Enhancing Asset-Oriented Digital Twins with AI-Agents for Intelligent Configuration and Decision Support

Asset-oriented digital twins (DTs) represent individual physical assets—such as machines, vehicles, or infrastructure components—by integrating real-time data, semantic models, and operational workflows.

The integration of AI agents into these DTs enables advanced assistance in configuring complex systems, orchestrating workflows, and automating maintenance cycles. Leveraging well-documented open-source tools (e.g., CWL, Argo, Airflow), AI agents can design and manage data pipelines, ensuring seamless data collation and contextualisation across heterogeneous sources.

By utilising ontologies and semantic enrichment, these agents lower the barrier to understanding technical data, making it accessible to engineers and operators. Furthermore, AI agents can analyse asset conditions, recommend optimal actions, and translate decisions into executable commands, thus enhancing operational efficiency and supporting predictive maintenance strategies.