Evaluating the Performances of Various Machine Learning Models for Accurate Production Forecasting in The Textile Industry

Charles Onyeka Nwamekwe¹, Nnamdi Vitalis Ewuzie², Nkemakonam Chidiebube Igbokwe³, Charles Chikwendu Okpala⁴

^1,2,3,4 Department of Industrial/Production Engineering, Nnamdi Azikiwe University, P.M.B. 5025 Awka, Anambra State – Nigeria

Abstract

Accurate production forecasting is essential for optimizing operations, minimizing waste, and improving resource allocation in the textile industry—a sector characterized by complex supply chains, fluctuating demand patterns, and dynamic market conditions. This research analyses the performances of various ML models for production forecasting, utilizing five years of historical production data alongside market and economic indicators. The research compares the predictive accuracy of Linear Regression, Decision Trees, Neural Networks, and a hybrid SARIMA-LSTM model, leveraging metrics like MSE and R2 values. A quantitative methodology was adopted, encompassing data preprocessing, model training, and evaluation. Results demonstrate that Linear Regression achieved the highest R-squared value (0.86), indicating its strong predictive capability, while Neural Networks and the hybrid model also showed competitive performance. Comparative analyses highlighted the trade-offs between accuracy, interpretability, and computational efficiency across the models. This research underscores the transformative potential of ML in addressing forecasting challenges within the textile industry, offering actionable insights for enhancing operational efficiency and sustainability. The findings contribute to the growing body of knowledge on leveraging ML techniques for data-driven decision-making in complex industrial contexts.

Keywords: Machine learning models, Production forecasting, Textile industry analytics, Time series analysis, Hybrid predictive models

References

1. Ansari, M. and Alam, S. (2024) ‘Hybrid SARIMA–BiLSTM model with Fourier terms and attention mechanisms for air quality forecasting’, Electronics, 14(3), 549. https://doi.org/10.3390/electronics14030549
2. Buban, J. and Choi, S. (2017) ‘Auto-encoders for noise reduction in scanning transmission electron microscopy’, Microscopy and Microanalysis, 23(S1), pp. 130–131. https://doi.org/10.1017/s1431927617001337
3. Chidiebube, I.N., Onyeka, N.C., Sunday, A.P. and Chiedu, E.O. (2025) ‘A comparative analysis of machine learning models for inventory demand forecasting in a food manufacturing SME’, Indonesian Journal of Innovation Science and Knowledge, 2(3), pp. 35-48. https://knowledge.web.id/index.php/ijisk/article/view/177
4. Elseidi, R., Mohamed, A. and Farouk, H. (2024) ‘A hybrid SARIMA–Prophet model for streamflow forecasting’, SN Applied Sciences, 6, 6083. https://doi.org/10.1007/s42452-024-06083-x
5. Ezeanyim, O.C., Ewuzie, N.V., Aguh, P.S., Nwabueze, C.V. and Nwamekwe, C.O. (2025) ‘Effective maintenance of industrial 5-stage compressor: a machine learning approach’, Gazi University Journal of Science Part A: Engineering and Innovation, 12(1), pp. 96–118. https://dergipark.org.tr/en/pub/gujsa/issue/90827/1646993
6. Igbokwe, N.C., Christiana, C., Nweke, C.O.N. and Onyeka, C. (2025) ‘Data-driven solutions for shuttle bus travel time prediction: machine learning model evaluation at Nnamdi Azikiwe University’, African Journal of Computing, Data Science and Informatics (AJCDSI), 1(1), pp. 31-55. https://journals.co.za/doi/abs/10.31920/2978-3240/2025/v1n1a2
7. Izza, Y., Ignatiev, A. and Marques-Silva, J. (2022) ‘On tackling explanation redundancy in decision trees’, Journal of Artificial Intelligence Research, 75, pp. 261–321. https://doi.org/10.1613/jair.1.13575
8. Juanga-Labayen, J., Labayen, I. and Yuan, Q. (2022) ‘A review on textile recycling practices and challenges’, Textiles, 2(1), pp. 174–188. https://doi.org/10.3390/textiles2010010
9. Kim, J., Kim, H., Kim, H., et al. (2025) ‘A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges’, Artif. Intell. Rev., 58, 216. https://doi.org/10.1007/s10462-025-11223-9
10. Li, X. (2024) ‘Inflation forecasting using a hybrid LSTM–SARIMA model based on discrete wavelet transform’, Advances in Economics, Management and Political Sciences. Retrieved from https://www.ewadirect.com/proceedings/aemps/article/view/10487
11. Liu, Z., Zhang, Z. and Zhang, W. (2025) ‘A hybrid framework integrating traditional models and deep learning for multi-scale time series forecasting’, Entropy, 27(7), 695. https://doi.org/10.3390/e27070695
12. Mahmoud, A. and Mohammed, A. (2024) ‘Leveraging hybrid deep learning models for enhanced multivariate time series forecasting’, Neural Process Lett., 56, 223. https://doi.org/10.1007/s11063-024-11656-3
13. Molinder, J., Scher, S., Nilsson, E., Körnich, H., Bergström, H. and Sjöblom, A. (2020) ‘Probabilistic forecasting of wind turbine icing-related production losses using quantile regression forests’, Energies, 14(1), 158. https://doi.org/10.3390/en14010158
14. Nwamekwe, C., Ewuzie, N., Igbokwe, N., U-Dominic, C. and Nwabueze, C. (2024) ‘Adoption of smart factories in Nigeria: Problems, obstacles, remedies and opportunities’, International Journal of Industrial and Production Engineering, 2(2), pp. 68 – 81. https://journals.unizik.edu.ng/ijipe/article/view/4167
15. Nwamekwe, C.O. and Chikwendu, O.C. (2025c) ‘Machine learning-augmented digital twin systems for predictive maintenance in highspeed rail networks’, International Journal of Multidisciplinary Research and Growth Evaluation, 6(01), pp. 1783-1795. https://hal.science/hal-04943345/
16. Nwamekwe, C.O., Chinwuko, C.E. and Mgbemena, C.E. (2020) ‘Development and implementation of a computerised production planning and control system’, UNIZIK Journal of Engineering and Applied Sciences, 17(1), pp. 168-187.
17. Nwamekwe, C.O., Ewuzie, N.V., Okpala, C.C., Ezeanyim, C., Nwabueze, C.V. and Nwabunwanne, E.C. (2025b) ‘Optimizing machine learning models for soil fertility analysis: insights from feature engineering and data localization’, Gazi University Journal of Science Part A: Engineering and Innovation, 12(1), pp. 36–60. https://dergipark.org.tr/en/pub/gujsa/issue/90827/1605587
18. Nwamekwe, C.O., Vitalis, E.N., Chidiebube, I.N. and Victoria, N.C. (2025a) ‘Evaluating Advances in machine learning algorithms for predicting and preventing maternal and foetal mortality in Nigerian healthcare: a systematic approach’, International Journal of Industrial and Production Engineering, 3(1), pp. 1-15. https://journals.unizik.edu.ng/ijipe/article/view/5161
19. Okpala, C., Chikwendu, U. and Onyeka, N.C. (2025) Artificial intelligence-driven total productive maintenance: The future of maintenance in smart factories’, International Journal of Engineering Research and Development, 21(1), pp. 68-74. https://hal.science/hal-05001680/
20. Onyeka, N.C., Chikwendu, O.C. and Chinwe, O.S. (2024b) ‘Machine learning-based prediction algorithms for the mitigation of maternal and fetal mortality in the Nigerian tertiary hospitals’, International Journal of Engineering Inventions, 13(7), pp. 132-138. https://hal.science/hal-04669409/
21. Onyeka, N.C., Vitalis, E.N., Chidiebube, I.N., Chikwendu, O.C. and U-Dominic, C.M. (2024a) ‘Sustainable manufacturing practices in Nigeria: optimization and implementation appraisal’, Journal of Research in Engineering and Applied Sciences, 9(3), pp. 769 – 774
22. PCS (2024) ‘Hybrid AR/SARIMA–ANN models for time series forecasting: benchmark study with Canadian lynx dataset’, Procedia Computer Science, 227, pp. 835–842. https://doi.org/10.1016/j.procs.2024.03.367
23. Prayuda, A. (2023) ‘Post consumed textile waste management and its impacts on the environment and economy in Bandung City’, IOP Conference Series: Earth and Environmental Science, 1257(1), 012009. https://doi.org/10.1088/1755-1315/1257/1/012009
24. Salman, D., Direkoglu, C. and Kusaf, M. (2024) ‘Hybrid deep learning models for time series forecasting of solar power’, Neural Comput. & Applic., 36, 9095–9112. https://doi.org/10.1007/s00521-024-09558-5
25. Sharma, V. and Patel, A. (2024) ‘A comprehensive review of standalone and hybrid models in financial time series forecasting’, Computer Modeling in Engineering & Sciences, 139(1), pp. 25–49. https://doi.org/10.32604/cmes.2024.055114
26. Vitalis, E.N., Nwamekwe, C.O., Chidiebube, I.N., Victoria, N.C., Nwabunwanne, E.C. and Ono, C.G. (2024) ‘Application of machine-learning-based hybrid algorithm for production forecast in textile company’, Jurnal Inovasi Teknologi dan Edukasi Teknik, 4(12), pp. 1-9  https://journal3.um.ac.id/index.php/ft/article/view/7140
27. Zhang, Z., Cheng, H. and Yu, Y. (2020) ‘Relationships among government funding, R&D model and innovation performance: a study on the Chinese textile industry’, Sustainability, 12(2), 644. https://doi.org/10.3390/su12020644