A Foundation Model for Material Fracture Prediction

TL;DR

This paper introduces a transformer-based foundation model for predicting material fracture across diverse materials, geometries, and loading conditions, unifying physics-based and machine learning approaches for scalable, adaptable failure prediction.

Contribution

The paper presents a versatile, multimodal transformer model that generalizes fracture prediction to new materials and scenarios with minimal data, surpassing traditional narrow ML models.

Findings

Model predicts fracture across multiple materials and conditions.

Requires minimal data for new materials, as few as one sample.

Unifies physics-based and ML methods into a single scalable framework.

Abstract

Accurately predicting when and how materials fail is critical to designing safe, reliable structures, mechanical systems, and engineered components that operate under stress. Yet, fracture behavior remains difficult to model across the diversity of materials, geometries, and loading conditions in real-world applications. While machine learning (ML) methods show promise, most models are trained on narrow datasets, lack robustness, and struggle to generalize. Meanwhile, physics-based simulators offer high-fidelity predictions but are fragmented across specialized methods and require substantial high-performance computing resources to explore the input space. To address these limitations, we present a data-driven foundation model for fracture prediction, a transformer-based architecture that operates across simulators, a wide range of materials (including plastic-bonded explosives, steel, aluminum, shale, and tungsten), and diverse loading conditions. The model supports both structured and unstructured meshes, combining them with large language model embeddings of textual input decks specifying material properties, boundary conditions, and solver settings. This multimodal input design enables flexible adaptation across simulation scenarios without changes to the model architecture. The trained model can be fine-tuned with minimal data on diverse downstream tasks, including time-to-failure estimation, modeling fracture evolution, and adapting to combined finite-discrete element method simulations. It also generalizes to unseen materials such as titanium and concrete, requiring as few as a single sample, dramatically reducing data needs compared to standard ML. Our results show that fracture prediction can be unified under a single model architecture, offering a scalable, extensible alternative to simulator-specific workflows.