Evaluating the Clogging Behavior of Pervious Concrete (PC) Using the Machine Learning Techniques

Pervious concrete (PC) is at risk of clogging due to the continuous blockage sand into it during its service time. This study aims evaluate and predict such behavior PC using hybrid machine learning techniques. Based on 84 groups dataset developed in earlier study, was determined by algorithm combing SVM (support vector machines) particle swarm optimization (PSO) methods. The PSO employed adjust hyperparameters verify performance 10-fold cross-validation. predicting results model were assessed coefficient determination (R) root mean square error (RMSE). importance influential variables evaluated as well. showed that can effectively be used construct predictive for PC. combined has advantage higher reliability validity than random selection. For verification process, able obtain values 0.9469 1.8148 R RMSE, showing accurately PC, guiding mix-design from perspective durability. size most important parameter thickness sample least significant factor affecting behavior. proportions smallest aggregate largest are two design parameters with consideration relatively scores, these aggregates should given special attention future anti-clogging purposes.