Dr. Tony Fast leads this Chalk & Talk discussion. Isolated improvements in algorithms, technology, and computation directly impact the landscape of information use in materials science.  A grander adoption of data-driven science is looming within materials science, as the materials science community rapidly generates large volumes of heterogeneous materials information.  These multi-modal and spatiotemporal packets of information are quickly outgrowing traditional empirical approaches to analysis.  This provides a new opportunity to create research profiles of users steeped in materials domain knowledge and data science; these users will interface data scientists with materials scientists to fill the space in-between.

The materials data scientist will use a diverse set of skills for data management, feature identification, feature representation/parametrization, data modeling, and reporting. The materials data scientist will understand reduced order modeling, statistics, algorithms, and coding.  The materials data scientist will also need to be social in order to  enable the next generation of bi-directional structure-property-processing linkages that are required to discover and develop advanced materials.

In this talk, a few early case studies in data-driven methods for solving materials science problems will be explored.  Emerging spatial statistics tools will be explored that enable an objective comparison of static and evolving 3-D material volumes from molecular dynamics simulation, micro-CT, and Scanning Electron Microscopy.  Also, the statistics will provide a foundation to create improved bottom-up homogenization relationships in fuel cell materials.  Lastly, applications of the Materials Knowledge System, a data-driven meta-model to create top-down localization relationships will be explored for phase field model and finite element model information.  Lastly, this lecture will touch on how to ensure subsequent generations of materials scientists are increasingly collaborative.