Optimizing Construction Sustainability: Utilizing Machine Learning for Analyzing the Strength Properties of Concrete with Partial Replacement of Cement by Calcium Carbonate Powder in Concrete Mixtures

With an ever-increasing focus on sustainable construction practices, there is a growing need for innovative methods that can effectively minimize environmental impact. This research investigates the application of advanced machine learning (ML) techniques to comprehensively evaluate the strength properties of concrete. By introducing partial cement replacement with calcium carbonate powder, the study thoroughly examines the alterations in compressive and flexural strength attributes across a spectrum of varied cement substitution ratios. Notably, the incorporation of calcium carbonate powder demonstrates a marked enhancement in concrete strength, highlighting its potential as a key component in optimizing sustainable construction methodologies. Additionally, the research incorporates comprehensive analysis techniques, such as R-squared analysis, encompassing a substantial dataset of 160 data points and a detailed assessment of 20 parameters. The utilization of scatter plots further accentuates the predictive capabilities of five distinct ML models, emphasizing the superior performance of the decision tree (DT) and extra tree (ET) models in accurately forecasting concrete strength. These findings underscore the significant role of these ML models in advancing the realm of material science and engineering, facilitating the development of improved concrete formulations and fostering sustainable construction practices.