AlloyLens: A Visual Analytics Tool for High-throughput Alloy Screening and Inverse Design

Susie S.Y. Li, Fernando Fajardo-Rojas, Diego Gomez-Gualdron, Remco Chang & Mingwei Li

bScatterplot matrix

  • Alloy
  • Matches filters
  • Recommended
  • Excluded
  • Surrogate model
Loading alloy data

Fig. 1 | Find alloys that meet a design target. a, Choose the variables for the columns and rows of the matrix. b, Each plot compares one variable against another. Drag the sliders along an axis to set a target range, or draw a box on a plot. Alloys inside every range turn magenta. If none fit, the closest alloys are recommended in yellow, fading to grey with distance. Double-click a slider to reset it. c, Export the selected alloys, overlay curves predicted by the surrogate model, and set how far the recommendation range reaches when no alloy fits. n = 2,000 alloys sampled from the 2025 IEEE SciVis Contest dataset by Bugelnig and Requena (CC BY 4.0).