One of the central challenges in industrial environments is capturing the know-how of experienced operators and turning it into actionable knowledge. Much of this expertise is tacit, stored in people’s minds, or scattered across documents and manuals. To make it accessible and usable by both humans and intelligent systems, PERKS (https://perks-project.eu) focuses on structured knowledge elicitation.
This process combines well-established methods such as expert interviews and workshops with modern approaches like natural language processing to extract information from existing texts. Crucially, domain experts remain at the heart of the process through human-in-the-loop validation, ensuring that the resulting knowledge is both accurate and trustworthy.
Within PERKS, this knowledge is formalized in a machine-understandable way, allowing it to be integrated into knowledge graphs and applied across different industrial use cases. The result is a digital foundation that enables operators to be guided step by step through complex procedures, supports managers in auditing and optimizing processes, and safeguards critical expertise for future generations.
Knowledge elicitation is therefore more than documentation: it is the gateway to transforming industrial know-how into transparent, explainable, and AI-enhanced solutions. This aligns closely with the PERKS vision of Industry 5.0, where human expertise and artificial intelligence work hand in hand to create safer, more efficient, and more resilient operations.
To explore this topic further, check out the PERKS podcast: How Knowledge Elicitation supports Industrial Applications available on the project website: https://perks-project.eu/podcasts-videos/