IDMM Lab

Generative AI · Multiphysics Modeling · Multiscale Modeling · Digital Twins · Design Automation · Advanced Manufacturing

생성형 인공지능 · 멀티피직스 모델링 · 멀티스케일 모델링 · 디지털 트윈 · 설계 자동화 · 첨단제조

The Intelligent Design for Materials & Manufacturing (IDMM) Lab (지능형 소재 및 제조 설계 연구실) at Soongsil University builds computational methods that connect generative AI, multiphysics modeling, and digital engineering. We design materials, parts, and manufacturing processes that are difficult to design by hand — and we build the digital infrastructure that makes those methods reusable across problems.

Our work cuts across four thrusts — Generative AI for Advanced Materials & Component Design, Multiphysics & Multiscale Modeling, Sparse-to-Rich 2D/3D Material Characterization & Reconstruction, and Digital Engineering for Advanced Manufacturing — with an application focus on metal additive manufacturing, microstructure design across diverse material systems (energy storage and conversion materials, advanced composites, and alloys), and next-generation aerospace component design. See the Research page for details, and Publications for our recent work.

We work closely with researchers in mechanical engineering, aerospace engineering, materials science, and applied machine learning, and we welcome students who enjoy living at the boundary between physics, computation, and hardware.


Join Us

We are recruiting postdoctoral researchers, graduate students (M.S. / Ph.D.), and undergraduate research interns interested in working at the intersection of AI, physics, and manufacturing. Please see the Join Us page for details on what to send.


숭실대학교 기계공학부 지능형 소재 및 제조 설계 연구실 (IDMM Lab)은 생성형 AI, 멀티피직스 모델링, 디지털 엔지니어링을 결합하여 재료·부품·첨단 제조 시스템을 설계하는 계산 방법론을 연구합니다. 박사후연구원, 석·박사 과정생, 학부 인턴을 모집하고 있습니다 — 자세한 사항은 Join Us 페이지를 참고해 주세요.

News

Selected Publications

  1. VPP
    vpp.jpg
    Modeling the hierarchical microstructure of L-PBF Ti-6Al-4V: a multiphysics framework for texture-induced mechanical anisotropy
    Kang-Hyun Lee and Gun Jin Yun
    Virtual and Physical Prototyping, 2026
    Accepted · IF 8.8 · JCR Top 10% · Engineering & Manufacturing
  2. CMAME
    CMAME.jpg
    MicroLad: 2D-to-3D microstructure reconstruction and generation via latent diffusion and score distillation
    Kang-Hyun Lee and Faez Ahmed
    Computer Methods in Applied Mechanics and Engineering, 2026
    IF 7.3 · JCR Top 2.6% · Mathematics & Interdisciplinary Applications
  3. EAAI
    EAAI.jpg
    A data-driven framework for designing microstructure of multifunctional composites with deep-learned diffusion-based generative models
    Kang-Hyun Lee, Hyoung Jun Lim, and Gun Jin Yun
    Engineering Applications of Artificial Intelligence, 2024
    IF 7.5 · JCR Top 2.5% · Engineering & Multidisciplinary
  4. npj CM
    npjCM.jpg
    Multi-plane denoising diffusion-based dimensionality expansion for 2D-to-3D reconstruction of microstructures with harmonized sampling
    Kang-Hyun Lee and Gun Jin Yun
    npj Computational Materials, 2024
    IF 11.9
  5. ACHM
    ACHM.jpg
    Enhanced lightweight, moisture-resistant, and thermoelectric cement composites using carbon nanotube and hollow glass microsphere-based hybrid clusters
    Daeik Jang, Jaewoong Choi, Jinho Bang, and 3 more authors
    Advanced Composites and Hybrid Materials, 2026
    IF 21.8 · JCR Top 1.5% · Materials Science & Composites · Lee, K.-H. (corresponding author)