I am a Structural Analysis Engineer at Standard Nuclear and hold a Ph.D. in Mechanical Engineering from the University of Tennessee, Knoxville. I work at the intersection of computational mechanics, scientific machine learning, and high-performance computing. My experience spans structural and thermal analysis, constitutive and phase-field model development, numerical PDE solvers, and graph neural network surrogates for material response and fatigue. I connect mechanics models with automated simulation and data workflows.
Research Interests
Graph neural networks for material properties and fatigue-critical extremes in polycrystalline alloys and fiber composites.
Fiber-composite graph surrogates (Caliskan et al., 2025). Research at ORNL.
Multiscale computational mechanics using finite-element, discontinuous Galerkin, and phase-field methods.
Alumina metamaterials under impact, using the Johnson-Holmquist 2 material model. Research at UTK.
High-strain-rate dynamics and failure of heterogeneous materials, including metals, ceramics, and composites.
Boron carbide/silicon carbide ceramic composite simulation. Research at UTK.
Scalable computing and reproducible datasets for materials simulation and scientific machine learning.
Designing, manufacturing, and experimentally testing components for nuclear thermal propulsion, supported by structural, thermal, and material modeling in Ansys. Additional work includes OpenFOAM simulations and a fuel container project.
Developed graph neural networks to predict the maximum fatigue indicator across a microstructural domain, with hierarchical grain-to-domain modeling and calibrated uncertainty. Also built elastoplastic and composite-property surrogates and coauthored the open CP2D dataset.
Developed constitutive and phase-field subroutines, analyzed numerical methods for elastodynamics, and automated large simulation studies of fracture and metamaterials under impact and blast loading.
Thesis: Design of Composite Monocoque Body for Lightweight Gasoline Vehicles.
Advanced Training
Argonne Training Program on Extreme-Scale Computing (ATPESC)
July 26 - August 7, 2026 · Argonne National Laboratory
Competitively selected for an intensive two-week program in extreme-scale computing at Argonne National Laboratory. Training covered GPU and MPI/OpenMP programming; scalable numerical methods (AMReX, PETSc, hypre); profiling, mixed precision, and parallel I/O; distributed ML, LLM training, and agentic simulation-AI workflows; scientific software testing and reproducibility.
Caliskan, E., Anto, A. D., Lupo Pasini, M., TerMaath, S., and Abedi, R. (2026). Multitask graph neural networks for elastoplastic response prediction in dual-phase polycrystals.Journal of Materials Science: Materials Theory, 10, 8. Published article
Caliskan, E., Cheney, W., Wang, W., Plaisted, T., Amirkhizi, A. V., and Abedi, R. (2026). Time domain analysis of locally resonant elastic metamaterials under impact.Mechanics of Advanced Materials and Structures, 33(1), 2619034. Article
Caliskan, E., Abedi, R., and Lupo Pasini, M. (2025). Graph Neural Networks for Mechanical Property Prediction of 2D Fiber Composites.Materials & Design, 257, 114500. Article
Abedi, R., Furey, C., Pourkamali-Anaraki, F., Huynh, G., Caliskan, E., and Amirkhizi, A. V. (2025). Analyzing fragmentation response of heterogeneous ring using the method of characteristics and machine learning techniques.Computer Methods in Applied Mechanics and Engineering, 436, 117709. Article
Caliskan, E. and Kirca, M. (2022). Tensile characteristics of boron nanotubes by using reactive molecular dynamics simulations.Computational Materials Science, 209, 111368. Article
Preprints and Manuscripts
Heidari Shirazi, A., Caliskan, E., Amirkhizi, A. V., and Abedi, R. A Higher-Order-Mode Scattering Framework for Periodic Elastic Slabs. Under review.
Caliskan, E. and Abedi, R. (2026). A unified dispersion and stability analysis of discontinuous Galerkin formulations for one-dimensional elastodynamics. Preprint. PDF
Caliskan, E., Anto, A. D., TerMaath, S., Abedi, R., and Lupo Pasini, M. (2026). Extreme-value-aware graph surrogates for fatigue localization in austenitic stainless steel. Preprint; manuscript in preparation for submission. PDF
Research Dataset
Anto, A. D., Caliskan, E., Lupo Pasini, M., TerMaath, S., and Abedi, R. (2026). CP2D Dataset: Dual-Phase Polycrystal SVEs with Elastic and Plastic QoIs. Zenodo. Dataset and citation
Open data supporting polycrystal surrogate modeling, including microstructure graphs, elastic and plastic quantities of interest, and a saved data split.
HPC resource scholar: UHeM and TRUBA, 2020 - 2022.
Conferences
Contributed to 20 conference abstracts, including presentations at SES, IMECE, WCCM, USNCCM, EMI, and related venues. (★) presenter. View all conference entries. Selected presentations:
Erdem Caliskan★, Anik Das Anto, Reza Abedi, and Massimiliano Lupo Pasini. Probabilistic multi-task graph neural network surrogates for elastic-plastic behavior and fatigue indicator prediction in polycrystalline alloys. In SES Conference 2025, Atlanta, Georgia, USA, October 12-15, 2025.
Erdem Caliskan★, Willoughby Cheney, Weidi Wang, Alireza Amirkhizi, and Reza Abedi. Design and Dynamic Response Analysis of Resonant Microstructured Media. In International Mechanical Engineering Congress and Exposition 2024 (IMECE 2024), Portland, Oregon, November 17-21, 2024.
Erdem Caliskan★, Weidi Wang, Willoughby Cheney, Alireza V. Amirkhizi, and Reza Abedi. Transient nonlinear response of resonant metamaterial arrays under impact loading. In 16th World Congress on Computational Mechanics and 4th Pan American Congress on Computational Mechanics (WCCM / PANACM 2024), Vancouver, British Columbia, Canada, July 21-26, 2024.
