Model validation and sensitivity analysis of regression-based downtime predictive systems
DOI:
https://doi.org/10.20961/3398nb16Keywords:
Downtime, Model validation, Parameters, Plastic manufacturing, Predictive model, Sensitivity analysisAbstract
Reliable predictive models are essential for optimizing maintenance strategies in manufacturing environments; however, their adoption is often hindered by inadequate model validation and limited understanding of parameter influence on prediction robustness. This study presents an in-depth validation and sensitivity evaluation of previously developed regression-based downtime predictive models for plastic manufacturing production lines. The validation framework integrates statistical performance metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), coefficient of determination (R²), Pearson correlation, and hypothesis testing using the Analysis of Variance (ANOVA) F-test. The results confirm strong predictive accuracy, high correlation between predicted and experimental downtimes, and statistically significant model structures at 5% significance level. Furthermore, a One-At-A-Time (OAT) sensitivity analysis was conducted to quantify the relative influence of key process parameters (uptime, product weight, cycle time, and quantity) on predicted downtime. Findings reveal that the developed regression models exhibited strong predictive performance, for the Cup model, MAE = 1.46min, RMSE = 1.92min, R² = 0.976, for the Plate model, MAE = 1.83min, RMSE = 2.40min, R² = 0.987, plus the ANOVA results for both models, indicates a high level of reliability. Sensitivity analysis revealed that uptime and cycle time were the most influential variables affecting downtime behavior, while quantity and weight demonstrate comparatively lower contributions. The combined validation and sensitivity framework establish model reliability, demonstrates robustness under parametric variation, and provides deeper insight into parameter significance, thereby supporting confident deployment of the models for predictive maintenance decision-making in industrial environments.
References
Akintona, E. O., Fredrick, O., & Augustine, N. M. (2021). Development of optimal design and circular broach modular design of disc cutters with the development of a parametric model. Journal of Engineering Research and Reports, 20(10), 15–21.
Backhaus, K., Erichson, B., Gensler, S., Weiber, R., & Weiber, T. (2025). Analysis of variance. In Multivariate analysis: An application-oriented introduction (pp. 149–205). Springer Fachmedien Wiesbaden.
Berry, M. W., Mohamed, A., & Yap, B. W. (2020). Supervised and unsupervised learning for data science. Springer. https://doi.org/10.1007/978-3-030-22475-2
Chigurupati, A., Thibaux, R., & Lassar, N. (2016). Predicting hardware failure using machine learning. In 2016 Annual Reliability and Maintainability Symposium (RAMS) (pp. 1–6). IEEE. https://doi.org/10.1109/RAMS.2016.7448045
Çınar, Z. M., Nuhu, A. A., Zeeshan, Q., Korhan, O., Asmael, M., & Safaei, B. (2020). Machine learning in predictive maintenance towards sustainable smart manufacturing in Industry 4.0. Sustainability, 12(19), 8211. https://doi.org/10.3390/su12198211
Dalzochio, J., Kunst, R., Pignaton, E., Binotto, A., Sanyal, S., Favilla, J., & Barbosa, J. (2020). Machine learning and reasoning for predictive maintenance in Industry 4.0: Current status and challenges. Computers in Industry, 123, 103298. https://doi.org/10.1016/j.compind.2020.103298
Igbokwe, N. C., Okeagu, F. N., Onyeka, N. C., Onwuliri, J. B., & Godfrey, O. C. (2024). Machine learning-driven maintenance cost optimization: Insights from a local industrial compressor case study. Jurnal Inovasi Teknologi dan Edukasi Teknik, 4(11), 1–10.
Kane, A., Kore, A., Khandale, A., Nigade, S., & Joshi, P. P. (2022). Predictive maintenance using machine learning (arXiv:2205.09402). arXiv. https://doi.org/10.48550/arXiv.2205.09402
Karkouch, A., Mousannif, H., Al Moatassime, H., & Noel, T. (2016). Data quality in Internet of Things: A state-of-the-art survey. Journal of Network and Computer Applications, 73, 57–81. https://doi.org/10.1016/j.jnca.2016.08.001
King, N. J., & Raja, V. T. (2012). Protecting the privacy and security of sensitive customer data in the cloud. Computer Law & Security Review, 28(3), 308–319. https://doi.org/10.1016/j.clsr.2012.03.003
Koković, V., Pavlović, K., Mijanović, A., Kovačević, S., Mačužić, I., & Božović, V. (2024). Exploring statistical and machine learning methods for modeling probability distribution parameters in downtime length analysis: A paper manufacturing machine case study. Journal of Big Data, 11(1), 162. https://doi.org/10.1186/s40537-024-00916-8
Maktoubian, J., Taskhiri, M. S., & Turner, P. (2021). Intelligent predictive maintenance (IPdM) in forestry: A review of challenges and opportunities. Forests, 12(11), 1495. https://doi.org/10.3390/f12111495
Mgbemena, C. E., & Okeagu, F. N. (2023). Development of an IoT-based real-time remote monitoring device for the maintenance of injection moulding machines in plastic industries. UNIZIK Journal of Engineering and Applied Sciences, 2(1), 260–278.
Mohan, T. R., Roselyn, J. P., Uthra, R. A., Devaraj, D., & Umachandran, K. (2021). Intelligent machine learning–based total productive maintenance approach for achieving zero downtime in industrial machinery. Computers & Industrial Engineering, 157, 107267. https://doi.org/10.1016/j.cie.2021.107267
Nnaemeka, O. F., Oluwadamilare, A. E., & Blessing, N. (2022). Impact of stakeholders on the success of project cost management in construction industry. International Journal of Latest Technology in Engineering, Management & Applied Science, 11(3), 1–7.
Okeagu, F. N., & Mgbemena, C. E. (2022). A systematic review of digital twin systems for improved predictive maintenance of equipment in smart factories. International Journal of Industrial and Production Engineering, 1(1), 1–20.
Onuoha, D. O., Mgbemena, C. E., Godwin, H. C., & Okeagu, F. N. (2022). Application of Industry 4.0 technologies for effective remote monitoring of cathodic protection systems of oil and gas pipelines: A systematic review. International Journal of Industrial and Production Engineering, 1(2), 1–15.
Peng, J., Manevski, K., Kørup, K., Larsen, R., & Andersen, M. N. (2021). Random forest regression results in accurate assessment of potato nitrogen status based on multispectral data from different platforms and the critical concentration approach. Field Crops Research, 268, 108158. https://doi.org/10.1016/j.fcr.2021.108158
Sanzana, M. R., Abdulrazic, M. O. M., Wong, J. Y., Maul, T., & Yip, C.-C. (2022). Analysis of machine learning techniques for predictive maintenance in cooler condition. In 2022 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) (pp. 1–4). IEEE. https://doi.org/10.1109/ISPACS57703.2022.10082814
Uzochukwukanma, M. C., Okeagu, F. O., & Orji, R. U. (2023). Design and optimization of 0.5-ton/hr foundry cupola furnace operations. Iconic Research and Engineering Journals, 9(2), 1–8.
Van Oudenhoven, B., Demerouti, E., Basten, R., & Van de Calseyde, P. (2025). Preparing for predictive maintenance: Employee perspectives on job demands and resources before and after its implementation. Applied Ergonomics, 129, 104561. https://doi.org/10.1016/j.apergo.2024.104561
Xia, C. S., & Zhang, L. (2022). Less training, more repairing please: Revisiting automated program repair via zero-shot learning. In Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE 2022) (pp. 959–971). Association for Computing Machinery. https://doi.org/10.1145/3540250.3549157
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Innovations in Science Education and Practice
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
