Prediction of Clean-out on Permeable Interlocking Concrete Pavers using Laboratory Experiments and Machine Learning
A major problem with permeable interlocking concrete pavers (PICPs) is the significant cost associated with its clean-out to restore the original functionality, which is substantial and discouraging for potential users and municipalities. The current investigation employs a novel understanding of variables affecting the sustainable and economically feasible maintenance of PICP. Two new models have been derived to predict more accurately the percent mass removal from the PICPs using Artificial Neural Network (ANN) and Gene Expression Programming (GEP). Four novel non-dimensional parameters were developed using five independent variables (cleaning equipment speed over the pavement; air speed in the cleaning jets; lateral depth of the cupule, top opening width of the cupule, and filter media gradation) that affect the cleaning of the permeable pavement. The findings of this research can be applied to the industrial application of Regenerative Air Street Sweepers, which is economically feasible PICP maintenance.