Artificial intelligence-based prediction of geotechnical impacts of polyethylene bottles and polypropylene on clayey soil

Abstract. This study aims to investigate the application of artificial intelligence (AI) methods in predicting the resilient modulus of soil mixtures with polyethylene (PE) bottles and polypropylene (PP). The AI methods used in the study are artificial neural network (ANN) and classification and regression random forest (CRRF), and the modeling was conducted using a database of 160 datasets. The study also evaluated the importance of different input parameters on the accuracy of the models. The results show that the CRRF model is more accurate than the ANN model in predicting the effects of materials PE and PP on soil resilient modulus. Additionally, the study found that the number of hidden layers and neurons in the ANN model should be optimized for the best performance and increasing their number does not always lead to increased accuracy. Finally, the study identified the most and least important input parameters for predicting the effect of PE and PP on the resilient modulus of the mixture using both AI models.