MAD3: Software Toolkit for Robust Predictions of Plastic Anisotropy in Metals

Votes: 0
Views: 57

MAD3 is an innovative software that leverages the power of machine learning to modernize the forming/stamping processes of sheet metals by predicting directional mechanical behavior of metal alloy (plastic anisotropy), 1000 times faster than existing solutions. Accurate characterization of plastic anisotropy in metal alloy is critical in major manufacturing industries (i.e., automotive, aerospace, metal manufacturers/suppliers) given the fact that it determines how a sheet of metal can be shaped to the desired component (fit and finish) and whether it will withstand the applied load (performance).

MAD3 allows manufacturers to characterize the deformation behavior of sheet metals in a manner that it was only available to research labs and universities who own specialized test equipments or High-Performance Computing resources (HPC), and have the necessary subject matter expertise. Therefore, MAD3 aids in drastically reducing development costs and significantly accelerating the design cycle and manufacturing time for a new formed, stamped, or forged component by enabling manufacturing companies to adequately characterize the deformation behavior of sheet metal alloys without the need to perform experiments or run complex and computationally expensive high-fidelity simulations.

MAD3 uses state-of-the-art and top-of-the-line data- driven and machine learning techniques to first extract a unique fingerprint descriptor of the metal alloy’s internal structure (crystallographic texture), then subsequently uses these descriptors to predict the parameters that characterize the deformation of a sheet metal alloy in an accurate and efficient manner. Moreover, MAD3 is conveniently packaged in a simple and easy-to-use graphical user interface (GUI) that is deployed using a web-based application. This means that the user does not need to install any program directly into their computer. This powerful feature maximizes portability and usability from any operating system. The MAD3 software solution is available for license from Sandia National Laboratories.

Given that the insights and predictions obtained with MAD3 are unique and significantly outperform any existing solution, there is a clear and unique market opportunity for MAD3. The total addressable market that MAD3 is a part of is approximately $1B (billion) dollars per year, as characterized by the revenue of the top 10 simulation software companies. The service-addressable market for MAD3 consists of $200M (million) dollars per year and corresponds to the revenues of the top 10 commercial software companies used by manufacturing companies to perform metal-forming analysis. Our estimated entry point market is $50M per year and corresponds to the estimated size of the service-addressable market focused on automotive and metal manufacturers, our initial target customers. The MAD3 team held discussions with more than 120 potential customers in different metal manufacturing companies, automotive companies, aerospace companies as well as national laboratories and top-tier research universities in order to develop a lean, customer-oriented product that has been recognized as one of the top 100 innovations of 2023 and named a “Market Disruptor” in the software industry by R&D World.

SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525. SAND No. SAND2024-07976O


Learn how to vote for your favorites.


  • Name:
    David Montes De Oca Zapiain
  • Type of entry:
    Team members:
    • David Montes de Oca Zapiain
    • Hojun Lim
  • Patent status: