Abstract
Ensuring the evolvability of data stewardship planning is a critical challenge, particularly in environments where requirements frequently change. Traditional data management plans (DMPs) often lack modularity, adaptability, and machine-actionability, making it difficult to automate their evaluation, update them in response to new policies, or tailor them to specific disciplines. This paper presents a framework for machine-actionable and evolvable data stewardship planning, leveraging Normalized Systems Theory (NST) to improve scalability, flexibility, and reusability. Our approach integrates semantic technologies, ontologies, and linked data principles to enable structured, interoperable DMPs that can be automatically assessed and adapted. We provide a prototype implementation that demonstrates the feasibility of this approach. The key contribution of this paper is a demonstration case study, which evaluates the framework’s effectiveness in real-world scenarios, illustrating how machine-actionability can reduce manual workload, improve guidance for researchers, and support automated assessment processes. The findings highlight the potential of our solution to advance Findable, Accessible, Interoperable, and Reusable (FAIR)-aligned data stewardship practices, making data management planning more dynamic and sustainable.