Compressive Strength Estimation of Waste Marble Powder Incorporated Concrete Using Regression Modelling

A tremendous volumetric increase in waste marble powder as industrial has recently resulted high environmental concerns of water, soil and air pollution. In this paper, we exploit the capabilities machine learning to compressive strength prediction concrete incorporating for future use. Experimentation been carried out using different compositions varying water binder ratios 0.35, 0.40 0.45 analysis. Effect dosages superplasticizer also considered. regression algorithms analyse effect on concrete, viz., multiple linear regression, K-nearest neighbour, support vector decision tree, random forest, extra trees gradient boosting, have exploited their efficacies compared various statistical metrics. Experiments reveal forest best model with an R2 value 0.926 mean absolute error 1.608. Further, shapley additive explanations variance inflation factor analysis showcase achieved optimizing use partial replacement cement concrete.