A refreshing view of soft computing models for predicting the deflection of reinforced concrete beams

Abstract

The efforts of this study are to address an essential technical issue in construction and civil engineering, namely predicting the deflection of reinforced concrete beams. Indeed, six new hybrid models (ensemble models) were developed to address this critical technical problem based on artificial intelligence models as well as machine learning algorithms, such as artificial neural network (ANN), support vector machine (SVM), and adaptive neuro-fuzzy inference system (ANFIS). Accordingly, the bagging (BA) technique was applied to create new ensemble models, including BA-SVM, BA-ANN, BA-ANFIS, SVM-ANN, SVM-ANFIS, and ANN-ANFIS models. They were developed based on 120 practical experiments on the deflection of reinforced concrete beams. A series of indicators of error, accuracy, as well as the statistical significance of the models, were analyzed to assess the overall efficiency of the forecasting models. The results showed that the ensemble models are capable of predicting the deflection of reinforced concrete beams with high accuracy, especially the SVM-ANFIS model. The results of this study have opened up many new research directions in the design and optimization of the structure of buildings, dangerous warning systems, and timely solutions to ensure the safety of buildings.