A New Marketplace for Materials Innovation

IDEAS:MD3 is the the latest center in Georgia Tech’s continued expansion of its Materials Innovation Ecosystem. Its organization was led by David L. McDowell and Surya Kalidindi with the mission of developing the initial data engineering and science foundation in that ecosystem, a key to unleashing new potential in solving 21st century grand challenges.

This materials innovation ecosystem builds on and leverages Georgia Tech’s thought leadership established by the Materials Genome Initiative (MGI), existing strengths in technology translation, strong relationships with original equipment manufacturers and their supply chains in transportation, health, energy and defense industries and research excellence in materials science and engineering, computational science and engineering, manufacturing, and materials data sciences and informatics.

Materials with enhanced performance characteristics have served as critical enablers for the successful development of advanced technologies throughout human history. The diversity of governing physics at multiple length and time scales in materials has resulted in highly customized and ad-hoc workflows for expressing, extracting, curating, and disseminating the core materials knowledge, culminating in a “variety” challenge of big data. Consequently, there is now a major economic incentive to synthesize and express this core material knowledge in structured/organized forms that facilitate easy integration into high value product design and design for commercial manufacturing enterprises. This gap has been identified as the primary hurdle in the accelerated insertion of high performance materials in advanced technologies, and has been specifically targeted in several recent federal initiatives and investments (e.g., MGI, Advanced Manufacturing Partnership (AMP), Manufacturing Innovation Institutes (IMI), Materials Development Platforms (MDP), Materials Innovation Platforms (MIP). These investments are expected to grow significantly in the coming years. Furthermore, the same gap has also been identified as a major challenge in successfully navigating the “valley of death” for materials development in the Integrated Computational Materials Engineering (ICME) paradigm outlined in 2008 by the National Academies. Closing the multiscale, multiphysics, materials-manufacturing gap has the potential to dramatically reduce the current timeline (15-20 years) for development of materials from the discovery and/or design stage to certification and deployment in products, while simultaneously realizing cost reduction

IDEAS:MD3 will focus sharply on closing this gap by synergistically fusing and leveraging GT’s widely acknowledged expertise in materials engineering and sciences with leadership class expertise in computational simulation and data engineering and sciences.

Through IDEAS:MD3 Georgia Tech has established the very first university-industry collaborative research center in the nation focused on materials data science- and cyberinfrastructure-enabled MD3 (IDEAS: MD3). The center also serves as an attractor for industry and startups in the space at the intersection of materials discovery, development and big data in the new HPC facility. The center will also establish the first comprehensive materials e-innovation ecosystem. This is serving as a national model, strengthening our relationships with various global industry partners, who expressed a strong interest in establishing synergistic partnerships. This collaboration is leading to application of the novel concepts and the emergent toolsets to their ongoing MD3 projects, allowing them to aggressively recruit a new cadre of GT Ph.D. students uniquely trained in the envisioned materials innovation ecosystems, as well as retraining of their existing workforce.

Another central activity of IDEAS:MD3 is to promote, collect, archive, curate, and disseminate open source, open access, data and code repositories related to hierarchical material systems. Existing materials databases do not address the complexities inherent to documenting the hierarchical materials structure and the governing multiphysics underlying the material properties/responses of interest. Multiscale experiments and models employed by experts in the field are easily capable of producing extremely large datasets. Even for a single material processed in a single manufacturing history, datasets can easily accumulate to 1-10 TB.  Georgia Tech plays a visionary and leadership role in this domain.  It is also apparent that the materials databases and code repositories of the future will be federated across many nodes with the high value metadata and knowledge systems aggregated in hubs. In this regard, we plan to leverage fully the activities of the recently formed Materials Accelerator Network (GT lead: McDowell) in connection with the MGI as well as the soon-to-be-formed South Big Data Regional Innovation Hub (GT lead: Srinivas Aluru). Our vision is to be become an application “spoke” in the Big Data Innovation Hub, while becoming the main national hub for Hierarchical Materials Informatics. Our vision is to establish GT as the foremost global authority on effective e-collaboration workflows that will produce the disruptive transformation envisioned by MGI/ICME. In the process, we envision creating completely new markets for materials data, analytic tools, and knowledge databases with many new opportunities for incubation of new commercial technologies/products created by GT researchers and their successful translation through various programs offered by GT’s EI2.

IDEAS:MD3 will identify a specific set of problems that address the broad interests of industry partners, while being relevant to the scientific and technological foundations described above. Based on our preliminary discussions over the past few years and consideration of Tech’s core materials strengths,  the following opportunities have been identified: (i) Gas turbine propulsion material systems, including additive manufacturing, for aerospace industry (e.g., elevated temperature jet engine components) that demand simultaneous considerations of the intricate coupling between chemistry, non-equilibrium and heterogeneous processing conditions, mesoscale structure, and macroscale geometry.

A New Materials Innovation Cyber-Ecosystem

IDEAS:MD3 will conduct research focused on materials data science and informatics approaches to accelerate materials design, development and deployment (MD3) and invites interested organizations to become members and participate in the research activities conducted therein.  IDEAS:MD3 will facilitate exploration of open source and commercial data science methods within the materials innovation ecosystem with a strong focus on workforce development and retraining.

Preparing the 21st century workforce for accelerated materials design, development and deployment through:

  • Low entry cost opportunity to familiarize and train your current employees in MD3

  • Increased exposure of your organization to modern data science tools and e-collaboration platforms for materials discovery and development

  • Networking with domain experts at the nexus of materials science, manufacturing, data science, and high throughput methods, including commercial data services vendors; closer linkage of OEMs and their supply chains

Advancing research and development in data science and informatics via:

  • Integrated projects making use of existing and emergent state-of-the-art best practices and methods, tailored to your needs

  • Methods to accelerate process development with data-driven decision support

  • Strategies to identify and organize important materials data and accumulated knowledge, including important electronic metadata in materials development

  • Accelerated qualification procedures for vendors and the supply chain


Additional Resources:

Institute for Materials

FLAMEL - Accelerating the design of high performance materials

MATIN - An e-collaborative platform in materials science

Material Informatics in the classroom

MINED - Material INformatics for Engineering Design