A Data Management Platform for AI-Driven Chemical Experimentation

Project description

The RobotLab project aims to revolutionize the chemical industry by integrating artificial intelligence (AI) into an automated laboratory framework, enabling efficient and autonomous experimentation. This project focuses on developing a sophisticated data management system to handle the extensive and diverse data generated by AI-driven experiments in the RobotLab. The main research question explores the extent to which existing data management frameworks can provide a sustainable solution for managing large amounts of data in laboratory settings, which is supported by sub-questions.

Context

The RobotLab project operates in chemical experimentation with a mission to redefine conventional practices in the chemical industry. Currently, experiments are designed and performed with human involvement, consuming considerable time and resources. The RobotLab project envisions a future where experimentation is not only designed but also executed by a self-driving laboratory, integrating automated machinery and artificial intelligence (AI). This ambitious project seeks to revolutionize the chemical industry by introducing an automated laboratory capable of efficiently designing and conducting experiments.

The project's context involves the development of a sophisticated data management framework to navigate the complexities of organizing, securing, and extracting meaningful data from the large amounts of data produced within this innovative laboratory setting.

As the chemical industry undergoes a technological evolution, the RobotLab project emerges as a pioneering force, emphasizing the critical need for advanced data management solutions to unlock the full potential of AI-driven experimentation. The project's context is characterized by the convergence of chemistry, artificial intelligence, and automation, with a central objective of ushering in a new era of efficiency and precision in chemical research and development.

Results

In the pursuit of designing a robust data management framework for the RobotLab project, a set of requirements, including both functional and non-functional aspects, was formulated. The thorough examination of the project's unique context led to the creation of Epics—strategic combinations of functional and non functional requirements linked to the RobotLab project's Key Performance Indicators (KPIs). These EPICS serve as a roadmap for the implementation of an effective data management system.

NOMAD, as part of the Dutch-German consortium FAIR Data Infrastructure, emerged as a candidate to address the challenges associated with managing the large and varied data generated in the RobotLab project's experiments. The selection of NOMAD has come to light because of its similar use cases and by its open-source nature, tools for data management, sharing, and publishing, and its alignment with FAIR principles (Findability, Accessibility, Interoperability, and Reusability).