Interpretable Predictive Modelling of Basalt Fiber Reinforced Concrete Splitting Tensile Strength Using Ensemble Machine Learning Methods and SHAP Approach

Çakıroğlu, Celal; Aydın, Yaren; Bekdaş, Gebrai̇l; Geem, Zong Woo

doi:10.3390/ma16134578

Cited by 16 publications

(7 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast, XGBoost and LightGBM models demonstrated strong predictive performance, achieving test set R 2 values of 0.93 and 0.94, respectively. This aligns with the findings of [85], who reported R 2 values of 0. (e) (f) (g) In Table 4, the values comparing the performance of the seven optimal models are presented, and Figure 5 illustrates the performance of the models based on Table 4.…”

Section: Final Model Selectionsupporting

confidence: 92%

“…In contrast, XGBoost and LightGBM models demonstrated strong predictive performance, achieving test set R 2 values of 0.93 and 0.94, respectively. This aligns with the findings of[85], who reported R 2 values of 0.93 for XGBoost and 0.94 for LightGBM. Secondly, looking at the RMSE and MAE values on the test set, the linear regression model…”

supporting

confidence: 92%

“…Previous studies have also demonstrated similar findings. For instance, Cakiroglu et al [85] used machine learning to study fiber-reinforced concrete and found that both model data samples remained within the ±10% deviation lines, while a study by Abdulalim Alabdullah et al [71] on high-strength concrete prediction using LightGBM and XGBoost found a strong correlation between experimental and predicted results for both models. Figure 4 compares the actual values and predicted values from the test dataset of seven optimal models.…”

Section: Final Model Selectionmentioning

confidence: 99%

See 2 more Smart Citations

A Meta-Analysis of the Effect of Moisture Content of Recycled Concrete Aggregate on the Compressive Strength of Concrete

Cho,

Brand

2024

Applied Sciences

View full text Add to dashboard Cite

To reduce the environmental impact of concrete, recycled aggregates are of significant interest. Recycled concrete aggregate (RCA) presents a significant resource opportunity, although its performance as an aggregate in concrete is variable. This study presents a meta-analysis of the published literature to refine the understanding of how the moisture content of RCA, as well as other parameters, affects the compressive strength of concrete. Seven machine learning models were used to predict the compressive strength of concrete with RCA, including linear regression, support vector regression (SVR), and k-nearest neighbors (KNN) as single models, and decision tree, random forest, XGBoost, and LightGBM as ensemble models. The results of this study demonstrate that ensemble models, particularly the LightGBM model, exhibited superior prediction accuracy compared to single models. The LightGBM model yielded the highest prediction accuracy with R2 = 0.94, RMSE = 4.16 MPa, MAE = 3.03 MPa, and Delta RMSE = 1.4 MPa, making it the selected final model. The study, employing feature importance with LightGBM as the final model, identified age, water/cement ratio, and fine RCA aggregate content as key factors influencing compressive strength in concrete with RCA. In an interaction plot analysis using the final model, lowering the water–cement ratio consistently improved compressive strength, especially between 0.3 and 0.4, while increasing the fine RCA ratio decreased compressive strength, particularly in the range of 0.4 to 0.6. Additionally, it was found that maintaining moisture conditions of RCA typically between 0.0 and 0.8 was crucial for maximizing strength, whereas extreme moisture conditions, like fully saturated surface dry (SSD) state, negatively impacted strength.

show abstract

Section: Final Model Selectionsupporting

confidence: 92%

“…In contrast, XGBoost and LightGBM models demonstrated strong predictive performance, achieving test set R 2 values of 0.93 and 0.94, respectively. This aligns with the findings of[85], who reported R 2 values of 0.93 for XGBoost and 0.94 for LightGBM. Secondly, looking at the RMSE and MAE values on the test set, the linear regression model…”

supporting

confidence: 92%

Section: Final Model Selectionmentioning

confidence: 99%

See 1 more Smart Citation

A Meta-Analysis of the Effect of Moisture Content of Recycled Concrete Aggregate on the Compressive Strength of Concrete

Cho,

Brand

2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The results show that the ensemble learning model is much better than the standalone model [24]. Cakiroglu et al used Extreme Gradient Boosting, Light Gradient Boosting Machine, Random Forest, and Categorical Boosting to predict the splitting tensile strength of concrete reinforced with basalt fibers [25]. Feng et al predicted the creep behavior of recycled aggregate concrete using ensemble learning combined with SHAP and performed feature importance analysis [26].…”

Section: Introductionmentioning

confidence: 99%

Accurate Prediction of Punching Shear Strength of Steel Fiber-Reinforced Concrete Slabs: A Machine Learning Approach with Data Augmentation and Explainability

Cheng,

Taffese,

2024

Buildings

View full text Add to dashboard Cite

Reinforced concrete slabs are widely used in building structures due to their economic, durable, and aesthetic advantages. The determination of their ultimate strength often hinges on punching shear strength. Presently, methods such as closed hoops, steel bending, and fiber reinforcement are employed to enhance punching shear strength, with fiber reinforcement gaining popularity due to its ease of implementation and efficacy in improving concrete durability. This study introduces a novel approach employing six machine learning algorithms rooted in decision trees and decision tree-based ensemble learning to predict punching shear strength in steel fiber-reinforced concrete slabs. To overcome experimental data limitations, a data augmentation approach based on the Gaussian mixture model is employed. The validation of the data augmentation is conducted through “synthetic training—real testing” and “real training—real testing”. Additionally, the best machine learning model is analyzed for explainability using Shapley Additive exPlanation (SHAP). Results demonstrate that the proposed data augmentation method effectively captures the original data distribution, enhancing the robustness and accuracy of the machine learning model. Moreover, SHAP provides better insights into the features influencing punching shear strength. Thus, the proposed data enhancement model offers a reliable approach for modeling small experimental datasets in structural engineering.

show abstract

“…Researchers have made many efforts to minimize this error. Some researchers use machine learning methods to predict the splitting strength [14][15][16][17]. Van Cauwelaert [18] provided an approximate analytical solution for the splitting tensile strength of rectangular test blocks by modifying the equation using a two-dimensional planar body stress function of infinite length in the horizontal direction.…”

Section: Introductionmentioning

confidence: 99%

Theoretical Investigation on Indirect Tensile Strength of Concrete with Rectangular Cross-Section under Locally Distributed Load

Wang,

Xu,

Sun

et al. 2023

Materials

View full text Add to dashboard Cite

The indirect tensile test plays a crucial role in experimental investigations of brittle material properties. In this study, a mechanical analysis model of the rectangular test block is established based on the theory of elastic mechanics for the characteristics of the indirect tensile test. The theoretical solution of the triangular series is derived for the rectangular test block under the locally distributed load. The finite element simulation results and splitting test results were compared with the theoretical results. The results of the study verify the accuracy of the theoretical solutions. Based on the proposed analytical solution, the effects of loading width and length-to-height ratio (h/l) of local loading on the measured tensile strength of test block are discussed. The results demonstrate that the tensile strength of the test block increases as the loading width expands, and the rate of growth in the recorded tensile strength gradually stabilizes. The variation in loading width affects the location of crack initiation points during the concrete test block splitting tests. When the loading width exceeds 6% of the side length of test block, the cracking point is positioned at the center of test block, ensuring the effectiveness of the splitting test. As the length-to-height ratio of the test block increases, there is a general upward trend in the measured tensile strength. When h/l < 0.6, the measured tensile strength initially increases before decreasing. However, when h/l > 0.6, the measured tensile strength consistently increases, with the rate of increase gradually diminishing until it stabilizes. The length-to-height ratio also significantly influences the location of the cracking point in the test block. As the length-to-height ratio increases, the cracking point initially shifts from around the center to the central point and then further from the center toward the edge. To ensure that the location of the crack initiation point is in the center of the specimen and that the tensile strength is close to the measured result, the length to height ratio can be chosen at around 0.85.

show abstract

Interpretable Predictive Modelling of Basalt Fiber Reinforced Concrete Splitting Tensile Strength Using Ensemble Machine Learning Methods and SHAP Approach

Cited by 16 publications

References 69 publications

A Meta-Analysis of the Effect of Moisture Content of Recycled Concrete Aggregate on the Compressive Strength of Concrete

A Meta-Analysis of the Effect of Moisture Content of Recycled Concrete Aggregate on the Compressive Strength of Concrete

Accurate Prediction of Punching Shear Strength of Steel Fiber-Reinforced Concrete Slabs: A Machine Learning Approach with Data Augmentation and Explainability

Theoretical Investigation on Indirect Tensile Strength of Concrete with Rectangular Cross-Section under Locally Distributed Load

Contact Info

Product

Resources

About