Comparison of Multilayer Perceptron and Other Methods for Prediction of Sustainable Optimum Design of Reinforced Concrete Columns

Abstract

Machine learning has become a popular science in recent years, as it produces concrete and fast solutions to solve many problems. The rapidly increasing world population and the developments in technology have caused large greenhouse gas emissions. The harmful effects of carbon dioxide emissions from cement in concrete on climate change and global warming are quite remarkable. In this study, the most commonly used machine learning (ML) models in the literature were used for CO2 minimization of reinforced concrete columns. Harmony search was employed to find the optimum dataset for machine learning. The performances of these algorithms were compared and the best algorithm was tried to be found. As a result of all, it is observed that Multilayer Perceptron (MLP) has higher performance than other algorithms. The R2 of the MLP is 0.999. According to this result, it was observed that MLP is the most successful ML model in the design of eco-friendly reinforced concrete columns. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.