Collaborative Research: Disciplinary Improvements: Creating a FAIROS Materials Research Coordination Network (MaRCN) in the Materials Research Data Alliance

Overview: A FAIROS-RCN administered by the Materials Research Data Alliance (MaRDA) will advance and coordinate FAIR data to support Open Science and bridge the fundamental gap between materials data and data-intensive methods including AI/ML. Closing this gap is central to the transformative acceleration of materials design and deployment envisioned in the U.S. Materials Genome Initiative (MGI). The central goal of Materials Science and Engineering is to discover and deploy innovative materials that serve society. unlocking critical applications in renewable energy and sustainability, health, agriculture, energy storage, quantum computing, and advanced manufacturing. MGI and parallel international efforts depend on broader understanding, adoption, and implementation of FAIR and Open Science practices with specific needs for functional data sharing built on metadata and data-format standards to facilitate interoperable systems that maximize investments in research infrastructure and efforts. Implementation of these practices requires partnerships in computational, data, and engineering development to create advanced data infrastructure and smart labs that leverage AI to accelerate experimental processing. Outcomes empowered by this work will in-turn create new engineering and computational opportunities to realize new classes and paradigms of applications and devices. The recent emergence of MaRDA as a community-driven, grassroots network that spans data stakeholders provides a unique platform to develop and administer an effective coordination network. MaRDA membership and participation bridges academia, industry, and publishing while providing liaison to federal agency participants in the MGI and international efforts. The proposed FAIROS-RCN will leverage MaRDA’s foundational structure and community leadership to provide workshops, hackathons, and training specifically focused on development of community metadata standards; open teaching and training materials; and broadened community FAIROS literacy and engagement. Intellectual Merit: The proposed RCN activity outcomes will accelerate the connection and diversification of the materials research community needed to create and utilize FAIR data. Activities will be organized in four synergistic areas: [FAIR Data] activities will convene academic and industry researchers to foster concurrent development of materials-data vocabularies and ontologies within communities of practice. Activities will focus on high-impact community data generation in sub-fields aligned with grand challenge problems identified in recent NSF-funded workshops and square tables. [FAIR Model] activities will coordinate sharing of models and ML best practices developed across the domain and implement extensions of FAIR for reuse and transparency. Activities will focus on defining model metadata standards and evaluation criteria to enable understanding of model performance, rapid reuse, and efficient comparison. [FAIR Train] will coordinate sharing of existing educational materials, implementation of FAIR suited to teaching and training materials, and host workshops geared to accelerate access and adoption of data-centric methods across curricula. Efforts will include specific activities to articulate needs in under-represented communities and creation of training that meet those needs. [FAIR Impact] activities will coordinate dissemination of results and support adoption of materials and resources created or utilized in the FAIROS-RCN activities. This will leverage existing