Optimization of Mix Design for Sustainable Concrete Using Artificial Neural Networks

Abstract

Sustainable concrete is necessary to lessen the environmental impact of building. In the construction industry, supplemental cementitious materials (SCMs) added to recycled aggregate concrete (RAC) provide a sustainable alternative. The compressive strength (CS) of RAC made using SCM was thoroughly examined in this work, which compiled a dataset of 1000 samples from published literature. Conventional concrete mix design techniques are based on trial-and-error and empirical methods, which results in inefficient material use and performance optimization. This study explores the use of Artificial Neural Networks (ANNs) to forecast and optimize concrete mix design parameters in order to increase sustainability and performance. This research illustrates how machine learning may greatly expedite the design process by training an ANN model on a dataset that includes several concrete mix compositions and their resulting properties. According to the results, ANN-based models can determine the ideal mix proportions with less cement and better workability, which is in line with sustainability objectives. They can also accurately forecast compressive strength.