Principles and Metrics for Digital Curation in Aviation

The aviation industry is leaning more and more into enterprise digital transformation as data needs to be shared at scale about the processes and engineering artifacts involved in the development process across many manufacturers and integrators. This requires the need for data-centric policies and a starting point is the four FAIR principles for curated data (Findable, Accessible, Interoperable, and Reusable). More recently, NASA has proposed FAIRUST, which builds on FAIR by adding Understandability, Security, and Trustworthiness. Similarly, the DOD has data strategy goals summarized as VAULTIS, to make data Visible, Accessible, Understandable, Linked, Trustworthy, Interoperable, and Secure. However, these principles and goals are abstract and not measurable and therefore too informal to support the kind of formal analysis that drives processes for certification (or qualification in military). In this paper, we propose concrete objectives and quantifiable metrics for FAIRUST and VAULTIS. These metrics should be published alongside the dataset so that data consumers can evaluate fitness for purpose of the data. We also evaluate a data curation platform we developed, called Rapid Assurance Curation Kit (RACK), and present how well our data curation workflow describes RACK, and conversely how well our current development of RACK scores against the expectations of the data curation workflow.