Dr. Özcan outlines how machine-learning–guided experiment design enables autonomous experimentation in MAPS and contributes to accelerating materials discovery. To operate effectively, this system requires strong connections between automated experiments, simulation-based material pre-screening, and early stages of upscaling and technology assessment.
A central point of her contribution is the need for aligned workflows, structured metadata, and consistent data recording practices. She explains that these elements are crucial for making autonomous experimental systems reliable and interoperable. Within this context, NFDI-MatWerk provides a valuable foundation. The MAPS environment serves as a testbed where the infrastructures, standards, and tools developed by NFDI-MatWerk can be applied, validated, and improved.
Her perspective illustrates how coordinated data practices and shared infrastructures support a more coherent and transparent process for generating and evaluating research data—an essential prerequisite for accelerating pathways towards new materials.