Prediction of Minimum CO2 Emission for Rectangular Shape Reinforced Concrete (RC) Beam

Abstract

Machine learning methodologies help to directly determine the desired design parameters of any structural design problem. In this regard, with the present study, a simply supported reinforced concrete (RC) beam with rectangular cross-section was handled for the prediction of the design outcomes as minimum carbon dioxide (CO2) emission and optimal cost level of structural materials. In this respect, firstly, an optimization process was performed with a well-known metaheuristic algorithm, and then a prediction application was realized via artificial neural networks (ANNs) to detect the mentioned parameters. Also, different concrete compression strengths, beam lengths, and concrete cover thicknesses were benefited to generate a dataset for training the ANNs. Finally, to validate the model's success, test data was created and evaluated in comparison with optimal results. Thus, with these processes, both generating of eco-friendly, cost-effective, and optimal designs were made possible to provide, and rapid, effective, and reliable decision-making applications were presented to determine the parameters in terms of the mentioned structural designs.