Big data-assisted digital twins for the smart design and manufacturing of advanced materials: from atoms to products

Motivated by the ever-increasing wealth of data boosted by national strategies in terms of data-driven Integrated Computational Materials Engineering (ICME), Materials Genome Engineering, Materials Genome Infrastructures, Industry 4.0, Materials 4.0 and so on, materials informatics represents a unique strategy in revealing the fundamental relationships in the development and manufacturing of advanced materials. Materials developments are becoming ever more integrated with robust data-driven and data-intensive technologies. In the present review, big data-assisted digital twins (DTs) for the smart design and manufacturing of advanced materials are presented from the perspective of the digital thread. In the introduction of the DT design paradigm in the ICME era, the simulation aspects of DT and the data and design infrastructures are discussed. Referring to the simulation and theoretical factors of DTs, high-throughput simulation and automation and artificial intelligence-assisted multiscale atomistic modeling are detailed through several cases studies. With respect to data and data mining technologies, entropy and its application for attribute selection in decision trees are discussed to emphasize knowledge-based modeling, simulation and data analysis in machine learning coherently. Guided by the perspectives and case studies of the digital thread, we present our recent work on the design, manufacturing and product service via big data-assisted DTs for smart design and manufacturing by integrating some of these advanced concepts and technologies. It is believed that big data-assisted DTs for smart design and manufacturing effectively support better products with the application of novel materials by reducing the time and cost of materials design and deployment.