Accelerating Materials Simulation with MAD3: A Machine Learning Breakthrough from Sandia National Laboratories

Video Url
Originally held on January 20, 2026 – Hosted by Sandia National Laboratories and the U.S. Department of Energy

About the Webinar

This National Lab Discovery Series webinar featured Materials Data Driven Design (MAD3), a powerful simulation platform developed by Sandia National Laboratories that is transforming how engineers predict material behavior in design and manufacturing. By combining machine learning with materials science, MAD³ enables fast, accurate prediction of how metals deform in real-world applications—without relying on costly testing or compute-heavy simulations.

Designed for ease of use and broad accessibility, MAD³ runs through a web-based interface and requires no specialized hardware, helping teams reduce development time, avoid costly failures, and optimize performance across a range of manufacturing workflows.

Key Topics Covered

  • How MAD³ predicts directional mechanical behavior (plastic anisotropy) in metals
  • Application of machine learning to overcome traditional simulation bottlenecks
  • Integration into forming, stamping, and structural simulation workflows
  • Opportunities to evaluate, license, and co-develop MAD³ with Sandia
  • Live Q&A with the Sandia development team

Featured Speakers

David Montes de Oca Zapiain, Ph.D. - R&D Scientist and Engineer, Sandia National Laboratories
Hojun Lim - R&D Scientist and Engineer, Sandia National Laboratories