Modeling Tacit and Implicit Procedural Knowledge with Knowledge Graphs: a Human‑Centric Semantic Approach

Invited talk for the GOBLIN COST Action given on February 26th, 2026 Industrial procedures rely simultaneously on tacit knowledge held by workers and implicit knowledge embedded in textual or semi-structured artefacts. This talk presents a semantic approach for formalising such heterogeneous procedural know-how through a dedicated Procedural Knowledge Ontology (PKO) and its instantiation as integrated knowledge graphs. The methodology combines expert elicitation, semi-automatic extraction techniques, and human-in-the-loop validation to ensure contextual fidelity and operational usefulness; the resulting KG is then exploited by AI-powered tools to support operators during task execution. The approach has been applied across three industrial scenarios (manufacturing, automation, energy), providing insights and lessons learned on how knowledge-graph-based representations can enhance reliability, onboarding, and decision support in manufacturing and industrial environments. Recording of this talk is available at: https://www.youtube.com/watch?v=Ewfrl7PVrs0.