Making computational materials data and workflows interoperable and reusable

Computational materials science spans many methods, software codes, and length and time scales, producing data that can be difficult to compare, combine, and reuse.

FAIRmat Area C develops common data models, workflows, and tools that make computational materials data interoperable and reproducible across methods. By expanding method coverage, improving data quality, and connecting calculations across scales, Area C enables reliable data reuse, benchmarking, and data-driven materials discovery.

Objectives

Make computational data interoperable and reusable across methods, software codes, and scales.
Enable reproducible representation of complex computational workflows and simulations.
Improve the quality and reliability of computational data through benchmarking and data-quality assessment.
Enable computational data to be explored and reused for machine learning, AI, and materials discovery.

Diagram showing the FAIRmat Areas with Area C Computation highlighted.

Tasks in Area C

C1: Ground-state and electronic structure

Developing data models, quality-assessment tools, and benchmarks for ground-state and electronic-structure calculations across different computational methods.

Task leader: Silvana Botti

C2: Multi-excitations and dynamics

Harmonizing data from excitation and dynamics methods to describe time-dependent phenomena, spectroscopy, nuclear dynamics, and coupled quasiparticle excitations.

Task Leader: Claudia Draxl

C3: Multiscale modeling

Developing interoperable data models that connect simulations across scales, including molecular dynamics, Monte Carlo methods, and model Hamiltonians.

Task Leader: Tristan Bereau