EXTENDED ABSTRACT: The rapid advancement of Industry 4.0 has created a growing demand for secure, efffcient, and collaborative data management in materials research and industrial community. However, existing platforms struggle with data silos, security concerns, and limited collaboration, which slow down innovation and hinder material discovery. To address these challenges, the National Material Data Management and Services (NMDMS) platform [1], which integrates over 14 million material data entries from more than 30 institutions across China [2], focuses on three key areas: security, storage, and collaborative computing. By integrating blockchain technology, swarm transfer learning [3], and a hybrid storage model, NMDMS enables secure sharing and management of large-scale, heterogeneous material data across institutions. Its hybrid storage architecture combines MongoDB for public data with decentralized IPFS for private data, allowing institutions to manage sensitive information securely while contributing to collaborative research. NMDMS also utilizes swarm transfer learning to align heterogeneous datasets, improving prediction accuracy and facilitating equitable collaboration across institutions of varying sizes. In summary, NMDMS offers a comprehensive, secure, and efficient solution for managing material data, driving collaboration, and accelerating material discovery through cutting-edge technologies, making it a key innovation in the ffeld of materials research.

Keywords: Material big-data; Collaborative intelligence; Data integration.

REFERENCES:

[1] Liu, S., Su, Y., Yin, H. et al., npj Comput Mater, 7, 88 (2021).

[2] Gong, H., He, J., Zhang, X. et al. Sci Data 9, 787 (2022).

[3] Wang, R., Xu, C., Zhang S. et al. Nat Commun, 15, 9066 (2024).