Mechanical Performance and Predictive Modeling of Hot Mix Asphalt Modified with Nanoclay

Abstract This study investigates the effect of nanoclay (NC) additives on the mechanical performance of hot mix asphalt (HMA). NC was incorporated at 0%, 2%, 4%, and 6% by binder weight, and the mixtures were tested for Marshall properties, indirect tensile strength (ITS), compressive strength, and resilient modulus (M R ) at 5°C, 25°C, and 40°C. Results revealed notable improvements: Marshall stability increased by 26.9% at 6% NC, while flow values decreased by 13.8% up to 4% NC. ITS and compressive strength increased by 24.7% and 27.4%, respectively. M R values improved by 66%, 81.6%, and 60% at 5°C, 25°C, and 40°C, respectively, for 6% NC. An Artificial Neural Network (ANN) model was developed using nanoclay content and temperature as inputs to predict M R . The model achieved high accuracy (R² = 0.9434, RMSE = 508.87 MPa), demonstrating its effectiveness in capturing the nonlinear response of nanoclay-modified mixtures and providing a reliable predictive tool.