Micro-scale mechanisms, such as inter-particle and particle-fluid interactions, govern the behaviour of granular systems. While particle-scale simulations provide detailed insights into these interactions, their computational cost is often prohibitive. At a recent Lorentz Center Workshop on “Machine Learning for Discrete Granular Media”, researchers explored how machine learning approaches can aid the development of constitutive laws and efficient data-driven surrogates for granular materials while also addressing uncertainty quantification. Attended by researchers from both the granular materials (GM) and machine learning (ML) communities, the workshop brought the ML community up to date with GM challenges. This position paper emerged from the workshop discussions. In this position paper, we define granular materials and identify seven key challenges that characterise their distinctive behaviour across various scales and regimes–ranging from gas-like to fluid-like and solid-like. Addressing these challenges is essential for developing robust and efficient models for the digital twinning of granular systems in various industrial applications. To showcase the potential of ML to the GM community, we present classical and emerging machine/deep learning techniques that have been, or could be, applied to granular materials. We reviewed sequence-based learning models for path-dependent constitutive behaviour, followed by encoder-decoder type models for representing high-dimensional data in reduced spaces. We then explore graph neural networks and recent advances in neural operator learning. The latter captures the emerging field evolution of interacting particles via efficient latent space representation. Lastly, we discuss model-order reduction and probabilistic learning techniques for high-dimensional parameterised systems, both of which are crucial for quantifying and incorporating uncertainties arising from physics-based and data-driven models. We present a typical workflow aimed at unifying data structures and modelling pipelines and guiding readers through the selection, training, and deployment of ML surrogates for granular material simulations. Finally, we illustrate the workflow’s practical use with two representative examples, focusing on granular materials in solid-like and fluid-like regimes.

The open Delft Advanced Research Terra Simulator (open-DARTS) framework is an open-source reservoir simulation software. The open-DARTS focused on energy transition applications, such as geothermal energy production and carbon sequestration. It enables the modeling of compositional thermal flow, coupled with a geomechanical solver based on the Finite Volume discretization and adjoints method for inverse modeling. The open-DARTS supports different grid types (structured, corner-point geometry, and unstructured), discrete fracture networks, contact mechanics, and various thermal-chemical interactions. The recently proposed generic nonlinear formulation supports the most general nonlinear PDEs designed for various energy transition applications. The open-DARTS has been implemented in C++ and Python to optimize hardware utilization while ensuring flexibility. The most computationally expensive part is written in C++ and compiled into libraries, which are subsequently exposed to Python using pybind11. This allows the extension and overriding of C++ functions by user-defined Python code. For example, using only a Python interface, one can adjust a timestep strategy, nonlinear solver, or properties output. Besides, the Python interface of open-DARTS provides straightforward coupling with other Python-based numerical modeling packages, including the meshing, file storage, caching, and visualization modules. The open-DARTS core uses the advantages of C++ language, such as efficient low-level memory management, object-oriented programming, compile-time polymorphism, and parallelization with OpenMP. One of the advantages of open-DARTS is the Operator-Based Linearization (OBL) technique, which can resolve challenges associated with complex physics and reduce the computation time, especially for ensemble-based simulations. We would also like to share our experience on the project, repository, and the development workflow configuration using gitlab.com, including the build system (cmake), handling merge requests, automated testing in CI/CD pipelines, documentation management (gitlab.io), wiki utilization, and release publishing. Additionally, Python’s integration into open-DARTS offers the advantage of straightforward installation via PyPI and simplifies defining requirements for users who prefer to avoid compiling code from source files.