A crucial factor in the efficient design of concrete sustainable buildings is the compressive strength (Cs) of eco-friendly concrete. In this work, a hybrid model of Gradient Boosting Regression Tree (GBRT) with grid search cross-validation (GridSearchCV) optimization technique was used to predict the compressive strength, which allowed us to increase the precision of the prediction models. In addition, to build the proposed models, 164 experiments on eco-friendly concrete compressive strength were gathered for previous researches. The dataset included the water/binder ratio (W/B), curing time (age), the recycled aggregate percentage from the total aggregate in the mixture (RA%), ground granulated blast-furnace slag (GGBFS) material percentage from the total binder used in the mixture (GGBFS%), and superplasticizer (kg). The root mean square error (RMSE) and coefficient of determination (R²) between the observed and forecast strengths were used to evaluate the accuracy of the predictive models. The obtained results indicated that—when compared to the default GBRT model—the GridSearchCV approach can capture more hyperparameters for the GBRT prediction model. Furthermore, the robustness and generalization of the GSC-GBRT model produced notable results, with RMSE and R² values (for the testing phase) of 2.3214 and 0.9612, respectively. The outcomes proved that the suggested GSC-GBRT model is advantageous. Additionally, the significance and contribution of the input factors that affect the compressive strength were explained using the Shapley additive explanation (SHAP) approach.

Unconfined compressive strength (UCS) can be used to assess the applicability of geopolymer binders as ecologically friendly materials for geotechnical projects. Furthermore, soft computing technologies are necessary since experimental research is often challenging, expensive, and time-consuming. This article discusses the feasibility and the performance required to predict UCS using a Random Forest (RF) algorithm. The alkali activator studied was sodium hydroxide solution, and the considered geopolymer source material was ground-granulated blast-furnace slag and fly ash. A database with 283 clayey soil samples stabilized with geopolymer was considered to determine the UCS. The database was split into two sections for the development of the RF model: the training data set (80%) and the testing data set (20%). Several measures, including coefficient of determination (R), mean absolute error (MAE), and root mean square error (RMSE), were used to assess the effectiveness of the RF model. The statistical findings of this study demonstrated that the RF is a reliable model for predicting the UCS value of geopolymer-stabilized clayey soil. Furthermore, based on the obtained values of RMSE = 0.9815 and R² = 0.9757 for the testing set, respectively, the RF approach showed to provide excellent results for predicting unknown data within the ranges of examined parameters. Finally, the SHapley Additive exPlanations (SHAP) analysis was implemented to identify the most influential inputs and to quantify their behavior of input variables on the UCS.

The construction industry has witnessed a substantial increase in the demand for eco-friendly and sustainable materials. Eco-friendly concrete containing Ground Granulated Blast Furnace Slag (GGBFS) and Recycled Coarse Aggregate (RCA) is such a material, which can contribute to a reduction in waste and promote environmental sustainability. Compressive strength is a crucial parameter in evaluating the performance of concrete. However, predicting the compressive strength of concrete containing GGBFS and RCA can be challenging. This study presents a novel XGBoost (eXtreme Gradient Boosting) prediction model for the compressive strength of eco-friendly concrete containing GGBFS and RCA, optimized using Bayesian optimization (BO). The model was trained on a comprehensive dataset consisting of several mix design parameters. The performance of the optimized XGBoost model was assessed using multiple evaluation metrics, including Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R²). These metrics were calculated for both training and testing datasets to evaluate the model’s accuracy and generalization capabilities. The results demonstrated that the optimized XGBoost model outperformed other state-of-the-art machine learning models, such as Support Vector Regression (SVR), and K-nearest neighbors algorithm (KNN), in predicting the compressive strength of eco-friendly concrete containing GGBFS and RCA. An analysis using Partial Dependence Plots (PDP) was carried out to discern the influence of distinct input features on the compressive strength prediction. This PDP analysis highlighted the water-to-binder ratio, the age of the concrete, and the percentage of GGBFS used, as significant factors impacting the compressive strength of the eco-friendly concrete.

Concrete is the most widely used building material, but it is also a recognized pollutant, causing significant issues for sustainability in terms of resource depletion, energy use, and greenhouse gas emissions. As a result, efforts should be concentrated on reducing concrete’s environmental consequences in order to increase its long-term viability. In order to design environmentally friendly concrete mixtures, this research intended to create a prediction model for the compressive strength of those mixtures. The concrete mixtures that were used in this study to build our proposed prediction model are concrete mixtures that contain both recycled aggregate concrete (RAC) and ground granulated blast-furnace slag (GGBFS). A white-box machine learning model known as multivariate polynomial regression (MPR) was developed to predict the compressive strength of eco-friendly concrete. The model was compared with the other two machine learning models, where one is also a white-box machine learning model, namely linear regression (LR), and the other is the black-box machine learning model, which is a support vector machine (SVM). The newly suggested model shows robust estimation capabilities and outperforms the other two models in terms of R² (coefficient of determination) and RMSE (root mean absolute error) measurements.

For the design or assessment of concrete structures that incorporate steel fiber in their elements, the accurate prediction of the shear strength of steel fiber reinforced concrete (SFRC) beams is critical. Unfortunately, traditional empirical methods are based on a small and limited dataset, and their abilities to accurately estimate the shear strength of SFRC beams are arguable. This drawback can be reduced by developing an accurate machine learning based model. The problem with using a high accuracy machine learning (ML) model is its interpretation since it works as a black-box model that is highly sophisticated for humans to comprehend directly. For this reason, Shapley additive explanations (SHAP), one of the methods used to open a black-box machine learning model, is combined with highly accurate machine learning techniques to build an explainable ML model to predict the shear strength of SFRC slender beams. For this, a database of 330 beams with varying design attributes and geometries was developed. The new gradient boosting regression tree (GBRT) machine learning model was compared statistically to experimental data and current shear design models to evaluate its performance. The proposed GBRT model gives predictions that are very similar to the experimentally observed shear strength and has a better and unbiased predictive performance in comparison to other existing developed models. The SHAP approach shows that the beam width and effective depth are the most important factors, followed by the concrete strength and the longitudinal reinforcement ratio. In addition, the outputs are also affected by the steel fiber factor and the shear-span to effective depth ratio. The fiber tensile strength and the aggregate size have the lowest effect, with only about 1% on average to change the predicted value of the shear strength. By building an accurate ML model and by opening its black-box, future researchers can focus on some attributes rather than others.