Conference Logo ProgramMaster Logo
Conference Tools for 2026 TMS Annual Meeting & Exhibition
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools

About this Abstract
Meeting 2026 TMS Annual Meeting & Exhibition
Symposium Additive Manufacturing Modeling, Simulation and Artificial Intelligence
Presentation Title Prediction of Pore-Driven Debit in Fatigue Performance of Additively Manufactured Materials Via Graph Neural Networks
Author(s) Luca Loiodice, Krzysztof S. Stopka, Yixuan Sun, Guang Lin, Michael D. Sangid
On-Site Speaker (Planned) Luca Loiodice
Abstract Scope Despite advances in additive manufacturing (AM), it remains unfeasible to eliminate pore defects in components, making it essential to understand their impact on fatigue performance. This study employs Graph Neural Networks (GNN) to predict the potential fatigue debit caused by individual pores in AM IN718. GNNs are well suited for modeling such microstructure-sensitive phenomena due to their ability to capture local grain–pore interactions through graph-based representations. Hundreds of pores were extracted from computed tomography scans, covering a large range of different sizes and morphologies, and multiple statistically equivalent virtual microstructure models were analyzed for each pore. Crystal plasticity simulations were conducted using an elasto-viscoplastic FFT solver, leveraging its high computational efficiency to generate a large training dataset comprising microstructural descriptors and a fatigue damage metric for lifetime assessments. The proposed GNN offers a rapid means to predict the influence of individual pores on the fatigue performance of AM materials.
Proceedings Inclusion? Planned:
Keywords Machine Learning, Additive Manufacturing, Computational Materials Science & Engineering

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

3D Surrogate Modeling of Elasto-Viscoplastic FFT Simulations for Porosity-Driven Fatigue Prediction in Additive Manufacturing
A Comprehensive Hot Cracking Model Development by Coupling CALPHAD and Machine Learning
A Framework for Efficient Part-Scale Microstructure Prediction in Laser Powder Bed Ti-6Al-4V Using Combined Physics-Based Modeling and Machine Learning Surrogate Methods
A Machine Learning-Based Model for Fatigue Behavior Prediction and Analysis of Surface Treated Laser Powder Bed Fused Nickel Based Super Alloys
A Statistics-Informed Efficient Micromechanical Model for Additively Manufactured Ti-6Al-4V
Adding Sub-Grain Features to Mesoscale Additive Manufacturing Microstructure Simulations
AI-Guided Resilience of Additive Manufacturing to Expeditionary Environments
AI Agent-Managed, Simulation-Guided Adaptive Control of Additive Manufacturing
An Experimentally Validated Defect Mapping Strategy for Laser Powder Bed Fusion Processed Nickel Alloys
BOF Endpoint Carbon Content Prediction Model Based on Data and Mechanism Driven
Bridging and Coupling Models with Microstructure and Thermo-Fluid Flow in Powder Bed Fusion
Columnar-To-Equiaxed Transition in Laser Fusion Additive Manufacturing
Combining Physics-Based and Machine Learning Models for Process Control in Laser-Based Metal Additive Manufacturing
Crystal Plasticity–Informed Surrogate Modeling for Predictive Laser Powder Bed Fusion Simulation of Ti-6Al-4V
Crystal Plasticity Finite Element Modeling of Polycrystalline Metals Produced by Additive Manufacturing
F-21: Advancing Ultrasonic Atomization through AI-Driven Process Control
F-22: An Artificial Neural Network Modeling Approach to Electroforming of Micro-Tubes in the Presence of Surfactants
F-23: Computational Thermodynamics-Based Optimization of Ti-6Al-4V via Al and V Substitution for Additive Manufacturing
F-24: Convolution Tensor Decomposition Method for Efficient High-Resolution Solutions to the Allen-Cahn Equation: Application to Grain Growth Simulations in Additive Manufacturing
F-25: Creep Property of SS316L Lattices Fabricated by Laser Powder Bed Fusion
F-26: Enhancing Binder Jetting Performance: A Multi-Scale Numerical Approach to Residual Porosity Reduction
F-27: Expanding Capabilities to Simulate Powder Bed Fusion Process With Additive Manufacturing Module
F-28: Experimental‑CFD-Based Observation of Thermo‑Fluid Mechanisms in LPBF with Nanoparticle‑Coated Powders
F-29: Impact of Intermittent Laser on Fluid and Thermal Response in Directed Energy Deposition: A Numerical Study
F-30: Influence of Inoculants on Aluminum Alloy Microstructure Evolution During Fast Solidification: A Computational Approach Under Laser Powder Bed Fusion Conditions
F-31: Lightweight, Bending-Resistant 3D Voronoi Structures Inspired from Feather Rachis and Optimized by Genetic Algorithm
F-32: Melt Pool Control and Segregation Using Magnetic Field in Additive Manufacturing
F-33: Modeling and Simulation of the Shock Response in AM Material Accounting for the Retained Porosity
F-34: Numerical Analysis of the Flow Behavior in the Mold Induced by a Real Clogged Nozzle and Symmetrically Reduced Nozzles
F-35: Optimization of Binder Jet Printing Parameters of an Irregular, Non-Spherical Copper Powder Manufactured Via Solid-State Machining and Comparison to a Simplified DEM Simulation
F-36: Predictive Multi-Physics Multi-Material Design of Refractory Materials in Laser Powder Bed Fusion Additive Manufacturing
F-37: Quantifying Process Uncertainty in Laser Powder Bed Fusion: A Modeling Approach for Surface Topography Prediction
F-38: Simulation and Experimental Investigation of Dual-Side Deposition Strategy for Deformation Control in DED-Arc
F-39: Simulation and Optimization Ceramic Mold Cleaning Process Using Transparent 3D-Printed Molds
F-6: Computational Prediction of Grain Structure Evolution in Thermoelectric Bulk Parts Processed by Laser Powder Bed Fusion
Identifying LPBF Anomalies and Creep Damage in Nickel Based Super Alloys
Image-Based Uncertainty Quantification of Geometric Deviations in Additive Manufacturing
Integrated Melt Pool Dynamics and Defect Prediction in Additive Manufacturing of Inconel 625
Integrating AI-Driven In-Situ Monitoring with Additive Manufacturing: A Multi-Angle Vision Approach for Defect Detection
Leveraging Laser Parameters and Layer Remelting to Tailor Microstructure in Laser Powder Bed Fusion
LPBF Microstructure Modeling for Multiple Materials
Machine Learning Based Adaptive Deposition Control for Wire Arc Additive Manufacturing Repair
Machine Learning Guided Prediction of Electrohydrodynamic Jet Printing by Incorporating High Speed Video Data
Mechanistic Understanding of Strengthening in WAAM NAB Alloys Through Dislocation Dynamics
Microstructure Validation of 316L Stainless Steel Additive Manufacturing Using CAFE with MOOSE/MALAMUTE Thermal Fields
Modeling Particle Transport in Laser Powder Bed Fusion Melt Pools
Overcoming Strength-Ductility Trade-Offs in Additively Manufactured Aluminum Alloys Through Integrated Computational Design
Prediction of Pore-Driven Debit in Fatigue Performance of Additively Manufactured Materials Via Graph Neural Networks
Quantitative Validation Methodology for Grain Growth Models Using Probabilistic Metrics
Residual Stress Evolution in a Nickel-Aluminum Bronze Wire-Arc Additive Manufacturing Build
SciML for Modeling Fatigue Indicator Parameters Near Voids in AM IN718
Simulating CMT Additive Manufacturing with CFD for Process Parameter Optimization
Simulation Calibration and Process Optimization for DED
Spatters in Laser Powder Bed Fusion: Physics-based Simulations, Formation Mechanisms, and Reduction Strategies
Thermomechanical Crystal Plasticity Study of HCP Ti Under Solid State Thermal Cycling
Toward Processing Compositional Gradients of Thermally Dissimilar Nickel and Copper Alloys with Laser Powder Directed Energy Deposition (LP-DED)
Questions about ProgramMaster? Contact programming@programmaster.org | TMS Privacy Policy | Accessibility Statement