Machine learning approach for mechanical property assessment of industrial waste-filled epoxy–jute composites
6 affiliations
- 1Department of Mechanical Engineering, Galgotias University, Greater Noida, Uttar Pradesh, 203201, India
- 2Department of Mechanical Engineering, Saveetha Engineering College, Thandalam, Chennai, Tamil Nadu, 602105, India
- 3Department of Mechanical Engineering, Graphic Era (Deemed to be University), Dehradun, Uttarakhand, 248002, India. [email protected]
- 4School of Engineering and Technology, CGC University, Mohali, Punjab, 140307, India
- 5Department of Mechanical Engineering, J.J. College of Engineering and Technology, Tiruchirappalli, Tamil Nadu, 620009, India
- 6Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, India. [email protected]
Abstract
The growing demand for sustainable materials has stimulated the development of bio-based composites, yet the combination of natural fibers and industrial waste fillers in polymer matrices has not been exploited synergistically. In line with Sustainable Development Goal 12 (SDG-12), the paper focuses on reusing the steel industry by-product Linz–Donawitz (LD) sludge to enhance the mechanical properties of epoxy composites when used with the jute fiber. Six composite specimens with constant jute fiber loading (20 wt%) and a range of LD sludge content (0–25 wt%) were prepared using hand lay-up technique. The best composition (60 wt% epoxy, 20 wt% jute, 20 wt% LD sludge) resulted in tensile strength of 61.84 MPa (28.8% better than neat epoxy), flexural strength of 31.81 MPa (41.8% better) and impact strength of 18.026 kJ/m 2 . Interfacial defects and agglomeration of particles led to a decrease in mechanical properties beyound 20 wt% sludge. This experimental data was used to train four machine learning models to forecast mechanical properties given compositional inputs. On training data, XGBoost achieved R 2 = 1.0000 with near-zero errors (MAE = 0.0005 MPa, RMSE = 0.0008 MPa). However, when trained on a small dataset of six specimens, this perfect fit is mostly due to memorization of the training data, as opposed to predictive power. The findings suggest the risk of overfitting, mainly in the cases of Decision Tree and Gradient Boosting models. More realistic estimates of model performance are given by cross-validation (R 2 = 0.94 ± 0.04 in the case of XGBoost). The ML models can thus be used to analyze exploratory composition-property trend analysis in this particular composition space, as opposed to extrapolative prediction. These results both validate the possibility of hybrid composites that use industrial waste to obtain mechanical performance equivalent to standard natural fiber composites and indicate that waste can be valorized, although any assertion of ML predictive capacity should be carefully hedged due to limitations in the datasets.
Related
- Physics-Informed Deep Learning Surrogate for Mechanical-Response Prediction and Failure Assessment of High-Performance Polymer Composites
- Hydration Characteristics and Mechanical Property of Microfine Cement–Silica Fume Composites Modified with Cellulose Nanocrystals
- Investigation on the Degradation Mechanisms and Mechanical Property Evolution of 3D-SiO2-Fiber-Reinforced Polyimide Composites After Thermal Aging
- Mechanical performance of natural fiber reinforced polymer composites with critical assessment of recent advances and future perspectives
- Machine learning-aided design of metallic mechanical metamaterials: fabrication-aware modelling, property prediction, and inverse design
BibTeX
@article{purohit2026machine,
title = {{Machine learning approach for mechanical property assessment of industrial waste-filled epoxy–jute composites}},
author = {Abhilash Purohit and S. Sathees Kumar and Pravat Ranjan Pati and Arvind Kumar and Suresh Palanimuthu and Byomakesh Mahapatra},
journal = {Scientific Reports},
year = {2026},
month = may,
doi = {10.1038/s41598-026-53330-9},
url = {https://doi.org/10.1038/s41598-026-53330-9},
note = {CC BY},
}
RIS
TY - JOUR
TI - Machine learning approach for mechanical property assessment of industrial waste-filled epoxy–jute composites
AU - Abhilash Purohit
AU - S. Sathees Kumar
AU - Pravat Ranjan Pati
AU - Arvind Kumar
AU - Suresh Palanimuthu
AU - Byomakesh Mahapatra
PY - 2026
DA - 2026/05/18
JO - Scientific Reports
DO - 10.1038/s41598-026-53330-9
UR - https://doi.org/10.1038/s41598-026-53330-9
AB - The growing demand for sustainable materials has stimulated the development of bio-based composites, yet the combination of natural fibers and industrial waste fillers in polymer matrices has not been exploited synergistically. In line with Sustainable Development Goal 12 (SDG-12), the paper focuses on reusing the steel industry by-product Linz–Donawitz (LD) sludge to enhance the mechanical properties of epoxy composites when used with the jute fiber. Six composite specimens with constant jute fiber loading (20 wt%) and a range of LD sludge content (0–25 wt%) were prepared using hand lay-up technique. The best composition (60 wt% epoxy, 20 wt% jute, 20 wt% LD sludge) resulted in tensile strength of 61.84 MPa (28.8% better than neat epoxy), flexural strength of 31.81 MPa (41.8% better) and impact strength of 18.026 kJ/m 2 . Interfacial defects and agglomeration of particles led to a decrease in mechanical properties beyound 20 wt% sludge. This experimental data was used to train four machine learning models to forecast mechanical properties given compositional inputs. On training data, XGBoost achieved R 2 = 1.0000 with near-zero errors (MAE = 0.0005 MPa, RMSE = 0.0008 MPa). However, when trained on a small dataset of six specimens, this perfect fit is mostly due to memorization of the training data, as opposed to predictive power. The findings suggest the risk of overfitting, mainly in the cases of Decision Tree and Gradient Boosting models. More realistic estimates of model performance are given by cross-validation (R 2 = 0.94 ± 0.04 in the case of XGBoost). The ML models can thus be used to analyze exploratory composition-property trend analysis in this particular composition space, as opposed to extrapolative prediction. These results both validate the possibility of hybrid composites that use industrial waste to obtain mechanical performance equivalent to standard natural fiber composites and indicate that waste can be valorized, although any assertion of ML predictive capacity should be carefully hedged due to limitations in the datasets.
SN - 2045-2322
ER -