Experiences with ICME information infrastructures for applying materials models in sequence to give accurate macroscopic property prediction

This paper draws on findings from the Material Data Management Consortium. The consortium, now in its 15th year, enables leading engineering enterprises (including among others GE, Boeing, Honeywell, Lockheed Martin, and NASA) to collaborate on best practice approaches to managing and applying critical materials information and technology with a key focus on traceable workflows for simulation and multi-scale modelling of materials, their processing and the resulting components. A key emphasis has been on verification and validation of macroscopic materials properties generated through sequentially-applied materials models; from ab-initio codes or models of microscopic unit cells through calculations at various length scales up to macroscopic properties. An example of this integration technology is described, using the MAC/GMC micromechanics code from NASA Glenn Research Center. Focus is given to the traceability of the simulated data and capture of sufficient metadata to ensure that the simulations can be recreated in future. The traditional method for understanding the effects of these different factors on materials innovation follows a set pattern where advantageous properties are pursued, iterating between process and structure while measuring the appropriate property. Progress relies on extensive and expensive physical testing campaigns, which generate vast quantities of highly complex data. There are hundreds of possible test types within materials engineering, all generating data of differing types, complexity, and different formats