Open Access

REVIEW

A Review on Machine Learning and Digital Twin Enabled Advancements in Biofiller/Fibre-Based Polymer Composites for Sustainable Engineering Applications

Rahul Kumar¹, Faladrum Sharma², Pradeep Kumar Karsh^3,*

1 Department of Mechanical Engineering, School of Engineering (SoE), Dayananda Sagar University, Devarakaggalahalli, Harohalli, Kanakapura Road, South Bengaluru Dt., Bengaluru, India
2 Centre for Indian Knowledge Systems, Indian Institute of Technology Guwahati, Guwahati, Assam, India
3 Department of Mechanical Engineering, Parul Institute of Engineering & Technology, Parul University, Vadodara, India

* Corresponding Author: Pradeep Kumar Karsh. Email:

(This article belongs to the Special Issue: Modeling Strategy and “Material-Structure-Function” Integrated Design for Composite Components)

Computer Modeling in Engineering & Sciences 2026, 148(2), 3 https://doi.org/10.32604/cmes.2026.086198

Received 26 May 2026; Accepted 05 August 2026; Issue published 28 August 2026

Abstract

The increasing environmental concerns after the United Nations’ push towards sustainable development goals (SDGs) and depletion of non-renewable resources have accelerated the global pursuit of sustainable materials. Within this framework, bio-based polymer composites have gained considerable attention for their ability to balance mechanical performance, cost-effectiveness, and environmental responsibility. Biofibres/fillers-based polymer composites reinforced with natural fibres like jute, bamboo, coconut coir, pineapple leaf fibre (PALF), and flax offer an attractive combination of mechanical performance, cost-effectiveness, and environmental sustainability. Moreover, additive manufacturing (3D printing), coupled with machine learning, digital twins, and data-driven material design, is transforming the development and optimization of natural fibre-reinforced polymer composites for advanced, sustainable engineering applications and enabled the realization of the material, structure and function paradigm for composite design. This review highlights major developments in bio-based composites through the perspective of integrated modelling strategy and material-design-function philosophy. It also outlines how machine learning and digital twin-based models accelerated the material discovery, prediction, optimization, and sustainability assessment. Using machine learning techniques, prediction of properties, optimization of processes, and detection of defects can be performed in a much faster way. Digital twins provide a virtual representation of components, enabling real-time monitoring, predictive maintenance, virtual testing, and full lifecycle analysis. Moreover, the purpose of this review is to organize the latest progress, highlight the main research gaps, and propose the next steps towards smart, eco-friendly, and Industry 4.0/5.0, compliant composite materials. Also, the integration of life cycle assessment data into digital twin platforms could enable continuous environmental performance assessment throughout the product lifecycle.

---

Keywords

---