Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

  • Saeed H. Moghtaderi
  • Prakash Thamburaja
  • Muhammad Alias Md Jedi
  • Michael Beer
  • Shahrum Abdullah
  • Ahmad Kamal Ariffin

Externe Organisationen

  • Universiti Kebangsaan Malaysia
  • Texas A and M University
  • The University of Liverpool
  • Tongji University
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)3151-3172
Seitenumfang22
FachzeitschriftJurnal Kejuruteraan
Jahrgang37
Ausgabenummer7
PublikationsstatusVeröffentlicht - 30 Okt. 2025

Abstract

The fracture mechanics of quasi-brittle materials, such as concrete, ceramics, and rocks, pose significant challenges due to their nonlinear stress-strain response, microstructural heterogeneity, and complex failure mechanisms. Traditional numerical and analytical methods often fall short in capturing the full intricacies of fracture propagation and damage evolution in such materials. However, recent advances in machine learning (ML) offer promising solutions to these limitations by enabling data-driven insights and enhanced computational performance. In this review paper, we explore the growing role of ML techniques in the fracture analysis of quasi-brittle materials. By leveraging large and diverse datasets from experiments and numerical simulations, ML models not only complement traditional fracture mechanics approaches but also introduce novel capabilities such as real-time damage prediction, adaptive modelling, and improved generalization across varying material conditions. The integration of data-driven models with physics-based frameworks, especially through hybrid techniques like physics-informed neural networks (PINNs), marks a significant shift in how fracture phenomena are modelled and understood. Key ML methods discussed include artificial neural networks (ANNs), convolutional neural networks (CNNs), and PINNs, with a focus on their respective advantages and implementation strategies. The review highlights how these approaches can enhance safety, efficiency, and predictive accuracy in engineering applications, ultimately making machine learning a transformative tool in the study of quasi-brittle fracture behaviour.

ASJC Scopus Sachgebiete

Zitieren

Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review. / Moghtaderi, Saeed H.; Thamburaja, Prakash; Jedi, Muhammad Alias Md et al.
in: Jurnal Kejuruteraan, Jahrgang 37, Nr. 7, 30.10.2025, S. 3151-3172.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Moghtaderi, SH, Thamburaja, P, Jedi, MAM, Beer, M, Abdullah, S & Ariffin, AK 2025, 'Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review', Jurnal Kejuruteraan, Jg. 37, Nr. 7, S. 3151-3172. https://doi.org/10.17576/jkukm-2025-37(7)-06
Moghtaderi, S. H., Thamburaja, P., Jedi, M. A. M., Beer, M., Abdullah, S., & Ariffin, A. K. (2025). Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review. Jurnal Kejuruteraan, 37(7), 3151-3172. https://doi.org/10.17576/jkukm-2025-37(7)-06
Moghtaderi SH, Thamburaja P, Jedi MAM, Beer M, Abdullah S, Ariffin AK. Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review. Jurnal Kejuruteraan. 2025 Okt 30;37(7):3151-3172. doi: 10.17576/jkukm-2025-37(7)-06
Moghtaderi, Saeed H. ; Thamburaja, Prakash ; Jedi, Muhammad Alias Md et al. / Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials : A Review. in: Jurnal Kejuruteraan. 2025 ; Jahrgang 37, Nr. 7. S. 3151-3172.
Download
@article{970fe40741fc4b95b83e4d31ffed54c2,
title = "Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials: A Review",
abstract = "The fracture mechanics of quasi-brittle materials, such as concrete, ceramics, and rocks, pose significant challenges due to their nonlinear stress-strain response, microstructural heterogeneity, and complex failure mechanisms. Traditional numerical and analytical methods often fall short in capturing the full intricacies of fracture propagation and damage evolution in such materials. However, recent advances in machine learning (ML) offer promising solutions to these limitations by enabling data-driven insights and enhanced computational performance. In this review paper, we explore the growing role of ML techniques in the fracture analysis of quasi-brittle materials. By leveraging large and diverse datasets from experiments and numerical simulations, ML models not only complement traditional fracture mechanics approaches but also introduce novel capabilities such as real-time damage prediction, adaptive modelling, and improved generalization across varying material conditions. The integration of data-driven models with physics-based frameworks, especially through hybrid techniques like physics-informed neural networks (PINNs), marks a significant shift in how fracture phenomena are modelled and understood. Key ML methods discussed include artificial neural networks (ANNs), convolutional neural networks (CNNs), and PINNs, with a focus on their respective advantages and implementation strategies. The review highlights how these approaches can enhance safety, efficiency, and predictive accuracy in engineering applications, ultimately making machine learning a transformative tool in the study of quasi-brittle fracture behaviour.",
keywords = "crack characterization, fracture mechanics, Machine learning, quasi-brittle materials",
author = "Moghtaderi, {Saeed H.} and Prakash Thamburaja and Jedi, {Muhammad Alias Md} and Michael Beer and Shahrum Abdullah and Ariffin, {Ahmad Kamal}",
note = "Publisher Copyright: {\textcopyright} 2025, National University of Malaysia. All rights reserved.",
year = "2025",
month = oct,
day = "30",
doi = "10.17576/jkukm-2025-37(7)-06",
language = "English",
volume = "37",
pages = "3151--3172",
number = "7",

}

Download

TY - JOUR

T1 - Recent Advances of Machine Learning in Fracture Mechanics of Quasi-Brittle Materials

T2 - A Review

AU - Moghtaderi, Saeed H.

AU - Thamburaja, Prakash

AU - Jedi, Muhammad Alias Md

AU - Beer, Michael

AU - Abdullah, Shahrum

AU - Ariffin, Ahmad Kamal

N1 - Publisher Copyright: © 2025, National University of Malaysia. All rights reserved.

PY - 2025/10/30

Y1 - 2025/10/30

N2 - The fracture mechanics of quasi-brittle materials, such as concrete, ceramics, and rocks, pose significant challenges due to their nonlinear stress-strain response, microstructural heterogeneity, and complex failure mechanisms. Traditional numerical and analytical methods often fall short in capturing the full intricacies of fracture propagation and damage evolution in such materials. However, recent advances in machine learning (ML) offer promising solutions to these limitations by enabling data-driven insights and enhanced computational performance. In this review paper, we explore the growing role of ML techniques in the fracture analysis of quasi-brittle materials. By leveraging large and diverse datasets from experiments and numerical simulations, ML models not only complement traditional fracture mechanics approaches but also introduce novel capabilities such as real-time damage prediction, adaptive modelling, and improved generalization across varying material conditions. The integration of data-driven models with physics-based frameworks, especially through hybrid techniques like physics-informed neural networks (PINNs), marks a significant shift in how fracture phenomena are modelled and understood. Key ML methods discussed include artificial neural networks (ANNs), convolutional neural networks (CNNs), and PINNs, with a focus on their respective advantages and implementation strategies. The review highlights how these approaches can enhance safety, efficiency, and predictive accuracy in engineering applications, ultimately making machine learning a transformative tool in the study of quasi-brittle fracture behaviour.

AB - The fracture mechanics of quasi-brittle materials, such as concrete, ceramics, and rocks, pose significant challenges due to their nonlinear stress-strain response, microstructural heterogeneity, and complex failure mechanisms. Traditional numerical and analytical methods often fall short in capturing the full intricacies of fracture propagation and damage evolution in such materials. However, recent advances in machine learning (ML) offer promising solutions to these limitations by enabling data-driven insights and enhanced computational performance. In this review paper, we explore the growing role of ML techniques in the fracture analysis of quasi-brittle materials. By leveraging large and diverse datasets from experiments and numerical simulations, ML models not only complement traditional fracture mechanics approaches but also introduce novel capabilities such as real-time damage prediction, adaptive modelling, and improved generalization across varying material conditions. The integration of data-driven models with physics-based frameworks, especially through hybrid techniques like physics-informed neural networks (PINNs), marks a significant shift in how fracture phenomena are modelled and understood. Key ML methods discussed include artificial neural networks (ANNs), convolutional neural networks (CNNs), and PINNs, with a focus on their respective advantages and implementation strategies. The review highlights how these approaches can enhance safety, efficiency, and predictive accuracy in engineering applications, ultimately making machine learning a transformative tool in the study of quasi-brittle fracture behaviour.

KW - crack characterization

KW - fracture mechanics

KW - Machine learning

KW - quasi-brittle materials

UR - http://www.scopus.com/inward/record.url?scp=105022194685&partnerID=8YFLogxK

U2 - 10.17576/jkukm-2025-37(7)-06

DO - 10.17576/jkukm-2025-37(7)-06

M3 - Article

AN - SCOPUS:105022194685

VL - 37

SP - 3151

EP - 3172

JO - Jurnal Kejuruteraan

JF - Jurnal Kejuruteraan

SN - 0128-0198

IS - 7

ER -

Von denselben Autoren