Quantifying the impact of chemical composition on pervious concrete strength: a comparative analysis using full quadratic model and artificial neural network

Abstract

The present study investigates the effect of the total amount of chemical constituents in cement and supplementary cementitious materials on the compressive strength of pervious concrete. Experimental datasets were collected from the literature on different types of pervious concrete specimens modified with supplementary cementitious materials. A total of 659 data observations were collected. These were then analysed and modelled using three different approaches: linear regression (LR), full quadratic (FQ) and artificial neural network (ANN) models. These models’ purpose was to predict pervious concrete's compressive strength. The accuracy of the models was evaluated using correlation coefficient (R2), root mean square error (RMSE), mean absolute error (MAE), a-20 index and error distribution. The ANN model demonstrated superior effectiveness and accuracy in predicting the compressive strength of pervious concrete, as indicated by the performance metrics. The analysis results showed that Al2O3 content and curing time were the most influential parameters in predicting the compressive strength of concrete. In addition, SHapley Additive exPlanations (SHAP) analysis could determine if each input variable had a positive or negative effect on compressive strength. Al2O3, CaO, curing time and water content positively affected the compressive strength of pervious concrete, while SiO2 and corresponding alkalis had a negative effect.