Development of hybrid SVM-FA, DT-FA and MLR-FA models to predict the flexural strength (FS) of recycled concrete

Recycled concrete from construction waste used as road material is a current sustainable approach. To provide feasible suggestions for civil engineers to prepare recycled with high flexural strength (FS) the pavement, present study proposed three hybrid machine learning models by combining support vector (SVM), decision tree (DT) and multiple linear regression (MLR) firefly algorithm (FA) computational optimization, named SVM-FA, DT-FA, MLR-FA, respectively. Effective water-cement ratio (WC), aggregate-cement (AC), aggregate replacement (RCA), nominal maximum size (NMR), normal (NMN), bulk density of (BDR), (BDN), water absorption RCA (WAR) NA (WAN) were employed input variables. determine predicting results varying models, root mean square error (RMSE) correlation coefficient (R) performance indexes. The showed that SVM-FA demonstrated highest R values lowest RMSE values, fitting effect predicted actual FS best. All above analysis proving SVM optimized FA hyperparameters has prediction accuracy can more accurate convenient tool evaluate concrete. sensitivity WC most significant influence on concrete, while weakest FS, which should be noticed when apply design in future.