Predicting the flexural strength of 3D-printed geopolymer reinforced concrete using machine learning techniques

Both geopolymer concrete and 3D printing are innovative trends in construction materials science. This study investigates the prediction of 3D printed geopolymer reinforced concrete due to lack of information and studies on the prediction of 3D printed geopolymer reinforced concrete. This study investigated for the first time the flexural strength of 3D printed reinforced concrete through compressive strength with concrete mix design. Rigid, Lasso, elastic net, random forest, gradient boosting, decision tree, support vector machine regression and k-nearest neighbor are examined in this study. Considering to this study, compressive strength and flexural strength have more than 0.97 relationship. Moreover, the best result was for gradient boosting, random forest and k-nearest neighbor with 0.85 and 0.89.