Numerical framework for asphalt pre-compaction optimization using a coarse-graining discrete element method (DEM)

Pre-compaction shapes the initial aggregate skeleton and strongly affects the long-term performance of asphalt pavements. Yet optimizing this stage remains difficult: field trials are costly, intrusive, and hard to repeat, while many numerical models oversimplify aggregate morphology and ignore temperature-dependent adhesion during compaction, limiting practical guidance. This study proposes a high-fidelity discrete element method (DEM)–coarse-graining strategy (CGS) framework for pre-compaction. It integrates 3D-scanned aggregate geometries with a temperature-evolving JKR contact model to capture realistic particle–interface mechanics. A validated CGS reduces particle numbers while preserving physical representativeness, enabling construction-scale simulations. The framework is calibrated and independently verified using SmartRock blending tests, then applied to full-scale pre-compaction to quantify how paving speed, paving angle, and layer thickness influence compaction behavior across different gradations and operating conditions. Results show the CGS-based DEM is reliable and scalable for evaluating and optimizing asphalt paving, supporting improved construction quality and intelligent compaction systems.