Multiscale, Multiphysics Simulation Meets AI-Based Science for Advanced Materials Design

The current approach to materials discovery and design remains dominated by experimental testing, often based on little more than trial and error. With the advent of ever more powerful computers, rapid, reliable, and reproducible computer simulations are beginning to represent a feasible alternative. As high-performance computing reaches the exascale, exploiting the resources efficiently presents interesting challenges and opportunities. Multiscale modelling and simulation of materials are extremely promising candidates for exploiting these resources based on the assumption of a separation of scales in the architectures of nanomaterials. We apply multiscale techniques, from the electronic to the continuum scale, to the efficient design of a range of nanocomposites made from polymers and 2d-nanomaterials. We have also built an interoperable, open-source and open development environment for validation, verification and uncertainty quantification which make such simulations actionable. A range of applications we are involved in cover not only advanced materials but fusion energy, weather/climate, biomedicine, epidemiology and human migration. Concurrently, we have developed hybrid machine learning (ML) and molecular dynamics (MD) workflows to accelerate drug discovery. The two <i>in silico</i> approaches each have their own complementary advantages and limitations. Here, we discuss an innovative cyberinfrastructure development that combines both approaches to accelerate drug discovery on supercomputers. Such workflows could also prove beneficial for advanced materials discovery.