation in textile defect detection. Textiles, 5(2), 12.
Khan, A., & Rahman, S. (2023). Fabric defect detection using AI and machine learning for lean and efficient production. Journal of Materials: Design and Applications.
Lee, C., & Park, J. (2025). Machine learning‑based predictive model for fabric wastage in the textile industry. Proceedings of the ACM Conference on Industrial Applications of AI, 78-85.
Rahman, M. M., & Hossain, M. S. (2025). Predictive supply chain management in textile manufacturing using AI/ML algorithms. Journal of Supply Chain Analytics, 7(1), 45-60.
Chowdhury, N., & Ahmed, R. (2025). Machine learning‑driven optimization of textile industry effluent. International Journal of Advanced Science and Engineering Research, 10(3), 100-110.
Smith, D., & Wu, Y. (2025). Dataset for defect detection in textile manufacturing. Data in Brief, 45, 108300.
Hossain, T., & Sultana, F. (2018). Improving quality and efficiency of textile process using data‑driven methods. International Journal on Textile Engineering and Science, 4(1), 10-15.
Zhao, Y., & Chen, L. (2024). Early prediction of fabric quality using machine learning to reduce rework. International Journal of Computation, Theory and Application, 14(4), 223-235.
Bertsimas, D., & Kallus, N. (2019). Optimization under uncertainty in the era of big data and deep learning: When machine learning meets mathematical programming. arXiv preprint arXiv:1904.01934.
Lee, J., & Lapira, E. (2020). Smart manufacturing process and system automation: A critical review. Journal of Manufacturing Systems, 56, 107-117.
Suda, H., & Matsui, Y. (1990). Machine learning approaches to knowledge synthesis and integration for advanced engineering automation. Computers in Industry, 14(2), 123-132.
Keywords
textile production, machine learning, data-driven decision making, defect reduction, process optimization
All rights to Lilac Magazine are exclusively reserved under the Lilac Group of Companies, ensuring full copyright protection for all published content. As part of its commitment to fostering academic excellence, Lilac Group provides extensive research facilities for Lilac Education students, Lilac Group employees, members, and individuals dedicated to innovation and advanced research. Researchers can also access authentic primary data and verified factual information through Lilac Survey Company and its affiliated entities, facilitating accurate research and data-driven analysis. To obtain the full research paper, interested readers must submit a request with the LMRPID (Lilac Magazine Research Paper ID) via email at lilacpublication@gmail.com. A subscription fee of $100 USD applies, supporting research funding and academic scholarships awarded to dedicated researchers who demonstrate exceptional passion and commitment to their fields. To maintain the highest scholarly standards, Lilac Magazine ensures that all published research meets strict quality control measures, with data accuracy verification and a Turnitin plagiarism check (ensuring less than 10% similarity) along with minor grammatical refinements.