For my master’s thesis, I used Bullet Physics to analyse the compaction behaviour of hot asphalt. The abstract is provided below, and the full text is available here.

Abstract

Asphalt concrete is one of the most widely used materials in modern road construction, making the understanding of its mechanical behaviour and compaction characteristics essential for infrastructure design and maintenance. Traditionally, these properties are investigated through laboratory experiments using actual asphalt mixtures. However, such physical testing is both costly and time-consuming. As an alternative, numerical modelling techniques—particularly the Finite Element Method (FEM) and the Discrete Element Method (DEM)—have been employed. Despite their success, FEM struggles to capture discrete particle interactions, while DEM faces limitations in accurately representing complex particle shapes.

This study explores the potential of physics engines, specifically Bullet Physics, as an efficient computational framework for modelling the compaction behaviour of hot asphalt mixtures. Using the open-source Python interface PyBullet, the superpave gyratory compaction process is digitally replicated. A complex viscoelastic contact model based on Burgers’ formulation is developed to describe bituminous interactions. A comprehensive parametric study is conducted to evaluate the sensitivity of compaction behaviour to material and process parameters that are difficult to isolate experimentally.

The research highlights both the promise and challenges of adapting Bullet Physics for engineering applications. While the platform offers exceptional computational efficiency and visualisation capabilities, implementing custom contact models within PyBullet is nontrivial. Attempts using external force application or custom integration schemes revealed either instability or excessive computation times. The study concludes that full integration of advanced contact laws requires direct modification of the Bullet Physics source code.

Simulations using a simplified contact model demonstrate strong consistency and provide valuable insights into asphalt compaction mechanisms. The effects of inertia and mould friction are found to be negligible, while aggregate displacement and contact area evaluation are effectively visualised. Overall, Bullet Physics shows great potential for large-scale, cost-effective simulation of asphalt behaviour, paving the way for further development of digital compaction models.