Data-driven and explainable AI models for evaluating bond strength in reinforced concrete at elevated temperatures

Rwayda Kh. S. Al-Hamd; Asad S. Albostami; Holly Warren

doi:10.1016/j.firesaf.2025.104514

Data-driven and explainable AI models for evaluating bond strength in reinforced concrete at elevated temperatures

Rwayda Kh. S. Al-Hamd^*, Asad S. Albostami, Holly Warren

^*Corresponding author for this work

Department of Built Environment and Life Sciences

Research output: Contribution to journal › Article › peer-review

10 Downloads (Pure)

Abstract

The bond between steel and concrete in reinforced concrete (RC) and fibre-reinforced concrete (FRC) structures is a multifaceted and intricate phenomenon. It refers to the adhesion and mechanical interlock between the steel reinforcement bars and the surrounding concrete matrix. The bond becomes more complex at elevated temperatures; however, having an accurate estimate is a crucial factor in design. Therefore, this paper employs advanced machine learning (ML) techniques to predict bond strength (T_b) at both ambient and elevated temperatures from a 394-point experimental database, which includes additional variables such as fibre content, geometric ratios, and thermal parameter conditions. Seven models were built and assessed, including Linear Regression (LR), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), k-nearest Neighbours (KNN), Decision Tree (DT), and Deep Learning (DLearning) Regressors. The GB, XGBoost, and DT models offered the best prediction results with R² above 0.95 for the testing datasets, lowest error metrics (mean absolute error (MAE) between 0.8 and 1.1 MPa), and highest reliability (a30%-index ≥ 90%), all outperforming those reported in earlier literature. According to SHapley Additive exPlanations (SHAP) analysis, the length-to-diameter ratio (^l/_d ) and failure surface temperature (T) dominated as the predictors, followed by concrete compressive strength (f_c), and cover-to-diameter ratio (^c/_d), which is according to the existing mechanics of bond and thermal degradation. This study presents resolutions regarding the promise of data-driven models to accurately, reliably, and interpretably predict bond strength in post-fire conditions, which is of great merit in terms of resilient design practice. Future work may investigate hybrid ML–mechanistic frameworks and the integration of full-scale fire testing to further enhance engineering applicability.

Original language	English
Article number	104514
Number of pages	15
Journal	Fire Safety Journal
Volume	158
Early online date	25 Aug 2025
DOIs	https://doi.org/10.1016/j.firesaf.2025.104514
Publication status	E-pub ahead of print - 25 Aug 2025

Keywords

Concrete-steel bond at elevated temperatures
Explainable Artificial Intelligence (AI)
Machine learning (ML)
Fibre-reinforced concrete (FRC)
SHapley Additive exPlanations (SHAP) analysis

Access to Document

10.1016/j.firesaf.2025.104514Licence: CC BY

Al-Hamd_Data-DrivenAndExplainableAIModels_Published_2025Final published version, 7.3 MBLicence: CC BY

Cite this

@article{44d16c51019d4a7e9f9764cb6a3bcc17,

title = "Data-driven and explainable AI models for evaluating bond strength in reinforced concrete at elevated temperatures",

abstract = "The bond between steel and concrete in reinforced concrete (RC) and fibre-reinforced concrete (FRC) structures is a multifaceted and intricate phenomenon. It refers to the adhesion and mechanical interlock between the steel reinforcement bars and the surrounding concrete matrix. The bond becomes more complex at elevated temperatures; however, having an accurate estimate is a crucial factor in design. Therefore, this paper employs advanced machine learning (ML) techniques to predict bond strength (Tb) at both ambient and elevated temperatures from a 394-point experimental database, which includes additional variables such as fibre content, geometric ratios, and thermal parameter conditions. Seven models were built and assessed, including Linear Regression (LR), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), k-nearest Neighbours (KNN), Decision Tree (DT), and Deep Learning (DLearning) Regressors. The GB, XGBoost, and DT models offered the best prediction results with R² above 0.95 for the testing datasets, lowest error metrics (mean absolute error (MAE) between 0.8 and 1.1 MPa), and highest reliability (a30\%-index ≥ 90\%), all outperforming those reported in earlier literature. According to SHapley Additive exPlanations (SHAP) analysis, the length-to-diameter ratio (l/d ) and failure surface temperature (T) dominated as the predictors, followed by concrete compressive strength (fc), and cover-to-diameter ratio (c/d), which is according to the existing mechanics of bond and thermal degradation. This study presents resolutions regarding the promise of data-driven models to accurately, reliably, and interpretably predict bond strength in post-fire conditions, which is of great merit in terms of resilient design practice. Future work may investigate hybrid ML–mechanistic frameworks and the integration of full-scale fire testing to further enhance engineering applicability.",

author = "Al-Hamd, \{Rwayda Kh. S.\} and Albostami, \{Asad S.\} and Holly Warren",

note = "{\textcopyright} 2025 The Authors. Published by Elsevier Ltd. This is an open access article distributed under the terms of the Creative Commons CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Data availability statement: Data will be made available on request.",

year = "2025",

month = aug,

day = "25",

doi = "10.1016/j.firesaf.2025.104514",

language = "English",

volume = "158",

journal = "Fire Safety Journal",

issn = "0379-7112",

publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - Data-driven and explainable AI models for evaluating bond strength in reinforced concrete at elevated temperatures

AU - Al-Hamd, Rwayda Kh. S.

AU - Albostami, Asad S.

AU - Warren, Holly

N1 - © 2025 The Authors. Published by Elsevier Ltd. This is an open access article distributed under the terms of the Creative Commons CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Data availability statement: Data will be made available on request.

PY - 2025/8/25

Y1 - 2025/8/25

N2 - The bond between steel and concrete in reinforced concrete (RC) and fibre-reinforced concrete (FRC) structures is a multifaceted and intricate phenomenon. It refers to the adhesion and mechanical interlock between the steel reinforcement bars and the surrounding concrete matrix. The bond becomes more complex at elevated temperatures; however, having an accurate estimate is a crucial factor in design. Therefore, this paper employs advanced machine learning (ML) techniques to predict bond strength (Tb) at both ambient and elevated temperatures from a 394-point experimental database, which includes additional variables such as fibre content, geometric ratios, and thermal parameter conditions. Seven models were built and assessed, including Linear Regression (LR), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), k-nearest Neighbours (KNN), Decision Tree (DT), and Deep Learning (DLearning) Regressors. The GB, XGBoost, and DT models offered the best prediction results with R² above 0.95 for the testing datasets, lowest error metrics (mean absolute error (MAE) between 0.8 and 1.1 MPa), and highest reliability (a30%-index ≥ 90%), all outperforming those reported in earlier literature. According to SHapley Additive exPlanations (SHAP) analysis, the length-to-diameter ratio (l/d ) and failure surface temperature (T) dominated as the predictors, followed by concrete compressive strength (fc), and cover-to-diameter ratio (c/d), which is according to the existing mechanics of bond and thermal degradation. This study presents resolutions regarding the promise of data-driven models to accurately, reliably, and interpretably predict bond strength in post-fire conditions, which is of great merit in terms of resilient design practice. Future work may investigate hybrid ML–mechanistic frameworks and the integration of full-scale fire testing to further enhance engineering applicability.

AB - The bond between steel and concrete in reinforced concrete (RC) and fibre-reinforced concrete (FRC) structures is a multifaceted and intricate phenomenon. It refers to the adhesion and mechanical interlock between the steel reinforcement bars and the surrounding concrete matrix. The bond becomes more complex at elevated temperatures; however, having an accurate estimate is a crucial factor in design. Therefore, this paper employs advanced machine learning (ML) techniques to predict bond strength (Tb) at both ambient and elevated temperatures from a 394-point experimental database, which includes additional variables such as fibre content, geometric ratios, and thermal parameter conditions. Seven models were built and assessed, including Linear Regression (LR), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), k-nearest Neighbours (KNN), Decision Tree (DT), and Deep Learning (DLearning) Regressors. The GB, XGBoost, and DT models offered the best prediction results with R² above 0.95 for the testing datasets, lowest error metrics (mean absolute error (MAE) between 0.8 and 1.1 MPa), and highest reliability (a30%-index ≥ 90%), all outperforming those reported in earlier literature. According to SHapley Additive exPlanations (SHAP) analysis, the length-to-diameter ratio (l/d ) and failure surface temperature (T) dominated as the predictors, followed by concrete compressive strength (fc), and cover-to-diameter ratio (c/d), which is according to the existing mechanics of bond and thermal degradation. This study presents resolutions regarding the promise of data-driven models to accurately, reliably, and interpretably predict bond strength in post-fire conditions, which is of great merit in terms of resilient design practice. Future work may investigate hybrid ML–mechanistic frameworks and the integration of full-scale fire testing to further enhance engineering applicability.

U2 - 10.1016/j.firesaf.2025.104514

DO - 10.1016/j.firesaf.2025.104514

M3 - Article

SN - 0379-7112

VL - 158

JO - Fire Safety Journal

JF - Fire Safety Journal

M1 - 104514

ER -

Data-driven and explainable AI models for evaluating bond strength in reinforced concrete at elevated temperatures

Abstract

Keywords

Access to Document

Fingerprint

Cite this