Guilherme Barros

Dr. Guilherme Barros serves as a Research Fellow at the University of Newcastle within the School of Engineering and Centre for Geotechnical Science and Engineering, specializing in computational geomechanics and machine learning applications for rockfall hazard prediction.

His academic credentials include:

• B.Sc. in Civil Engineering from Fluminense Federal University (UFF)
• M.Sc. in Civil Engineering from Pontifical Catholic University of Rio de Janeiro (PUC-Rio)
• Ph.D. in Civil Engineering from the University of Newcastle

Barros' research centers on Modelling and simulation (30% focus), Geomechanics and resources geotechnical engineering (30%), and Deep learning (20%), with additional expertise in Granular mechanics and Solid mechanics. His work pioneers hybrid approaches combining physics-based numerical methods (BEM, DEM, FEM) with machine learning to address rockfall hazards in mining, seismic wave propagation in unbounded domains, and computational limit analysis. This interdisciplinary framework significantly enhances predictive accuracy for geotechnical risk assessment in complex terrain.

His publication record reveals a clear trajectory toward advanced multi-scale numerical coupling, particularly BEM-DEM integration for dynamic geomechanical problems across 1D to 3D domains. These works address critical challenges in infinite domain modeling, wave propagation discontinuities, and rockfall fragmentation prediction, demonstrating strong applicability to mining safety and civil infrastructure resilience.

His scientific recognition includes:

• AGS NSW Research Award (Australian Geomechanics Society, 2025)

Barros has co-supervised two Master's students on rockfall analysis projects: one at the University of Parma investigating material layer impacts on hazard prediction, and another at Newcastle comparing machine learning techniques (Random Forest, KNN) for trajectory modeling. His international research network spans Brazil (6 publications), Australia (5), and Poland (5), supported by funding from the Australian Geomechanics Society.

As an active contributor to the Centre for Geotechnical Science and Engineering, he develops computational frameworks for real-world mining safety challenges, with current projects focusing on machine learning-enhanced rockfall prediction systems for NSW and QLD open-pit mines.