Generative AI in Knowledge Management for Multilingual Work Environment: Case Study in Finnish Pulp Mills
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Perustieteiden korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
2024-03-13
Department
Major/Subject
Analytics and Data Science
Mcode
SCI3109
Degree programme
Master’s Programme in Industrial Engineering and Management
Language
en
Pages
75 + 12
Series
Abstract
Like many other European countries, Finland faces an increased demand for international talent to contribute to its welfare system and support companies’ expansion plans. However, this means enabling a new, multilingual work environment in the Finnish companies. This thesis is the first study to investigate the integration of generative AI, particularly Large Language Models (LLMs), into pulp mills to facilitate a multilingual work environment. Addressing the communication challenges in an industry dominated by Finnish language, this research aims to bridge the gap between native and non-native speakers, enhancing operational efficiency and inclusivity. Employing a mixed-methods approach, the study combines qualitative assessments of language barriers with quantitative analysis of AI-driven communication solutions. The key findings reveal that LLMs, enriched with company-specific data, can significantly mitigate language barriers in areas like shift kick-offs, problem-solving, and onboarding processes. These models, particularly when augmented with techniques like Retrieval Augmented Generation (RAG), demonstrate a potent capability to translate technical documentation and assist in real-time communication. However, the research highlights the critical need for up-to-date, explicit knowledge databases to ensure the AI's effectiveness and safety in the high-risk pulp mill environment. The study concludes that while generative AI presents a ground-breaking solution to language barriers, its success hinges on the comprehensive documentation of processes and continual updating of AI training material. Recommendations for future implementation include a strategic roadmap for operational management, emphasising the importance of digital literacy, updated documentation practices, and inclusivity in workforce development. This thesis contributes to the understanding of AI's potential in industrial settings and lays the groundwork for future explorations into AI-enhanced multilingual workplaces.Description
Supervisor
Luoma, JukkaThesis advisor
Käki, AnssiKeywords
Generative AI, GPT, RAG, knowledge management, strategic language management