Large Language Models for Tacit Knowledge Elicitation in Industry 5.0: A Literature Review

Abstract

As technological advances drive rapid change towards a human-centric Industry 5.0, integrating human expertise into intelligent systems is essential for adaptive, efficient and resilient operations. This paper investigates the role of Large Language Models (LLMs) in knowledge management, focusing on their ability to elicit tacit knowledge. Through a literature review, current methods for elicitation are explored in dynamic manufacturing environments and it is examined how LLMs can support this process.Tacit knowledge has long been a critical but elusive asset in manufacturing. Traditional methods of eliciting tacit knowledge require significant resources in time and personnel. In this context, LLMs emerge as a promising tool by using natural language processing to engage with operators.The paper examines key challenges, including ensuring operator acceptance of conversational agents. By incorporating operator insights, manufacturers can build an ever-expanding knowledge base that enhances decision-making and operational support. The extracted knowledge can serve as the basis for improving human-machine collaboration and allows continuous refinement of the knowledge base.By providing a thorough review of the current state of tacit knowledge acquisition in manufacturing and analyzing LLM applications, this paper highlights the challenges and opportunities for future developments. Addressing these challenges enables LLMs to bridge the gap between human expertise and increasingly complex production systems, thereby supporting the human-centric vision of Industry 5.0.

Keywords: Tacit Knowledge, Industry 5.0, Human-centric systems, Large Language Models (LLMs), Knowledge Management

DOI: 10.54941/ahfe1006404