Virtual Process Dossier: A Process-Aware Data Catalogue

2026-07-30

AI Summary

This work addresses the challenge in multi-stage manufacturing scenarios where downstream AI tasks struggle to accurately retrieve stage-specific data due to the absence of explicit provenance about data generation processes. To bridge this gap, the authors propose the Virtual Process Dossier (VPD) framework, which constructs a knowledge graph–based data catalog grounded in FAIR principles and enriched with workflow-level provenance information. Key innovations include the design of a manufacturing-oriented VPD ontology, integration of a production-environment-compatible provenance capture mechanism, and development of an interactive knowledge graph interface. The framework enables process-level organization and traceable access to manufacturing data, substantially improving data suitability for AI applications. The associated code and ontology have been made openly available.

Abstract

We propose the Virtual Process Dossier (VPD), a Knowledge Graph-based data catalogue that also captures workflow provenance. We developed VPD for multi-stage manufacturing use-cases where downstream AI-based optimization tasks require to distinct between datasets generated during individual workflow steps. VPD provides these datasets in a FAIR manner and makes both prospective and retrospective workflow provenance explicit. Our contributions are: (1) the VPD ontology that serves as the catalogue's semantic core; (2) the VPD provenance framework that integrates ontology instantiation into the production environment; and (3) the VPD user interface that provides human-centered interaction with the VPD Knowledge Graph. The ontology and code are available at https://github.com/kubeluk/VirtualProcessDossier .

Problem

• data catalogue
• workflow provenance
• multi-stage manufacturing
• FAIR data
• Knowledge Graph

Innovation

• Knowledge Graph
• Workflow Provenance
• FAIR Data
• Ontology
• Data Catalogue