COMPLAS 2027

1100 Computational Plasticity: Formulations and Integration Algorithms
The return-mapping algorithm on the multiplicative decomposition F = F^e F^p — comprising elastic trial, closest-point projection, and consistent algorithmic tangent — is the innermost kernel of every implicit finite-strain plasticity code. Classical backward-Euler integrators lose robustness for non-smooth or multi-surface yield criteria, large strain increments in crystal plasticity, and coupled plasticity–damage or plasticity–phase-field formulations. This mini-symposium addresses algorithms that improve, extend, or replace the standard return map and the global solvers it couples with. Topics include: (1) Semi-implicit integration and operator splitting — explicit–implicit splits on GL+(3), IMPL-EX extrapolation, staggered crystal-plasticity algorithms with dynamic relaxation across slip systems. (2) Optimization-based and non-smooth return mapping — SQP, augmented Lagrangian, trust-region, and interior-point formulations; semi-smooth Newton with NCP functions for vertex-type yield surfaces; variational constitutive updates. (3) Data-driven integration and automated model discovery — PINN-based Gauss-point solvers, neural-ODE for internal-variable evolution, physics-constrained NN yield functions (ICNNs, polyconvex PANNs), symbolic-regression discovery of hardening laws. (4) Stochastic plasticity integration — polynomial chaos and Bayesian inference of internal-variable trajectories. (5) Gradient-enhanced and nonlocal finite-strain plasticity — implicit-gradient, micromorphic, strain-gradient formulations; Lip-field regularization. (6) Phase-field fracture coupled with finite-strain plasticity — staggered vs. monolithic solvers, Anderson and Schur-complement acceleration, ductile fracture with kinematic hardening, phase-field fatigue. The unifying thread is the finite-strain plasticity integration algorithm — from the local return map to the global coupled solver. The aim is to cross-fertilize strategies across constitutive frameworks and communities (computational plasticity, optimization, machine learning, stochastic mechanics).
Organized by: P. Areias (Instituto Superior TĂŠcnico, Portugal), T. Rabczuk (Bauhaus University Weimar, Germany) and N. Silvestre (Instituto Superior TĂŠcnico, Portugal)
Keywords: constitutive integration algorithms, data-driven model discovery, finite-strain plasticity, non-smooth and optimization-based return mapping
The understanding of the mechanical behavior of a material under the operational environment has long been an important issue for the design and safety assessment. In recent years, particular attention has been given to research with applications to multiaxial cyclic loading. Under the multiaxial cyclic loading, the yield surface of the material translates, expands, and distorts. Towards the accurate prediction of yield surface evolution, several new plasticity models have been proposed and studied in recent years [1, 2]. Current work suggests, combining the kinematic, the isotropic, and the distortional hardening rules would a crucial point. However, this development greatly increases demands on computing power and require careful examination of numerical convergence. Furthermore, dedicated optimizations of numerical implementation are crucial for computational efficiency of mixed hardening predictions, numerical verification, and calibration. This MS provides a platform for researchers to exchange his/her work which is especially related to but not limited to the investigation on the distortional hardening/softening in metals, polymers, biological materials, and other materials. The computational investigation includes not only the continuum-based but also the particle-based approach whereas the computational modeling contains single-scale as well as multi-scale simulations. Theoretical research and experimental study are all welcome. REFERENCES [1] Reyne B., Barlat F., A new concept for continuum distortional plasticity, International Journal of Plasticity, Vol. 155, 103303, 2022. [2] Choi, H., Yoon, J.W., A new simplified distortional hardening model for nonlinear strain paths, International Journal of Plasticity, Vol. 165, 103617, 2023.
Organized by: L. Liu (National Taiwan University, Taiwan) and S. Parma (Institute of Thermomechanics of the CAS, Czechia)
Keywords: Constitutive modelling
1200 Advanced Constitutive Modelling of Nonlinear Materials
Advanced constitutive models for polymeric materials, thin films, and interlayers in laminated systems have been developed by research groups for several decades. A key challenge in applying such models in engineering practice is the complex interplay of physical mechanisms governing the mechanical response of polymers under varying stress states, strain rates, and temperature conditions. Furthermore, experimental data often exhibit limited resolution, significant scatter, and strong sensitivity to microstructure, processing conditions, and loading history. In recent years, non-local and peridynamic constitutive formulations have emerged as promising approaches for modeling deformation, damage, and failure in polymeric materials. At the same time, advances in artificial intelligence (AI) and machine learning are increasingly supporting the development, calibration, and validation of constitutive models. This session aims to bring together recent developments in constitutive modeling of polymers, with particular focus on inelastic deformation, viscoelasticity, viscoplasticity, and damage evolution. Contributions addressing hybrid physics-machine learning approaches, uncertainty quantification, data-driven model discovery, and experimental-computational integration are especially encouraged.
Organized by: D. Juhre (Otto von Guericke University Magdeburg, Germany), K. Naumenko (Otto von Guericke University Magdeburg, Germany) and M. Johlitz (Uni BW Munich, Germany)
Keywords: damage, failure, polymers, viscoelasticity, viscoplasticity
This mini-symposium aims to provide a forum for sharing and discussing research works in the general area of non-Newtonian fluid and multiphase matter, especially those inter-disciplinary studies that cross the traditionary boundary between solids and fluids. Many natural and industrial processes involve dynamic motions of both fluids and solids, forming a complex multiphase matter which is often further complicated by non-Newtonian / viscoelastic / viscoplastic behaviours, phase transitions, chemical reactions, and the presence of porous media. Examples include welding and casting, 3D printing, polymer injection moulding, fresh concrete placement, oil and grease lubrication, debris flow, sedimentation, dust storm, and many more. • This session welcome, but is not limited to, the following topics: • Physical and mathematical models of multiphase systems and processes • Numerical modelling of multiphase matter and non-Newtonian fluids • Computer simulation of complex systems involving multiple fluids and solids • Numerical and experimental studies of materials and processes involving phase transition and/or chemical reaction • Applied studies that cross the traditional boundary of solids and fluids: 3D printing, injection moulding, concrete placement, debris flow, sandstorm, welding and casting, food processing, glass forming, and others.
Organized by: C. Li (Zienkiewicz Institute for Modelling, Data and AI, Swansea University Bay Campus, Swansea SA1 8EN, United Kingdom)
Keywords: Constitutive modelling
Predictive simulation of nonlinear materials relies on constitutive models that capture complex time- and history-dependent phenomena, including viscosity, damage, fatigue, aging, phase transformation, microstructural evolution, and multiphysics effects. As these models grow in fidelity and complexity, robust methods for identifying material parameters, assessing model adequacy, and selecting among competing constitutive descriptions become essential for reliable computational mechanics. This session will focus on approaches for parameter identification and model selection for advanced nonlinear material models, with emphasis on materials exhibiting path dependence, rate effects, hysteresis, degradation, and evolving internal state variables. Contributions are invited on experimental design for parameter calibration, inverse methods, uncertainty quantification, Bayesian inference, optimization-based identification, sensitivity analysis, surrogate and reduced-order modeling, machine learning-assisted calibration, and physics-informed data-driven methods. Particular interest will be given to approaches that address non-uniqueness, parameter correlation, limited or noisy experimental data, computational cost, and transferability across loading conditions and scales. The session aims to bring together developers of constitutive models, experimental characterization methods, inverse analysis techniques, and verification and validation frameworks. Applications may include polymers, metals, composites, geomaterials, energetic materials, biological tissues, architected materials, and other nonlinear material systems where history-dependent behavior plays a central role. By highlighting both methodological advances and practical case studies, the session will provide a forum for discussing how model identification and selection can improve predictive capability, quantify confidence in simulations, and guide the development of more interpretable and reliable constitutive models for complex material behavior. Sandia National Laboratories is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.
Organized by: C. Alleman (Sandia National Laboratories, United States), J. Fuhg (The University of Texas at Austin, United States), R. Jones (Sandia National Laboratories, United States) and M. Stender (Sandia National Laboratories, United States)
Keywords: Constitutive modelling, Data-driven modelling, Optimization, Surrogate Modelling
Shape memory alloys offer distinctive opportunities for adaptive structures, actuators and energy dissipation through their shape memory effect and superelasticity, both leading to the evolution of inelastic, so-called transformation strains. From the start, models describing shape memory alloys have been tightly coupled to computational plasticity, with transformation surfaces and flow rules governing phase transformation. Recent advances in manufacturing, especially additive manufacturing, have opened new areas of application ranging from solid-state cooling to patient-specific biomedical implants. At the same time, the process-induced heterogeneity of the microstructure, porosity, local defects and a non-trivial coupling of printing parameters, printed geometry and material behaviour pose a challenge to classical modelling techniques. Together with well-known challenges such as non-proportional loading, functional degradation under cyclic loading and the interaction between the superelastic response and plasticity, the accurate prediction of superelastic material behaviour under practically relevant conditions requires interdisciplinary research on manufacturing processes, constitutive modelling and numerical simulation, as well as the experimental characterization of shape memory alloys. Contributions are consequently invited from both communities, working experimentally or numerically on shape memory alloys and focusing on process-material relations, thermodynamically consistent constitutive formulations, finite strain approaches or robust numerical algorithms. By integrating advances in computational plasticity with manufacturing and experimental evidence, the mini-symposium seeks to clarify the interplay between recoverable transformation strains and irreversible deformation, identify outstanding modelling challenges, and advance the reliable simulation and design of shape memory alloy components.
Organized by: M. Brodbeck (University of Stuttgart, Germany), S. Seyedpour (University of Stuttgart, Germany) and A. Celebi (The University of Toledo, United States)
Keywords: additive manufacturing, Constitutive modelling, Finite Element Method (FEM), Shape Memory Alloys, Superelasticity
1300 Crystal Plasticity and Micromechanics
The collective motion and interaction of dislocations are the fundamental mechanisms underlying plastic deformation in crystalline materials. Capturing these processes in predictive models remains a major challenge because atomistic-scale mechanisms ultimately govern macroscopic mechanical behavior across many orders of magnitude in length and time. This challenge becomes even more pronounced in complex alloy systems and under nontrivial loading conditions. Here, multiple interacting deformation mechanisms give rise to highly coupled, high-dimensional behavior that is difficult to isolate experimentally and challenging to represent within a single modeling framework. Developing a mechanistic understanding of this complex parameter space requires modeling approaches that span multiple length and time scales, linking dislocation-scale processes to engineering-scale material response. Addressing this need has led to a diverse range of methodologies, from physics-based homogenization and multiscale frameworks to data-driven and AI-assisted approaches, each providing unique capabilities for capturing different aspects of plastic deformation. This symposium brings together researchers working on the modeling and simulation of dislocation-mediated plasticity, with particular emphasis on scale-bridging strategies that enhance physical understanding and enable efficient information transfer across scales. Contributions are invited from a broad range of approaches, including discrete dislocation dynamics, crystal plasticity, continuum and homogenization methods, multiscale modeling frameworks, data-driven and physics-informed methods, and machine learning or AI-based techniques for surrogate modeling, accelerated simulations, and the replacement of computationally intensive modeling components.
Organized by: M. Sudmanns (RWTH Aachen University, Germany), G. Po (University of Miami, United States) and Y. Cui (Tsinghua University, China)
Keywords: AI-based surrogate modeling, crystal plasticity, data-driven scale bridging, discrete dislocation dynamics, homogenization methods, physics-informed modeling
Myriads of plasticity and damage models successfully reproduce experimental observations but far fewer can reliably predict material behavior beyond the testing domain used for calibration. Fewer still can capture local responses, where heterogeneity, localization, and failure arise. Models that capture the dominant mechanisms that drive deformation and damage may support the extrapolation within that mechanistic domain. In practice, nearly every prediction involves a transfer from incomplete evidence to previously unobserved conditions. We seek to identify and discuss the modeling attributes required to render material behavior predictive. This symposium is motivated by the view that predictive capability depends less on the amount of information available than on whether that information can diagnose the mechanisms controlling material behavior. While agreement with experimental results is a necessary condition, the mere availability of larger quantities of data is not sufficient to ensure predictive fidelity. Trustworthy predictions require models grounded in mechanisms that retain their validity under changing conditions, experiments designed to reveal the dynamics of those mechanisms, and formulations and calibration strategies that yield interpretable and identifiable parameters. Equally important are transparent assessments of uncertainty and clear articulation of model limitations. We seek to bring together modeling and experimental perspectives that advance the predictive capability of plasticity and damage models beyond the conditions used for their calibration. Relevant themes include governing mechanisms of deformation and failure, physics-aware calibration and interpretable parameterization, mechanism-revealing experimentation, prediction of local fields and failure initiation, and the quantification of predictive uncertainty and model transferability or generalizability across materials, microstructures, and loading conditions.
Organized by: A. Tallman (Florida International University, United States), K. Sedighiani (Tata Steels & Eindhoven University of Tech., Netherlands) and G. Castelluccio (Cranfield University, United Kingdom)
Keywords: Constitutive modelling, Coupled plasticity and damage, crystal plasticity, damage, discrete dislocation dynamics, fatigue, Multiscale modelling, physics-informed modeling
The (multi)phase-field method is widely established for the simulation of microstructure evolution, since it provides a numerically efficient treatment of moving interfaces without explicit interface tracking. In many applications, however, the diffuse interface is not merely a numerical transition region. It is the region where mechanical compatibility, balance laws, constitutive assumptions, and dissipative processes have to be formulated consistently. The objective of this invited session is to discuss recent developments in diffuse-interface approaches to inelastic and coupled material behavior. Particular emphasis is placed on the treatment of plastic, viscoplastic, viscous, damage-like, and damage and fracture-related processes within diffuse interface regions, as well as on their coupling to phase evolution, heat conduction, diffusion, and chemical driving forces. Contributions may address phenomenological or mechanism-based (crystal) plasticity theories, strain inheritance during phase transformations, chemo-mechanical and thermo-mechanical coupling, approaches relating phase-specific fields to an effective/interpolated field, and the relation between sharp-interface and diffuse-interface formulations. The session aims to bring together researchers working on continuum-mechanical foundations, thermodynamically consistent formulations, numerical implementations, and applications to evolving microstructures in crystalline, polymeric, composite, and multiphase materials.
Organized by: A. Prahs (Karlsruhe Institute of Technology (KIT), Germany), D. Schneider (Karlsruhe Institute of Technology (KIT), Germany), B. Nestler (Karlsruhe Institute of Technology (KIT), Germany), Y. Le Bouar (UniversitĂŠ Paris-Saclay, ONERA, CNRS, France) and B. Appolaire (UniversitĂŠ de Lorraine, CNRS, France)
Keywords: plasticity theories, (Multi)phase-field method, inelastic material behavior
1400 Multi-Scale Modelling and Scale Bridging
This invited session focuses on multi-scale studies utilising advanced computational modelling and characterisation techniques to elucidate the deformation and strengthening mechanisms of structural materials. Various computational frameworks, including emerging data-driven methodologies, have been developed to bridge multiple length scales, spanning the quantum and atomistic levels to the macroscopic scale, in an effort to better predict material response and structural behaviour. These approaches are fundamental to improving our understanding of the underlying physics of materials and are essential for accelerating the design of next-generation materials with optimised properties. In this context, advanced characterisation, particularly in-situ and data-intensive time-resolved studies, is increasingly vital for validating these computational frameworks and discovering novel mechanisms. This session aims to showcase recent research employing computational, experimental, and hybrid approaches to study structural materials, especially under complex loading conditions where the material microstructure evolves during deformation. Topics of interest include: +Scale-bridging and multiscale computational frameworks +Advanced modelling and characterisation of microscale material behaviour +Integration of theoretical, experimental, and data-driven methodologies +Practical applications of the integrated approaches
Organized by: I. Watanabe (National Institute for Materials Science, Japan), J. Fonseca (University of Manchester, United Kingdom), B. Holmedal (Norwegian University of Science and Technology, Norway) and M. Kasemer (University of Alabama, United States)
Keywords: Constitutive modelling, Data-driven modelling, Multiscale modelling
Computational design of advanced engineering materials with complex underlying structure requires key enabling numerical techniques such as multi-scale analysis and generalized continuum theories. This is the case of high-performance engineering materials, architected materials, metamaterials or natural materials exhibiting complex phenomena such as fracture across the scales. In these scenarios, dedicated simulation techniques that account for microscale mechanics and its interaction with upper scales are crucial for a successful design process. The objective of this symposium is to provide a place for discussion and exchange of ideas regarding the modelling, design and analysis of materials, taking into account their microstructure and their (possibly) non-classical behaviour at different scales. On the one hand, recent advances on the numerical description of this class of materials are foreseen, with focus on the multi-scale modelling through homogenisation schemes, techniques for optimal design of macro or microstructure and constitutive modelling based on generalised continua (Cosserat, micromorphic, strain gradient, ...). On the other hand, there is also place to share the application and potential industrial transference of this sort of techniques to specific classes of materials, like multi-phase materials, metamaterials, fibre reinforced composites, polycrystalline materials, biological structures, and architected materials, not to be exhaustive. Contributions addressing but not limited to the topics listed in what follows are welcomed: ďż˝.- Multi-scale models based on first and second-order homogenisation, micromorphic homogenisation, and homogenisation of generalised continua; ďż˝.- Modelling of fracture and damage across the scales; .- Analysis of size effects across the scales; ďż˝.- Data-driven, reduced-order and surrogate models for multi-scale and/or generalised continua; .- Constitutive modelling and parameters calibration in generalised continua; .- Multi-scale design and topology optimisation of high-performance materials and metamaterials; ďż˝ .- Analysis of the influence of strain gradients in materials behaviour; ďż˝ .- Numerical methods to solve generalised continua and multi-scale problems; ďż˝ .- Industrial multiscale analysis of complex and high-performance engineering materials.
Organized by: O. Lloberas-Valls (CIMNE, Spain), I. Rodrigues Lopes (INEGI, Portugal), W. Dos Santos (Universidade de SĂŁo Paulo, Brazil) and F. Andrade Pires (INEGI, Portugal)
Keywords: AI-based surrogate modeling, homogenization methods, Multiscale modelling
In recent years, considerable progress has been made in connecting the micro- and mesoscopic mechanics of materials to the structural engineering level supported by advances in multiscale modeling. To this end, different classes of computational scale-bridging methods have been developed, spanning various disciplines, e.g. engineering, computational mechanics, mathematics, physics, and chemistry. Although these methods are usually designed for specific research problems, from a methodological point of view, similarities and distinctive features can be identified. This invited session intends to bring together scientists from different disciplines working on scale bridging problems (both spatial and temporal) in materials and structures. The topics addressed in this invited session include, but are not limited to: • Homogenization-based methods: mathematical homogenization, computational homogenization etc. • embedded domain and domain decomposition methods, global-local techniques • heterogeneous multiscale method (HMM), equation-free method • (non-equilibrium) thermodynamics-based coarse graining methods • methods for bridging distinct models, e.g. atomistics-to-continuum • methods for phenomena with (partially) non-separating scales, e.g. localization, damage and fracture or transient phenomena • methods for bridging temporal scales • multiscale methods for coupled multi-field phenomena • methods for interfaces and contact conditions • model reduction techniques and reduction of computational costs associated with multiscale algorithms and complex microstructures • methods for integration of experimental data (e.g. imaging) into multi-scale numerical algorithms • data-driven and machine learning based multi-scale approaches • quantum-assisted solution methods and numerical algorithms
Organized by: D. Balzani (Ruhr Universität Bochum, Germany), C. Oskay (Vanderbilt University, United States), V. Kouznetsova (Eindhoven University of Technology, Netherlands), J. Schröder (Universität Duisburg-Essen, Germany) and K. Terada (Tohoku University, Japan)
Keywords: Advanced numerical methods, composite structures, data-driven scale bridging, homogenization methods, Multiscale Modeling
Microstructure plays a decisive role in determining the macroscopic properties and performance of materials. To enable resource-efficient material design and processing, reliable and computationally efficient prediction of material behavior is essential. However, the underlying microstructure and phenomena at smaller length scales - such as crystal plasticity, phase transformations, and dynamic recrystallization - strongly influence the macroscopic response and often evolve dynamically during processing. The complexity of these problems is further increased by coupled and multifield loading conditions, including thermo-mechanical, electro-magnetic, and electro-chemical interactions. Resolving such phenomena across multiple length and time scales can lead to substantial computational costs, motivating the development of efficient multiscale and model-order-reduction strategies that retain the relevant physical mechanisms while reducing computational effort. This minisymposium brings together researchers working on advanced computational and experimental approaches for predicting microstructural evolution and its impact on material behavior. It aims to provide a platform for discussing recent developments, emerging methodologies, and challenges in the efficient simulation and validation of multiscale material systems. Particular emphasis will be placed on, but is not limited to, the following topics: • Multiscale methods for evolving microstructures, including crystal plasticity, phase transformations, dynamic recrystallization, and related phenomena • Multiscale methods for coupled and multifield problems, including mechanical, chemical, thermal, and electrical interactions • State-of-the-art approaches for improving the computational efficiency of multiscale simulations, including FFT-based Lippmann-Schwinger solver and model order reduction techniques • Data-driven and machine-learning approaches for accelerating and enhancing two-scale and multiscale methods • Generation and characterization of representative volume elements (RVEs) for multiscale simulations • Experimental characterization and validation of multiscale models and predictions
Organized by:
Keywords: Multiphysics modeling, Multiscale Modeling
2100 Damage, Fracture and Failure Mechanics
Recent advances in computational mechanics have significantly enhanced the fidelity of models for inelastic deformation, damage evolution, fracture, and failure in engineering materials and structures. This is especially timely given the rapid proliferation of architected metamaterials and lattice structures, where the interplay between geometric design and inelastic response opens new and largely uncharted territory for computational damage mechanics. This invited session aims to provide a forum for contributions on coupled plasticity and damage, with an emphasis on physics-based, data-driven, and multiscale approaches. The scope includes constitutive modelling, multiscale and multiphysics simulation, inverse problems, scientific machine learning, reduced order and surrogate modelling, and advanced discretization methods. The session welcomes both fundamental and application-oriented studies, including work connected to experiments, digital twins, and industrial simulation. Topics of interest include, but are not limited to, the following: • Physics-based and data-driven modelling strategies • Coupled plasticity and damage constitutive models • Fracture, failure, and degradation across scales • Multiscale and multiphysics approaches • Scientific machine learning and surrogate modelling • Inverse problems and parameter identification
Organized by: E. Kakouris (University of Warwick, United Kingdom), S. Triantafyllou (National Technical University of Athens, Greece), H. Wei (Ansys, part of Synopsys Inc., United States), G. Lin (Ansys, part of Synopsys Inc., United States), F. Pavia (Ansys, part of Synopsys Inc., Switzerland), Y. Huang (The Hong Kong Polytechnic University, China, Hong Kong) and F. Azhari (Monash University, Australia)
Keywords: Computational damage mechanics, Constitutive modelling, Data-driven modelling, Fracture and failure , Multiscale modelling, Scientific machine learning, Coupled plasticity and damage
Cracks in brittle and quasi-brittle materials often initiate and grow in unpredictable ways when the material is subjected to complex combinations of loads. A central challenge is determining where a crack will form and how it will propagate, especially when multiple forces interact and lead to unstable or sudden failure. Fracture mechanics provides mathematical tools to describe this process, which involves the creation of new surfaces inside the material at locations and orientations that are not known in advance. These theoretical models form the basis for numerical simulations, where accuracy can be assessed and improved through controlled computational strategies. Crack growth is typically governed by physical criteria related to the material reaching its local strength limits. More recently, energy-based formulations have enabled the development of new computational approaches. Many widely used methods rely on finite element models, where cracks are explicitly represented by inserting discontinuities between or through elements. Phase-field methods have gained attention because they allow cracks to emerge naturally without tracking their geometry explicitly. Boundary element methods have also proven effective in certain cases. In addition, newer mesh-free approaches avoid the need to modify the computational grid as cracks evolve, simplifying implementation. Some advanced models combine multiple numerical techniques or operate across different scales to capture complex fracture behavior more accurately. This invited session at COMPLAS aims to bring together researchers working on both theoretical and computational aspects of fracture. The goal is to discuss emerging methods for predicting crack initiation and tracking crack paths, as well as improvements and applications of established techniques.
Organized by: A. Pandolfi (Politecnico di Milano, Italy), K. Weinberg (University of Siegen, Germany) and M. Ortiz (Caltech, United States)
Keywords: Computational damage mechanics, Fracture and failure, Multiscale modelling
Cementitious materials underpin civil engineering infrastructure, and understanding their fracture, damage, and failure mechanisms is essential to meet today’s requirements for durability, service life, and sustainable design. This symposium provides a concise overview of current advances and emerging directions in the mechanics and modelling of cement-based materials. Topics include: Fracture and damage mechanics in cement-based materials, including heterogeneity effects. Multi-scale and computational modelling linking micro/meso/macro behaviour. Durability and environmental/time-dependent degradation (freeze–thaw, chemical attack, long-term loading). Fibre-reinforced concrete (FRC) and other advanced cementitious materials, including ultra-high-performance concrete (UHPC), under varied loading regimes. Emerging technologies and resilience strategies, including 3D-printed cementitious materials, self-healing mechanisms, and non-destructive testing/monitoring for predictive maintenance. High‑value utilization of industrial solid wastes (e.g., coal gangue, recycled aggregates) in low‑carbon cementitious composites.
Organized by: H. Zhang (North University of China, China) and R. Yu (University of Castilla-La Mancha, Spain)
Keywords: Computational damage mechanics, Constitutive modelling, Fracture and failure, Multiscale modelling
In the 26 years since its inception, the phase-field approach to fracture has established itself as a staple of computational fracture mechanics. Indeed, owing to its ease of implementation, the original framework has served as the computational basis for a plethora of extended, damage-like models that incorporate diverse physical mechanisms. These developments have taken different and often diverging conceptual paths, particularly regarding the connection to sharp-crack theories and the variational structure of the original formulation. This invited session offers a platform to present and discuss current trends and developments in phase-field fracture modeling. General areas of interest include, but are not limited to the following: • theoretical foundations, • energy decompositions, • nucleation/strength criteria, • cohesive fracture, • inelasticity, • fatigue, • dynamics, • multiphysics and multiscale processes. Contributions that critically examine the foundations, limitations, and future directions of phase-field fracture methods are especially welcome.
Organized by: J. Ulloa (University of Michigan, United States), R. Alessi (University of Pisa, Italy), L. De Lorenzis (ETH Zürich, Switzerland), J. Sargado (University of Copenhagen, Denmark) and K. Yoshioka (Technical University of Leoben, Austria)
The symposium is dedicated to understanding the relationship between strength and toughness in material failure. While engineering design has traditionally relied on strength-based approaches, fracture mechanics has highlighted the importance of toughness in determining structural integrity. Connecting these two perspectives remains a fundamental challenge, particularly for materials containing defects, heterogeneities, or evolving damage. The symposium will explore experimental, theoretical, and computational approaches that link strength and toughness across materials and scales. Emphasis will be placed on the mechanisms governing this relationship, methods for characterizing both properties, and their application to the design of safer and more damage-tolerant materials and structures. Contributions are invited on topics including: • Damage mechanisms, crack initiation, and crack propagation. • The role of microcracking, defect interactions, and fracture process zones in fracture. • Transitions between strength-controlled and toughness-controlled failure. • Experimental methods for measuring strength, toughness, and fracture processes. • Numerical and theoretical approaches, including coupled criterion, finite fracture mechanics, phase-field and cohesive zone models, discrete methods. • Fracture challenges in interfaces, composites, additively manufactured materials, architected materials, and small-scale testing. • Material and structural design strategies that exploit the correlation between strength and toughness. By bringing together researchers from physics, materials science, mathematics, and mechanics, the symposium aims to advance a unified understanding of how strength and toughness govern material failure and how this knowledge can be translated into engineering practice.
Organized by: G. Molnar (CNRS / INSA Lyon, France), A. Doitrand (INSA Lyon, France) and D. Kondo (Sorbonne University, France)
Keywords: Computational damage mechanics, Damage and Fracture, Localization Phenomena
Phase-field modelling has emerged as a powerful computational framework for simulating complex fracture processes without explicit tracking of evolving crack surfaces. Despite its versatility in capturing crack initiation, propagation, branching and coalescence, conventional phase-field approaches remain computationally demanding, particularly for three-dimensional, multiscale and fatigue-fracture problems. At the same time, rapid developments in deep learning and scientific machine learning are creating new opportunities to enhance the efficiency, scalability and predictive capability of phase-field fracture models. This invited minisymposium focuses on recent developments at the intersection of deep learning and phase-field modelling for fracture, damage and failure mechanics. Particular attention will be given to deep neural network formulations for solving phase-field fracture problems, including deep energy methods, physics-informed neural networks (PINNs), neural operators and other physics-informed learning strategies. Contributions exploring hybrid approaches in which machine learning is integrated with conventional finite-element phase-field formulations are also particularly encouraged. The minisymposium will address both brittle and cohesive fracture and welcomes extensions to ductile fracture, fatigue, heterogeneous and composite materials, and coupled multiphysics failure. Topics of interest include deep-learning-based solution of coupled displacement–phase-field equations, data-driven identification of phase-field parameters and constitutive behaviour, surrogate and reduced-order models, neural operators for rapid fracture prediction, adaptive learning strategies, and computational acceleration of large-scale phase-field simulations. A central objective is to examine how deep learning can overcome some of the computational limitations of conventional phase-field methods while retaining physical consistency, robustness and generalisability. Contributions addressing accuracy, convergence, interpretability, training efficiency and comparisons with established finite-element solutions and experimental observations are especially welcome. By bringing together researchers in phase-field fracture, computational mechanics and scientific machine learning, the minisymposium aims to promote new computational paradigms for efficient and physically informed prediction of fracture, damage and structural failure.
Organized by: J. Ghaffari Motlagh (Leeds Beckett University, United Kingdom)
Keywords: (Multi)phase-field method, Damage and Fracture, Fracture and failure, physics-informed modeling, Scientific machine learning
2200 Fatigue
Understanding and controlling fatigue-driven damage in structural materials remains one of the most challenging problems in computational mechanics, largely owing to the inherently multiscale and stochastic nature of crack initiation, propagation, or even possible self-healing. Classical continuum models, while powerful, struggle to connect the atomistic origins of unit mechanisms (grain-boundary migration, nanovoid formation, dislocation emission) to the emergent, macroscopic fatigue response observed in experiments. Bridging these scales demands not only advanced multiscale simulation strategies, but also new probabilistic frameworks for learning, interpreting, and eliciting the complex interactions that drive fatigue processes. Machine learning and data-driven methods are increasingly offering new paths forward, not simply as replacements for physics-based models, but as powerful complements that can accelerate simulation, extract structure from high-dimensional data, uncover governing relationships that are not amenable to closed-form derivation, and make probabilistic predictions in high-dimensional spaces. This invited session brings together researchers working at the intersection of fatigue and fracture mechanics with machine learning, interpretable AI, generative methods and data-driven modeling. This session will highlight contributions spanning the full spectrum of this intersection: surrogate and reduced-order models for cycle-by-cycle fatigue evolution; neural operators and foundation models for crack propagation and damage field prediction; symbolic regression and equation-learning approaches that recover interpretable fatigue laws from simulation and/or experimental data; probabilistic and Bayesian frameworks for multiscale bridging and uncertainty quantification in fracture predictions; diffusion and flow matching models and multimodal data fusion strategies that bridge experiments and simulations into shared representations of damage mechanisms. Contributions addressing model validation, generalizability across material systems and loading conditions, and the challenge of limited or noisy experimental data are particularly encouraged. This session aims to foster dialogue across computational mechanics, materials science, statistical physics, and machine learning communities around a shared question: how do we build predictive, interpretable, and trustworthy models of fatigue and fracture that are grounded in physics yet empowered by data?
Organized by: R. Dingreville (Sandia National Laboratories, United States), K. Garikipati (University of Southern California, United States) and I. Srivastava (Lawrence Berkeley National Laboratory, United States)
Keywords: Computational damage mechanics, data-driven model discovery, Fracture and failure, Multiscale Modeling, Scientific machine learning
The study of the mechanisms governing fatigue crack initiation and propagation continues to represent a major research challenge, owing to its substantial implications for the economic impact, safety, and sustainability of both existing and future engineering structures and mechanical components. This session addresses recent advances in fatigue and plasticity modelling of metallic materials and structures. Particular attention is placed on the role of local plasticity in fatigue crack initiation, damage accumulation and the transition from low-cycle to high-cycle fatigue regimes. Contributions are invited on cyclic hardening and softening, ratcheting, mean stress relaxation, local strain approaches, crystal plasticity, finite element implementation of constitutive models, and coupled crack initiation-propagation strategies. The session welcomes both experimental and numerical studies, including material parameter identification, validation under strain-controlled cyclic loading, local fatigue indicators, residual stress effects, and applications to notched components, welded joints and structural details. The objective is to discuss how advanced cyclic plasticity models and local fatigue approaches can improve fatigue life prediction beyond conventional nominal stress-based methods.
Organized by: R. Fincato (Padua University, Italy) and S. Tsutsumi (The University of Osaka, Japan)
Keywords: Computational Modelling, Constitutive modelling, fatigue, Finite Element Method (FEM)
Predicting crack development and fatigue in structural components is often challenging in the presence of defects, residual stresses, heterogeneity and coupled environmental effects. These challenges are especially relevant to increasingly complex, multiphysical material systems, including additively manufactured metals and alloys, energy-storage materials, high-performance composites and other multifunctional material systems. Computational and physics-based artificial intelligence models are increasingly utilized to provide powerful and robust predictive capabilities to capture these multiphysics phenomena and failure across multiple spatial and temporal scales. The aim of this mini-symposium is to bring together recent advances in physics-based, data-driven and hybrid computational approaches for the modelling of fracture and fatigue in structures. Particular emphasis will be put on approaches including, but not limited to, variational and phase-field formulations for brittle, ductile, cohesive and fatigue fracture, as well as multiphysics phenomena such as hydrogen-assisted cracking, corrosion fatigue, thermal fracture and electro-chemo-mechanical degradation in energy materials. We also strongly encourage contributions at the intersection of computational fracture mechanics and machine learning approaches, such as physics-informed neural networks (PINNs), graph neural networks (GNNs), and neural operators (e.g., DeepONet), as well as their use in data-driven constitutive modeling and reduced-order or surrogate modeling of fracture processes.
Organized by: E. Azinpour (Faculty of Engineering, University of Porto, Portugal), R. Darabi (Imperial College London, United Kingdom) and J. Cesar de Sa (Faculty of Engineering, University of Porto, Portugal)
Keywords: AI-based surrogate modeling, Data-driven modelling, fatigue, Fracture and failure, Hydrogen embrittlement
2300 High Strain Rate, Impact and Dynamic Failure
Large-scale numerical analyses are essential for the design, assessment, and optimisation of engineering structures across many industrial sectors. Reliable predictions require accurate modelling of nonlinear material and joint behaviour, including plasticity, damage, and fracture. This mini-symposium aims to highlight recent advances in computational and experimental–numerical methods for modelling materials and joints in large finite element analyses, where accuracy, robustness, and efficiency must be balanced. Contributions are invited on constitutive modelling, failure criteria, validation strategies, and industrial applications, particularly in transportation sectors such as aerospace, automotive, naval, offshore, and railway engineering. Papers addressing joining technologies, including mechanical fastening, welding, adhesive bonding, and hybrid joints, are also welcome. The session also seeks to bring together international researchers, engineers, and industrial practitioners to promote discussion, knowledge transfer, and future collaborations.
Organized by: M. Costas (SIMLab / NTNU, Norway), D. Morin (SIMLab / NTNU, Norway) and B. Erice (University of MondragĂłn, Spain)
Keywords: finite element analyses, industrial applications, large-scale modelling, Plasticity and failure
3100 Advanced Discretization Techniques
Simulations involving fracture, localization, shocks, and fluid-like plastic flow require a departure from traditional Lagrangian finite elements. Meshfree, peridynamic, and particle methods have proven successful here, using point-based approximations without fixed mesh connectivity. They offer a natural framework for problems that would otherwise require erosion, mesh-moving, and remeshing with traditional methods. Their approximations allow controllable continuity and completeness, offering flexibility in incorporating enrichments that capture the physics of various problems. Peridynamic methods can likewise model systems with or without discontinuities and share many features with meshfree and particle methods. All are versatile enough to naturally embed and couple machine-learning techniques. This session aims to promote collaboration among academia, government/national labs, and industry to share recent advances and lessons learned and to facilitate knowledge transfer. Contributions are solicited on all subjects related to these methods, including: -Fundamental theoretical and algorithmic developments, including mixed formulations and cell-based schemes -Integration with machine learning, AI, and data-driven analysis; Ritz and PINN approaches -HPC and efficient solvers, including multigrid -VR physics, real-time simulation, digital twins, visual computing, and reduced-order modeling -One-to-one comparisons of meshfree and mesh-based methods in challenging applications -Strong and weak discontinuities, e.g., strong shocks, crack propagation, and material interfaces -Industry and lab applications where these methods outperform conventional ones, such as natural disasters, manufacturing, additive manufacturing, and cardiovascular/biomechanics -Treatment of extreme distortion, fragmentation, contact, and impact -Novel discretizations such as wavelets, multiresolution, and enriched methods for composite microstructures, fluid-structure interaction, multi-phase, and multi-scale problems -Enhanced collocation and integration, including variationally consistent integration, strong-form collocation, and solutions to rank, kernel, and zero-energy mode instability -Parallel computing, including GPUs, APUs, heterogeneous architectures, and large-scale simulations
Organized by: M. Hillman (Karagozian and Case, Inc., United States), J. Chen (UC San Deigo, United States), S. Chi (University of Illinois Chicago, United States) and P. Seleson (Oak Ridge National Laboratory, United States)
Keywords: additive manufacturing, advanced discretization techniques, contact mechanics, damage, data-driven modeling, digital twins, enrichment methods, extreme deformation, finite strain, fluid-structure interaction, fracture mechanics, fragmentation, GPU computing, high-performance computing, impact, machine learning, material point method, multiphysics, particle methods, peridynamics, physics-informed neural networks, plasticity, real-time simulation, reduced-order modeling, reproducing kernel methods, shock dynamics, SPH, stabilization, strong discontinuities, strong-form collocation, zero-energy modes, Cardiovascular Biomechanics, Computational damage mechanics, Computational Mechanics, Computational Modelling, Coupled plasticity and damage, damage, Damage and Fracture, Experimental Validation, GPU Computing, meshfree, meshfree methods, Particle Methods, Strain Localization
The simulation of complex physical phenomena in heterogeneous media presents major computational challenges. Traditional discretization methods frequently encounter limitations related to mesh generation and distortion, artificial directional bias, topological re-meshing, and prohibitive computational costs associated with fine-scale and repetitive Representative Volume Element evaluations. This Minisymposium focuses on the integration of novel spatial discretization schemes and complexity reduction strategies to construct high-fidelity, computationally efficient formulations for non-linear physical processes across spatial and temporal scales. Advanced schemes for complexity reduction include, for instance, Model Order Reduction (MOR) and Reduced Basis (RB) methods, as well as low-rank and tensor decompositions for high-dimensional parametric spaces. Their field of application overlaps with numerical acceleration paradigms such as quantum computing, data-assisted, surrogate, and operator-learning multiscale solvers. Novel discretization methods encompass unfitted and cut-background mesh methods, polyhedral elements with hanging nodes, advanced interface formulations, Virtual Element Methods (VEM, CutVEM), and Hybrid High-Order (HHO) formulations on general polytopal meshes. Contributions to this Minisymposium will showcase the application of these or any other novel discretization and complexity reduction frameworks to: - regularized fracture and damage models (nonlocal, peridynamics, phase-field, cohesive zone modeling). - Elasto-plasticity, viscoplasticity, shear band formation, and plastic dissipation. - Multiscale composite failure, delamination, grain-boundary sliding in polycrystals, thermo-plasticity, chemo-mechanical degradation, and real-time failure assessment within digital twin architectures.
Organized by: E. Benvenuti (University of Ferrara, Italy), G. Manzini (LANL, United States) and N. Sukumar (UC Davis, United States)
Keywords: advanced discretization techniques, complexity reduction methods, Damage and Fracture, Multiscale Modeling, plasticity
3200 Contact Mechanics and Interfaces
Over the past decade, Nitsche-based formulations have emerged as a powerful alternative to classical penalty and Lagrange-multiplier methods for contact mechanics. By enforcing interface constraints weakly, these methods provide consistent discretizations without introducing additional contact unknowns or imposing compatibility conditions between contacting meshes. This talk reviews recent developments in the mathematical analysis, numerical treatment, and industrial application of Nitsche methods for frictional contact problems. Particular attention will be given to large-scale three-dimensional simulations involving Coulomb friction, elastoplastic constitutive laws, and non-matching high-order finite-element discretizations. The presentation will first examine recent theoretical results concerning the well-posedness, stability, and convergence of Nitsche formulations for unilateral contact and frictional problems. The algorithmic challenges associated with nonlinear constitutive updates, semismooth Newton methods, projection operators, and monolithic solution strategies will then be discussed. In particular, the interaction between local constitutive integration schemes and global contact solvers remains a central issue for the robustness and efficiency of large-scale computations. A significant part of the talk will focus on industrial applications and numerical results obtained for challenging three-dimensional contact configurations. The discussion will emphasize convergence properties, robustness with respect to mesh non-conformity, and comparisons with classical penalty and augmented-Lagrangian approaches. Finally, the talk will outline several perspectives toward a rigorous mathematical and algorithmic framework for coupling Nitsche contact formulations with nonlinear elastoplastic constitutive operators.
Organized by: C. Pozzolini (FRAMATOME, France), M. Pignol (Framatome, France), Y. Renard (INSA LYON, CNRS, France) and B. Souloumiac (Framatome, France)
Keywords: Computational Modelling, Computational Plasticity, finite element analyses, Finite Element Method (FEM), non-smooth and optimization-based return mapping
Repeated impacts with friction occur in many industrial systems subjected to vibrations and mechanical clearances, and lead to irreversible phenomena such as wear, damage accumulation and surface degradation. Predicting their long-term effects remains challenging, because the underlying impact–friction laws are difficult to identify from observed mechanical responses. Identifying wear laws directly from experimental data is a very difficult task. We therefore start with a better-understood phenomenon, impact combined with friction, and aim to identify the impact–friction law from the trajectories of the system, as a first step toward the identification of wear. The study case is a viscoelastic Timoshenko beam impacting two rigid obstacles with friction. The data are not experimental: they are generated by numerical simulations driven by a known impact–friction law, so that the identified law can be compared with the reference one. Since this inverse problem is highly nonlinear and ill-posed, we restrict attention to impact–friction models deriving from a dissipation potential, which we seek to recover in an explicit and interpretable form. In doing so, we aim to extend to this nonsmooth setting data-driven identification methods developed for smooth dynamical systems, such as SINDy.
Organized by:
Keywords: Data-driven modelling
Interfaces play a fundamental role in many engineering systems. Although a lot of progress has been achieved, the research field is still growing and diversifies in many directions This session is devoted to recent developments on the various aspects of contact mechanics: • Interface behavior: unilateral contact, friction, adhesion, viscosity, fretting, wear, peeling, rolling contact, contact in biomechanics, fluid flow in contact interface • Computational models: multilevel approaches (molecular and nano-mico-macro models), multi-physics (thermo, piezo, etc.), coupled multi-field formulations, fractal surface characterization, homogenization, bi-potential • Computational methods: fast solvers, multi-grid, isogeometric analysis, NURBS, contact detection, third media contact • Dynamics of structures and of rigid bodies, instabilities • Mathematical progress • Industrial applications involving interface and contact conditions. Besides presentations of new results and new contributions to the understanding of contact mechanics, this session will provide an opportunity to discuss and exchange ideas on the various topics related to contact mechanics in science and engineering.
Organized by: P. Wriggers (Leibniz University Hannover, Germany), M. Puso (Lawrence Livermore National Laboratory, United States) and G. Zavarise (Politecnico di Torino, Italy)
Keywords: contact mechanics, Finite Element Method (FEM), interfaces, virtual element method
3300 Model Reduction, High-Performance Computing and Digital Twins
Simulating elastoplastic, crystal-plasticity, and other strongly nonlinear behaviour increasingly involves large-scale, computationally demanding problems — finite-strain kinematics, history-dependent constitutive updates, contact, and adaptive discretization — arising in both finite-element (FE) and mesh-free methods such as Smoothed Particle Hydrodynamics (SPH), and challenging conventional CPU implementations. Graphics Processing Units (GPUs), built on the Single Instruction, Multiple Threads (SIMT) paradigm, offer substantial acceleration, yet raise open questions on parallelizing constitutive integration, point-location search, and nonlinear solvers, as well as on exploiting automatic differentiation for inverse characterization and digital-twin applications. Recent advances span multi-GPU material point methods for extreme deformation, differentiable GPU-accelerated finite-element and crystal-plasticity solvers achieving substantial speedups, and GPU-resident mesh and point-location algorithms for particle tracking and contact detection. This rapidly maturing landscape — spanning computational mechanics, high-performance computing, and machine learning — underscores GPU computing's promise not only for faster forward simulations but also for optimization, uncertainty quantification, and digital-twin workflows increasingly reliant on GPU-accelerated reduced-order and surrogate models for real-time simulation. This Invited Session gathers researchers developing GPU-accelerated methods for computational plasticity to share strategies, benchmarks, and open challenges, such as: matrix-free/assembly-free GPU FE solvers, GPU-parallel constitutive integration, GPU-accelerated SPH and particle methods for large-deformation and impact plasticity, differentiable and AI-augmented frameworks, GPU-accelerated contact and mesh-search algorithms, reduced-order and surrogate modeling for real-time simulation, and next-generation HPC strategies, while comparing scalability across frameworks and identifying transferable algorithmic strategies for memory management, data locality, and heterogeneous CPU-GPU architectures. The session aims to clarify best practices, identify common bottlenecks, and foster collaborations advancing robust computational methods for nonlinear solid mechanics.
Organized by: A. Hashemi (CIMNE, UPC, Spain) and N. Dialami (CIMNE, UPC, Spain)
Keywords: Computational Plasticity, GPU Computing, High-Performance Computing (HPC), Reduced-Order and Surrogate Modeling
3400 Data-Driven Modelling, Machine Learning and AI
This mini-symposium brings together researchers from mechanics, applied mathematics, and related engineering disciplines to explore the integration of machine learning into computational mechanics. The focus is on leveraging data-driven methods to enhance modeling, simulation, and design across a wide range of scales and physical phenomena. Particular emphasis is placed on physics-based machine learning approaches that incorporate physical principles and constraints to improve consistency, interpretability, and generalization. We welcome theoretical developments, algorithmic innovations, and application-oriented studies that utilize machine learning to advance understanding and computation in mechanics. The goal is to foster interdisciplinary exchange and highlight cutting-edge advances that push the boundaries of computational mechanics through intelligent, physically grounded learning strategies.
Organized by: F. Aldakheel (Leibniz Universität Hannover, Germany), E. Cueto (Universidad de Zaragoza, Spain), S. Sun (Columbia University, United States), M. Bessa (Brown University, United States), Y. Heider (Universität Kassel, Germany), O. Weeger (TU Darmstadt, Germany) and D. Kochmann (ETH Zurich, Switzerland)
Keywords: Scientific machine learning
Driven by recent advances in scientific machine learning, operator learning is emerging as a powerful framework for discovering and representing complex material behavior while enabling accelerated prediction and simulation. Material discovery and model prediction remain bottlenecks in engineering design: accurately capturing how materials respond across varying loading conditions, histories, parameters, and microstructures typically requires extensive iterative experiments and simulations. Operator-learning approaches overcome this by learning families of material responses instead of isolated constitutive relations for individual cases. Combined with physical constraints—objectivity, material symmetries, thermodynamic principles, conservation laws, and evolution structures—learned operators enable both rapid discovery of transferable material representations and fast, efficient forward prediction. This mini-symposium focuses on path- and rate-dependent materials (viscoplasticity, viscoelasticity, creep, damage, stress relaxation), where history dependence is central. We convene researchers from computational mechanics, materials science, and scientific machine learning to exchange developments in physics-constrained operator learning for material discovery and accelerated simulation. Key Topics: • Physics-constrained operator learning for material discovery • Identification of constitutive operators from experimental and computational data • Learning of history-dependent and path-dependent material response • Discovery of free energies, dissipation mechanisms, internal variables, and evolution operators • Neural operators for microstructure–property and process–structure–property relations • Multiscale and microstructure-sensitive operator learning • Differentiable simulation and inverse material characterization • Accelerated prediction and forward simulation: surrogate models, digital twins, parametric studies • Uncertainty quantification, interpretability, and generalization of learned operators • Applications to nonlinear, inelastic, multiphysical, and evolving materials The primary focus is on methods combining learned mappings with physical structure to identify transferable material representations, enable efficient forward prediction across loading paths and scales, and accelerate both model discovery and real-time simulation.
Organized by: S. Rezaei (Access e.V., Germany), H. Holthusen (Computational Mechanics in Medicine, RWTH Aachen University, Germany), M. Flaschel (Institute of Applied Mechanics, University of Erlangen–Nuremberg, Germany), J. Simon (Computational Applied Mechanics, School of Architecture and Civil Engineering, Bergische Universität Wuppertal, Germany), M. Kästner (Institute of Solid Mechanics, TU Dresden, Germany), E. Kuhl (Department of Mechanical Engineering, Stanford University, United States) and D. Raabe (Max Planck Institute for Sustainable Materials, Germany)
Keywords: data-driven model discovery, Data-driven modelling, Scientific machine learning
3500 Agentic AI in Computational Mechanics
The emergence of large language models and agentic AI frameworks is reshaping the way engineers design, model, and simulate. While AI has already been significantly and widely applied in computational mechanics, from surrogate modelling and topology optimization to data-driven constitutive laws, the advent of autonomous agent systems opens qualitatively new possibilities. Agentic AI is not merely another tool within existing workflows; it represents a shift toward systems that can reason, plan, and act across complex multi-step tasks with minimal human intervention. Agentic AI offers new possibilities across materials modelling and characterization. By combining the generative and reasoning capabilities of LLMs with physics-based FEM engines and reduced-order models, these systems can automate and orchestrate modelling workflows that previously demanded deep expert knowledge, simplifying model setup and configuration, enabling new approaches to the development of material subroutines and user-defined models, and assisting in parameterization, geometry creation, mesh generation, solver selection, and post-processing. Beyond workflow automation, autonomous agents open new frontiers in materials characterization: driving inverse identification of material parameters from experimental data, interpreting multimodal geometry inputs for FEM preprocessing, and interfacing with simulation environments through natural language, lowering barriers for non-specialist users while accelerating the expert throughput. Contributions to this session will cover agentic architectures for simulation, LLM-driven modelling assistants, automated characterization pipelines, automated end-to-end modelling workflows, uncertainty management, and trust in autonomous simulation systems.
Organized by: J. Gomez-Sanchez (ADVANCED MATERIAL SIMULATION, Spain) and B. Erice (University of Mondragon, Spain)
Keywords: Agentic-AI, Large language models (LLMs), Simulation AI-agent, Workflow orchestration agent
Artificial intelligence is advancing beyond surrogate modeling and is beginning to take an active role across the simulation workflow, configuring solvers, coupling codes, and checking results. A central challenge to the broader adoption of AI in predictive computational mechanics is demonstrating its reliability. This Invited Session is devoted to AI methods that operate within simulation workflows for nonlinear materials, damage, fracture, and coupled multiphysics problems, together with the verification and validation practices needed to establish confidence in their results. Contributions are invited on: (i) agentic and language-model-based frameworks that set up, execute, couple, and verify simulations across one or several codes, ranging from early prototypes to mature systems; (ii) physics-informed learning, neural operators, surrogate models, and learned world models for representing and accelerating complex physical systems, including phase-field, peridynamics, and continuum damage formulations, as well as degradation processes such as corrosion and fatigue; (iii) machine learning for coupling, adaptivity, and information transfer between local and nonlocal descriptions; and (iv) verification, validation, and uncertainty quantification for AI-enabled simulation, including benchmarks, reproducibility, and computational cost. The session aims to connect advances in AI-enabled simulation with rigorous methods for assessing their performance and reliability, supporting credible computational workflows for complex mechanics and multiphysics problems.
Organized by: A. Hermann (Hamburg University of Technology, Germany) and A. Shojaei (Helmholtz-Zentrum Hereon, Germany)
Keywords: Agentic-AI, AI, AI-based surrogate modeling, Computational damage mechanics, Damage and Fracture, Multiphysics modeling, physics-informed modeling, Scientific machine learning
3600 Inverse Problems, Optimization and Uncertainty Quantification
Inverse design of heterogeneous materials, microstructures, and architected solids is becoming an increasingly important topic in computational solid mechanics. The central challenge is to identify material architectures that achieve prescribed mechanical behavior while navigating high-dimensional design spaces, non-unique solutions, and expensive forward evaluations. Recent advances in artificial intelligence, data-driven modeling, and optimization methods offer new tools for addressing these challenges and for expanding the range of structures that can be explored and designed. This Invited Minisymposium will focus on strategies for efficient exploration and optimization of complex material architectures. Of particular interest are approaches that reduce the dependence on large datasets generated through costly high-fidelity simulations or experiments. Surrogate models, reduced-order representations, and neural operators can accelerate repeated structure–property evaluations, while active learning and Bayesian optimization can guide the selection of informative or promising candidates under limited computational or experimental budgets. Generative models offer a complementary route for addressing the non-uniqueness of inverse design by representing multiple candidate structures compatible with a desired response. Recent developments such as diffusion models and flow-matching approaches are particularly relevant for the generation of microstructures with prescribed effective properties or mechanical behavior. Learned latent and implicit representations, together with differentiable models and gradient-based optimization, further enable efficient search over complex geometric parameterizations. Contributions involving alternative learning-based, physics-guided, optimization-based, or hybrid strategies are equally encouraged. Applications may encompass composites, porous and stochastic microstructures, and architected materials designed for prescribed stiffness, anisotropy, nonlinear response, stability, energy absorption, or other effective properties. The minisymposium aims to bring together researchers in computational solid mechanics, material modeling, optimization, and machine learning to exchange recent advances and foster reliable, data-efficient, and physically consistent approaches to inverse material design.
Organized by: H. Danesh (Technische Universität Braunschweig, Germany), H. Eivazi (Technische Universität Braunschweig, Germany) and H. Wessels (Technische Universität Braunschweig, Germany)
Keywords: Data-driven modelling, Inverse Design, Optimization, Reduced-Order and Surrogate Modeling
4100 Metamaterials and Computational Material Design
Metamaterials are engineered structures designed to exhibit exceptional properties not found in naturally occurring materials. Their unique macroscopic behaviors—ranging from negative Poisson's ratios to acoustic cloaking and tunable electromagnetic responses—stem from their intricate microstructural architectures. As the geometric and functional complexity of these architectures grows, empirical trial-and-error approaches to their design have become increasingly unfeasible. Consequently, advanced numerical modeling is essential to accurately predict, control, and optimize their performance. This Mini symposium explores the critical role of numerical simulations in the design and analysis of architected metamaterials. To accurately capture the complex, synergistic phenomena that govern these materials across multiple scales, researchers must move beyond isolated physical models. We invite contributions that discuss novel computational methods, multiscale homogenization techniques, data-driven modeling, and the experimental validation of numerical frameworks for metamaterial design. Numerical simulations enable the design of metamaterials for highly specific, complex environments. The session will highlight the latest advancements in integrated modeling frameworks, including, but not limited to: • Large strain modeling • Plasticity and viscoplasticity in metamaterials • FSI models • Biomechanical implants • Electromechanical models • Thermomechanical models
Organized by: M. Carraturo (University of Pavia, Italy), S. Morganti (University of Pavia, Italy) and F. Auricchio (University of Pavia, Italy)
Keywords: Data-driven modelling, Multiscale modelling
Improving and accelerating materials development is a major goal in science and industry, as tailored and optimized materials are key enablers of technological innovation. Recent advances in modeling, simulation, and machine learning have brought the vision of designing materials from desired properties closer to reality. By tailoring chemical composition and microstructure, materials can be engineered to achieve specific functional and mechanical performance. Beyond performance enhancement, these capabilities also enable the development of materials that are better suited for recycling, more tolerant to contamination, and less dependent on critical or scarce raw materials, thereby contributing to a more sustainable and resilient materials economy. This inverse design process requires computational techniques to characterize and reconstruct material microstructures and to quantify and model the behavior of phases and constituents at the microscale. Computational homogenization provides a powerful framework for linking these microstructural features to effective material properties and for understanding the mechanisms governing complex material behavior across scales. Data analysis, machine learning, and reduced-order modeling offer new opportunities to exploit the vast amount of information generated by simulations and experiments. Surrogate models can efficiently approximate process–structure–property relationships and enable large-scale exploration of the design space. More importantly, they pave the way for true inverse design strategies, where desired material properties, performance targets, sustainability requirements, or manufacturing constraints are used as inputs to identify suitable microstructures, compositions, or processing routes. Topics of interest covered within this mini-symposium include but are not limited to: • techniques for exploration and inversion of process-structure-property linkages, • microstructure characterization and reconstruction, • inverse design approaches for micro-heterogeneous and architected materials, • machine learning and data-driven approaches to constitutive modeling and multiscale simulations.
Organized by: M. Kästner (TU Dresden, Germany), D. Balzani (Ruhruniversität Bochum, Germany), B. Xu (TU Darmstadt, Germany) and D. Raabe (MPI SusMat, Germany)
Keywords: Data-driven modelling, Inverse Design, Multiscale Modeling, Sustainability
Architected metamaterials achieve tailored mechanical performance via the geometry and hierarchy of their microstructure. Auxetic architectures are a prominent example, combining negative effective Poisson’s ratios with unusual deformation modes, high specific energy absorption, and tunable stiffness. For load-bearing applications, however, their decisive behavior often emerges beyond the elastic regime, where material nonlinearity interacts with geometric instability, contact, localization, damage, and fracture across several length scales. This Invited Session addresses computational methods for the nonlinear, inelastic, and failure-dominated response of auxetic and architected metamaterials, with particular emphasis on the link between microstructural design and structural performance. Contributions are invited on finite-strain elastoplasticity and viscoplasticity, acoustic performance, instability and pattern transformation, damage and fracture, dynamics and vibration including high-rate and impact response, multiscale homogenization and generalized continua, and coupled multiphysics phenomena. The session also welcomes topology and shape optimization, sensitivity analysis, uncertainty quantification, model reduction, and data-driven or physics-informed surrogate models for efficient exploration of large design spaces. Digital shadows and digital-twin concepts for model updating, prediction, and reliability assessment are likewise within scope. A central objective is to connect high-fidelity simulation with fabrication and experimental validation. Relevant topics therefore include additive and advanced manufacturing, process- and defect-informed modelling, digital image correlation, micro-CT-based characterization, and inverse identification. Application-oriented studies may address lightweight civil, flight, and space structures, vibration and wave control, impact protection, energy absorption, structural sensing, and energy harvesting. By bringing together computational mechanics, materials science, optimization, manufacturing, and experiments, the session aims to clarify how nonlinear mechanisms can be deliberately exploited rather than merely tolerated, enabling robust and verifiable architected metamaterial structures from unit-cell design to component scale.
Organized by: T. Ricken (University of Stuttgart, Germany) and S. Seyedpour (University of Stuttgart, Germany)
Keywords: Additive Manufacturing, Architected Materials, Auxetic Metamaterials, Computational Mechanics, Experimental Validation, Machine Learning, Meta Materials, Multiscale Modeling, Nonlinear Mechanics, Structural Optimization, Uncertainty Quantification
Architected mechanical metamaterials and metastructures are attracting increasing attention due to their ability to achieve combinations of properties that are difficult or impossible to obtain using conventional materials. Their unique performance derives from the interaction between geometry, material behaviour and structural response across multiple length scales, creating new possibilities for lightweight, multifunctional and damage-tolerant systems in aerospace, defence, space and advanced engineering applications. Recent advances in computational mechanics have significantly expanded the capability to analyse and design metamaterials exhibiting tailored mechanical behaviour, including controlled fracture and crack propagation, enhanced toughness, vibration attenuation through band-gap engineering, wave manipulation and functionally graded responses. Furthermore, the maturation of additive manufacturing technologies has played a key enabling role in the metamaterials field, providing unprecedented freedom to fabricate architected materials with complex geometries and spatially varying properties. Numerical methodologies such as finite element formulations, multiscale modelling, topology optimization, phase-field approaches and reduced-order models are enabling the efficient exploration of increasingly complex architectures. In parallel, data-driven and physics-based artificial intelligence techniques are emerging as powerful tools to accelerate design processes while preserving physical consistency. Physics-informed machine learning, surrogate modelling, digital twins and hybrid computational frameworks are providing new opportunities for optimisation, uncertainty quantification and real-time prediction of metamaterial performance. This Invited Session aims to bring together researchers working on the modelling, simulation, optimisation and experimental validation of architected metamaterials and metastructures. Contributions addressing computational methods for fracture control, nonlinear behaviour, multiscale analysis, vibration attenuation, wave propagation, topology optimisation, additive manufacturing-oriented design and physics-based AI are particularly welcome. The session seeks to foster interdisciplinary discussion between computational mechanics, materials science and artificial intelligence communities, highlighting emerging challenges and opportunities for next-generation engineering structures.
Organized by: F. Montåns (Universidad PolitÊcnica de Madrid, Spain), J. Benítez Baena (Universidad PolitÊcnica de Madrid, Spain), J. García-Martínez (Instituto Nacional de TÊcnica Aeroespacial, Spain) and Á. García Juårez (Universidad de Castilla La Mancha, Spain)
Keywords: Fracture Mechanics, Mechanical Metamaterials, Physics-Informed Artificial Intelligence, Topology Optimization, Wave Propagation
4200 Non-metallic materials: Polymers, Composites and Soft Materials
The increasing use of composite materials in lightweight and high-performance structures has intensified the need for reliable numerical tools for the prediction of failure and long-term durability. Due to their anisotropic and heterogeneous nature, damage and failure mechanisms in composites involve complex interactions between matrix cracking, fibre degradation, interface debonding, and delamination across multiple scales, under both monotonic and cyclic loading conditions. The predictive simulation of these mechanisms remains particularly challenging due to the strong coupling between intralaminar and interlaminar damage processes, material anisotropy, path dependency, and the interaction between multiple length scales, ranging from constituent-level degradation to the structural response. Particular interest is devoted to models capable of capturing the interaction between different failure mechanisms under complex loading paths, including multiaxial stress states, variable amplitude loading, and coupled thermo-mechanical effects. This invited session aims to gather recent advances in the computational modelling of failure, delamination, and fatigue in composite materials and structures. Contributions on constitutive modelling, multiscale approaches, interlaminar damage and delamination, fatigue damage evolution, life prediction methodologies, numerical implementation, and experimental validation are particularly welcome. Special attention will be given to advanced numerical methodologies, including finite element formulations, cohesive zone and phase-field approaches, multiscale homogenization techniques, reduced-order models, and data-driven or hybrid computational strategies. Contributions addressing virtual testing, model calibration and validation, uncertainty-aware life prediction methodologies, as well as the influence of manufacturing-induced defects, porosity, residual stresses, and geometrical imperfections, are also encouraged. Applications to engineering structures in aerospace, automotive, civil, and energy sectors, including design, durability assessment, and certification-oriented virtual simulations, are particularly welcome.
Organized by: S. JimÊnez (Universitat Politècnica de Catalunya, Spain) and A. Cornejo (Universitat Politècnica de Catalunya, Spain)
Keywords: composite structures, damage, delamination, failure, fatigue
In this study, we investigated the effect of adding cork powder on the mechanical behavior of a hybrid fiber–metal laminate (HFML) made of glass fibers, cork-filled epoxy resin, and aluminum sheets as outer layers. The preparation of the FMLs started with the production of cork powder collected in the Dellys region (Boumerdes, Algeria). After cutting, boiling to remove impurities, drying, and grinding, a powder smaller than 200 µm was obtained and then treated with a NaOH solution (2 wt.%) to increase roughness and improve adhesion with the epoxy. At the same time, the aluminum sheets were prepared by mechanical polishing and alkaline treatment (NaOH 5 wt.%, 40 °C) to enhance surface roughness and strengthen fiber–metal adhesion. Three types of laminates were then produced by hand lay-up: FML-0 (without cork), FML-3 (3 wt.%), and FML-6 (6 wt.%). The prepared specimens were tested under three-point bending and Charpy impact. The results showed a modest improvement in maximum specific flexural strength for FML-3 (97.55 MPa/(g/cm³)) and a decrease for FML-6 (74.75 MPa/(g/cm³)) compared to FML-0 (95.93 MPa/(g/cm³)). On the other hand, a reduction in specific stiffness was observed for the cork-filled FMLs. For the impact tests, FML-3 showed a significant improvement in absorbed energy (3.92 J), specific absorbed energy (1.82 J/(g/cm³)), and impact strength (12.54 J/cm²) compared to the reference (3.64 J, 1.60 J/(g/cm³), and 12.28 J/cm², respectively), while FML-6 showed degradation (3.28 J, 1.58 J/(g/cm³), and 9.05 J/cm², respectively). These results suggest that a moderate cork powder content (3 wt.%) offers the best balance between lightness, flexibility, and energy absorption in hybrid fiber–metal laminates (HFMLs).
Organized by: F. Cherradi (École Militaire Polytechnique, Algeria)
Keywords: composite structures, Damage and Fracture, polymers
4300 Biomechanics
Cardiovascular biomechanics increasingly relies on the integration of mechanistic models with heterogeneous sources of experimental, imaging, and clinical data. This interplay is crucial to better characterize cardiovascular tissues, identify subject-specific properties, improve model predictions, and ultimately bridge fundamental biomechanical understanding with clinically relevant applications. Advances in data-driven methodologies, machine learning, inverse methods, and reduced-order modeling are providing new ways to extract information from data and incorporate it into biomechanical models. Rather than replacing physics-based descriptions, these approaches can complement established modeling frameworks by supporting parameter identification, constitutive characterization, model personalization, uncertainty quantification, and efficient prediction. At the same time, their integration raises important methodological questions concerning physical consistency, interpretability, generalization, and the effective combination of data with prior mechanical knowledge. This Invited Session aims to bring together researchers working at the interface between cardiovascular biomechanics, data integration, and hybrid modeling methodologies. The session will provide a forum to discuss recent developments and emerging approaches for combining data and models across different cardiovascular applications and spatial and temporal scales. Topics of interest include data-informed and machine-learning-enhanced biomechanical modeling, integration of experimental, imaging, and clinical data, inverse problems and parameter identification, constitutive modeling and tissue characterization, patient-specific modeling, surrogate and reduced-order models, uncertainty quantification, multiphysics approaches, and predictive modeling of cardiovascular function, adaptation, and disease. The session seeks to foster interactions between established biomechanical modeling approaches and emerging data-centered methodologies, highlighting their current capabilities as well as the challenges toward more reliable, interpretable, and predictive models in cardiovascular biomechanics.
Organized by: A. Mastrofini (University of Rome Tor Vergata, Italy), M. Marino (University of Rome Tor Vergata, Italy) and C. Gasser (KTH Royal Institute of Technology, Sweden)
Keywords: Cardiovascular Biomechanics, Data Integration, Data-driven modelling, Physics-Based Modeling
Computational biomechanics and robust numerical methods support the investigation of disease mechanisms and treatment strategies in living active systems. Multiphysics and multiscale models combine mechanics with biochemistry, electrophysiology and transport processes to support predictive simulations. These models connect subcellular activity with macroscopic continuum mechanics and account for tissue responses to mechanical stimuli, growth and remodelling. Interactions between functional units must also be considered to explain tissue and organ behaviour. This minisymposium focuses on these challenges and extends its scope to patient-specific digital twins of individual organs and coupled organ systems. The emphasis is on representing biological mechanisms, synchronizing models with longitudinal observations, and establishing the credibility of physiological predictions. Digital twins repeatedly update states, parameters, and, where necessary, model structure from individual data and assess predictions against subsequent observations. This requires handling heterogeneous, incomplete and asynchronous measurements while maintaining consistent model states and physiological coupling conditions. Representative applications include the heart, brain, lungs, liver, and kidneys and their functional interactions. Topics of interest include: • Active tissue mechanics, mechanobiology, growth and remodelling • Multiscale and multiphysics modelling • Multimodal data assimilation, personalisation and identifiability • Conservative organ coupling and adaptive multifidelity co-simulation • Scalable solvers, model reduction and physics-informed surrogates • Verification, independent validation and uncertainty propagation • Reproducible benchmarks, interoperable software and FAIR data and models We welcome interdisciplinary contributions, combining mechanical modelling, numerical methods, data science, and clinical applications. Contributions from experimental biomechanics are also welcome, since they provide data for mechanistic understanding, parameterisation and validation. The objective is to compare approaches for efficient and physiologically consistent predictions, including quantified aleatoric and epistemic uncertainties and explicit limits of applicability.
Organized by: T. Ricken (University of Stuttgart, Germany), O. RÜhrle (University of Stuttgart, Germany) and S. Budday (Friedrich-Alexander-Universität Erlangen-Nßrn, Germany)
Keywords: Biomechanics, Constitutive modelling, digital twins, Multiphysics modeling, Multiscale Modeling, Uncertainty Quantification
4400 Geomechanics and Porous Media
Geomaterials, including soils, rocks, concrete, snow and ice, are porous, heterogeneous and often multiphase media that exhibit nonlinear deformation and failure under natural and engineering conditions. Their behaviour is governed by plasticity, damage, fracture, strain localization and coupled thermal, hydraulic, mechanical and chemical processes, which are central to geohazards, underground energy storage, carbon sequestration, tunnelling, foundations, slopes and resilient geo-infrastructure. Predicting these processes requires advanced computational plasticity frameworks capable of capturing material nonlinearity, multiscale failure, multiphysics coupling and field-scale complexity. This invited session aims to bring together researchers and practitioners to discuss theoretical, numerical, experimental and data-driven advances in computational plasticity for geomaterials and geosystems, with emphasis on robust constitutive models and computational methods for understanding and predicting deformation and failure across scales. Topics of interest include, but are not limited to: elastoplastic, viscoplastic and hypoplastic constitutive modelling; strain localization, shear banding, compaction banding, damage and fracture in geomaterials; coupled THM and THMC modelling of porous media; multiscale and micromechanical approaches; phase-field, peridynamic, meshfree, particle-based and hybrid numerical methods; model calibration, verification and validation using laboratory and field data; uncertainty quantification, inverse analysis and probabilistic modelling; machine-learning-enhanced constitutive models, surrogate models and model updating; digital twins and computational frameworks for monitoring and field-scale applications. The Mini-Symposium will provide an interdisciplinary platform for exchanging recent advances between computational mechanics, plasticity theory, geomechanics, data science and engineering practice. It aims to advance reliable, predictive and efficient computational tools for safer, more sustainable and resilient interactions with the Earth’s subsurface systems.
Organized by: Y. Wang (Universität fßr Bodenkultur Wien, Austria), J. Choo (Seoul National University, Republic of Korea), R. Borja (Stanford University, United States) and W. Wu (Universität fßr Bodenkultur Wien, Austria)
Keywords: Computational Geomechanics, Damage and Fracture, Data-driven modelling, Geomaterials and geosystems, Multiphysics coupling, Multiscale modelling, Strain Localization
Geophysical materials such as ice, snow, rock, and frozen soils play a central role in some of the most consequential natural hazards and climate-related processes of our time. The fracture of glaciers and ice sheets critically determines rates of sea-level rise, avalanche release mechanisms pose direct risks to mountain communities and infrastructure, and the failure of rock slopes and permafrost threatens both natural and built environments. Despite their importance, the computational modelling of failure in geophysical materials remains an open challenge, owing to their complex, rate- and temperature-dependent constitutive behaviour, multi-phase nature, and the vast range of spatial and temporal scales involved. The objective of this invited session is to address modelling issues encountered in the mechanics and computational modelling of fracture and failure in geophysical materials, and to provide a platform for discussion of recent advances and open challenges in this field. Contributions addressing any aspect of failure in these materials are welcome, including, but not limited to, ¡ Damage and fracture mechanics for geophysical materials ¡ Constitutive modelling of multi-physically coupled processes in geophysical materials ¡ Multi-scale approaches and data-driven methods for failure in geophysical materials The session aims to foster an interdisciplinary exchange between computational mechanics, geosciences, and natural hazard research, and to contribute to the development of predictive tools that address environmental and societal challenges.
Organized by: A. Tsakmakis (Technical University of Darmstadt, Germany), A. Trauth (Technical University of Darmstadt, Germany) and R. MĂźller (Technical University of Darmstadt, Germany)
Keywords: Computational damage mechanics, Constitutive modelling, Damage and Fracture, Data-driven modelling, Fracture and failure, Multiscale modelling
Porous materials are found in a wide range of engineering applications, including civil engineering, mechanical engineering, and bioengineering to name few. They consist of a deformable porous solid skeleton and one or more fluids occupying the pore space. Several continuum theoretical frameworks have been widely used to describe coupled multiphysics processes in these materials. However, under certain environmental or engineered conditions, advanced modelling approaches, including plasticity and fracture, become necessary. Processes such as ice lens formation, salt crystallization, hydraulic fracturing, and damage localization in porous material can lead to the initiation and propagation of discontinuities within an initially continuous porous medium. These phenomena involve strong localization effects and pose major challenges for both theoretical understanding and computational modeling. This MS aims to bring together researchers working on theoretical and computational aspects of coupled multiphysics processes in porous materials with evolving discontinuities. Topics of interest include multiphysics fracture initiation and propagation, phase change, localization phenomena, and coupled hydro-mechanical, thermo-hydro-mechanical, and chemo-mechanical processes, validation of coupled process as well as multiscale approaches. The MS particularly welcomes contributions on topics, including, but not limited to: • multi-scale modeling of porous media • phase-field models for fracture and phase change • interface formulations • material point methods, and other novel numerical approaches. • inverse design of porous metamaterials • recent techniques applied to study porous media coupled behaviors, such as physics-guided neural networks.
Organized by: R. WILLIAMS MOISES (University of Kassel, Germany), Y. HEIDER (University of Kassel, Germany), F. ALDAKHEEL (Leibniz University Hannover, Germany), G. XOTTA (University of Padova, Italy) and B. MARKERT (RWTH Aachen University, Germany)
Keywords: Localization Phenomena, Multiphysics Coupling, Multiscale Modeling, Machine Learning in Porous Media, Porous Media
5100 Multi-Physics and Coupled Problems
Environmentally assisted damage arises from the combined effects of mechanical loading and aggressive environments on susceptible materials. The interaction between these factors can significantly reduce the degradation resistance of materials, promoting crack initiation and premature failure with potentially catastrophic consequences. Localized corrosion, stress corrosion cracking, and hydrogen embrittlement are among the most widely recognized forms of environmentally assisted material degradation. These complex and often hard to assess destructive failure mechanisms affect metallic engineering components across a wide range of industries. Computational modeling has emerged as an effective method for assessing the degradation resistance of engineering materials and as a powerful tool for designing components against environmentally assisted failure. Recent advances in computational power have enabled the development of sophisticated multiphysics models and hybrid digital methods for predicting environmentally assisted damage. These approaches can track the evolution of the material-environment interface and account for the synergetic interactions between mechanical loading and environmental conditions that drive material degradation. They overcome the long-standing challenge of simulating coupled degradation processes that evolve across multiple interacting spatial and temporal scales. Advances in coupled modeling frameworks now enable the natural simulation of complex phenomena such as pit-to-crack transition and crack propagation in arbitrary domains. This session will feature presentations and discussions on the modeling of environmentally assisted damage (pure fatigue is excluded). It will showcase state-of-the-art computational approaches for simulating a range of degradation and failure mechanisms, including localized corrosion, stress corrosion cracking, and hydrogen embrittlement. The session will cover recent methodological developments and advances in the field through presentations by leading experts in environmentally assisted degradation and computational materials engineering.
Organized by: D. Hoeche (Helmholtz-Zentrum hereon GmbH, Germany) and S. Kovacevic (University of Oxford, United Kingdom)
Keywords: Computational damage mechanics, Corrosion, Data-driven modelling, Environmentally assisted damage, Hydrogen embrittlement, Localized damage, Multiphysics modeling, Stress corrosion cracking
Cracks and discontinuities play a fundamental role in all aspects of the behaviour of quasi-brittle materials such as concrete and rock, and their complete understanding and accurate modelling is, to date, still an open challenge. Beyond certain threshold of mechanical loading these materials may develop distributed micro-cracks, some of which may eventually coalesce into localized macro-cracks while the rest unload. But degradation may be caused as well by environmental loading, and the effect of the latter may be crucial because of the interaction effects between the environmental reactive transport phenomena such as moisture movement, radiation or chemical activity, and the mechanical response and the cracking. In a similar way, pressurized fluid flow through the material porosity and the cracks, may generate crack propagation and lead to strong coupling. A proper understanding and assessment of all these phenomena require multi-physics sound models, which include mechanical and the flow/diffusion/reaction/transport behavior, in general considering the time variable as well. The highly heterogeneous nature of quasi-brittle materials may also play a key role in deterioration studies. In that case, multi-scale models, such as micro- or meso-mechanical models, where the material heterogeneities are explicitly represented, may help to considerably simplify the constitutive description, at the cost of a higher computational effort, which in turn may require parallel high-performance computing. This mini-symposium is intended to gather contributions on all those and related topics, including mainly numerical modeling, but also related experimental and theoretical studies. Classical models based on a continuum or a discrete approach, as well as more recent techniques such as XFEM, phase field, etc. are all welcome.
Organized by:
Keywords: Fracture and failure, Multiphysics Coupling, Multiscale Modeling
5200 Manufacturing and Industrial Applications
Manufacturing processes are present in a wide range of engineering fields, such as automotive, aerospace, aeronautical, naval industries, among others. Many of these processes rely on the elastoplastic behavior of metallic materials. Therefore, understanding the resistance and hardening characteristics of materials is crucial for optimizing these processes. Therefore, this “Invited Minisymposia” welcomes contributions focused on the study of manufacturing processes involving the behavior of metallic alloys in the plastic regime, including metal cutting, sheet and bulk metal forming, characterization of materials from additive manufacturing, the use of AI for manufacturing process optimization, tube and sheet bending, among others. Contributions related to the analysis of repurposing natural gas pipelines for the transportation and distribution of hydrogen will also be considered.
Organized by: L. Malcher (INEGI, Portugal), M. Gonçalves (INEGI, Portugal), L. Araújo (INEGI, Portugal), S. Miranda (INEGI, Portugal), O. Emadinia (INEGI, Portugal) and J. Pereira (INEGI, Portugal)
Keywords: Constitutive modelling, Coupled plasticity and damage, Fracture and failure, Manufacturing processes
This Invited Organized Session aims to provide a forum for researchers and scientists to present and discuss recent advances in the computational modeling of manufacturing and material processing using particle-based and meshless methods. The session will focus on state-of-the-art mathematical formulations, numerical methodologies, computational strategies, and industrial applications involving large deformations, complex multiphysics couplings, and evolving material interfaces. The objective of the session is to foster interdisciplinary exchanges and promote the development of robust and predictive computational approaches capable of addressing both fundamental and industrial challenges encountered in modern manufacturing and material processing technologies. Topics of interest include, but are not limited to: • Particle-based, meshless, and hybrid computational methods, including PFEM, SPH, DEM, MPM, PIC, LBM, FCM, EFG, and coupled FEM–particle approaches. • Computational modeling of manufacturing and material processing operations such as additive manufacturing, casting, solidification, forging, rolling, extrusion, stamping, deep drawing, leveling, superplastic forming, thixoforming, welding, friction stir welding, friction melt bonding, high-speed forming, metal deposition, and automated fiber placement. • Advanced mathematical formulations and numerical discretization techniques. • Nonlinear and multiphysics solution strategies and implementation aspects. • Stabilization techniques and multiscale computational approaches. • Constitutive modeling at the macro-, meso-, micro-, and multiscale levels, including high strain-rate and high-temperature behavior. • Contact mechanics, friction, and lubrication modeling. • Damage, fracture, failure, and crack propagation simulations. • Coupled thermomechanical, metallurgical, and multiphysics models. • Simulation and optimization of manufacturing process chains. • Large-scale computing, parallel algorithms, and high-performance computing for manufacturing simulations. The session welcomes contributions addressing theoretical developments, algorithmic innovations, verification and validation procedures, as well as industrial applications and benchmark studies demonstrating the capabilities of advanced computational methods for manufacturing and material processing problems.
Organized by: J. Ponthot (Liège University, Belgium), J. Carbonell (CIMNE, Spain) and A. Menzel (TU Dortmund, Germany)
Keywords: Manufacturing processes, multiphysics, Particle Methods
Additive Manufacturing (AM) is increasingly recognized as a transformative technology for fabricating complex components with enhanced performance and design flexibility. Processes such as Wire Arc Additive Manufacturing (WAAM), Directed Energy Deposition (DED), and Laser Powder Bed Fusion (LPBF) involve strongly coupled thermal, mechanical, and metallurgical phenomena across multiple spatial and temporal scales, posing significant challenges for modelling, simulation, and process control. This Invited Session aims to highlight recent developments in computational modelling, experimental characterization, and data-driven approaches for the analysis and optimization of metal AM processes. The focus is on advancing predictive capabilities at the part and component scales, enabling improved process understanding, reliability, and performance of additively manufactured structures. The session seeks to bring together researchers and practitioners from academia and industry to exchange ideas on robust and efficient simulation strategies, as well as their integration with experimental and monitoring techniques. Topics of interest include, but are not limited to: • Advanced numerical methods for part-scale process simulation. • Efficient time integration and discretization techniques for large-scale AM problems. • Reduced-order modelling and surrogate-based approaches. • Multi-physics and multi-scale frameworks for capturing process–structure–property relationships. • Thermo-mechanical and microstructural material modelling. • Simulation and mitigation of residual stresses, distortion, and defects. • Integration of simulation with in-situ monitoring and experimental validation. • Data-driven and machine learning methods for process control and optimization. • Design and optimization of scanning strategies and process parameters. Contributions addressing different AM technologies (e.g., LPBF, DED, WAAM) and material systems are welcome. This session aims to provide a platform for discussing emerging trends and fostering collaboration toward the development of reliable, high-fidelity simulation tools that support the broader industrial adoption and certification of additive manufacturing.
Organized by: C. Moreira (CIMNE, Spain), M. Chiumenti (UPC, Spain) and M. Caicedo (UPC, Spain)
Keywords: Computational Modelling, Defect Prediction, Direct Energy Deposition (DED), Finite Element Method (FEM), High-Performance Computing (HPC), In-Situ Monitoring, Laser Powder Bed Fusion (LPBF), Machine Learning, Metal Additive Manufacturing, Multi-Scale Modelling, Part-Scale Analysis, Reduced-Order Modelling (ROM), Surrogate Modelling, Wire Arc Additive Manufacturing (WAAM)
5300 Structural Mechanics and Dynamics
Following the introduction of the vehicle–bridge interaction (VBI) element in 1995, VBI was rapidly extended to the extraction of bridge frequencies from the dynamic response of a moving test vehicle. This technique, known as the vehicle scanning method (VSM), has been applied to the detection of damping, mode shapes, and damage in bridges. It offers a key advantage in that no vibration sensors need to be mounted on the bridge; instead, only one or a few sensors are required on the test vehicle. Compared with the conventional direct method, which relies entirely on the response of a sensor-instrumented bridge, VSM demonstrates clear advantages in mobility, cost-effectiveness, and efficiency. In recent years, a rapidly growing body of research has been conducted along the lines of VSM for bridge monitoring, leading to significant advances in various applications, including the extraction of bridge modal parameters, the investigation of frequency variation and dynamic coupling, experimental validation, and three-dimensional modeling. In this context, this minisymposium aims to provide a forum for researchers worldwide to present recent developments in vehicle–bridge interaction and its applications in structural health monitoring.
Organized by: J. YANG (National Yang Ming Chiao Tung University, Taiwan), Y. YANG (Chongqing University, China), J. YAU (Tamkang University, Taiwan) and S. URUSHDAZE (Institute of Theoretical and Applied Mechanic, Czechia)
Keywords: structural health monitoring, vehicle-bridge interaction, structural dynamics, vehicle scanning method
Beam and shell structures are used as primary components in various engineering applications. Their prevalence is further encouraged by new materials that enable the design of highly optimised shapes. These structures can exhibit complex mechanical responses driven by plasticity and other inelastic material behaviours, coupled with geometrical nonlinearities and strongly influenced by imperfections. Consequently, developing numerical approaches that offer robustness, efficiency, and accuracy in analysing these types of structures is a topic of significant interest in computational mechanics, involving modelling, discretization methods, and nonlinear solvers. Based on these premises, this mini-symposium aims to bring together scientists worldwide working on advanced methods for the nonlinear analysis of structures used in civil, mechanical, marine, aerospace, and biomedical engineering applications. Topics to be addressed within this session include (but are not limited to): ● Enhanced structural models for beam and shell structures experiencing material nonlinearities and large deformations. ● Discretization methods using strong formulations (e.g., collocation and differential quadrature methods, inverse differential quadrature methods) and weak formulations (e.g., finite element method, boundary element method, isogeometric analysis, virtual element method). ● Advanced computational methods for evaluating the nonlinear behavior of beams and shells in both static and dynamic scenarios. ● Path-following strategies in static and dynamic analyses. ● Efficient and stable time integration schemes (both implicit and explicit). ● Reduced-order models. ● Nonlinear coupled problems (e.g., magneto-electro-thermo-mechanical problems, fluid-structure interactions). ● Multi-level and multi-scale analysis. ● Numerical methods for reliable safety assessments, including cases involving exceptional loads. ● Structural optimization and control considering nonlinear behavior. ● Limit and shakedown analysis. ● Delamination in composite beams and shells.
Organized by: G. Garcea (UniversitĂ  della Calabria, Italy)
Keywords: Advanced numerical methods, Beams;, Optimization, Shells;
5400 Experimental Techniques and Computational–Experimental Integration
Composite materials are increasingly employed in mechanical engineering and aerospace applications due to their high strength-to-weight ratio and excellent specific mechanical properties. However, composite structures are susceptible to internal damage mechanisms, including matrix cracking, delamination, and porosity, particularly under low-velocity impact loading. Such damage can significantly reduce structural integrity and mechanical performance while remaining difficult to detect through conventional visual inspection. Non-Destructive Testing (NDT) techniques, such as ultrasonic inspection and infrared thermography, are widely used to detect and characterize internal damage without compromising the integrity of composite structures. Nevertheless, the analysis and interpretation of the large and complex datasets generated by these techniques remain challenging and often require substantial expert intervention. To address these limitations, this work investigates the integration of Artificial Intelligence (AI) with experimental mechanical testing and NDT data for the automated detection, characterization, and prediction of internal damage in composite structures. The proposed methodology combines low-velocity impact testing, non-destructive inspection, image-based damage characterization, and AI-based predictive modeling to establish relationships between measurable damage features and the resulting degradation of mechanical properties. The developed approach aims to improve the accuracy and efficiency of internal damage assessment while reducing reliance on manual interpretation. The results highlight the potential of AI-based methods for intelligent damage detection and prediction, with promising applications in structural health monitoring and predictive maintenance of composite structures.
Organized by: M. Benouis (Ecole militaire polytechnique, Algeria)
Keywords: Artificial Intelligence, Composite materials, Damage prediction, Internal damage, Low-velocity impact, Non-Destructive Testing, Structural health monitoring