Market and Production Management of Air Bearings using Ontology-based Knowledge Graph and NLP

Abstract

This research presents an ontology-based framework designed to enhance market and production management for specialized Small and Medium-sized Enterprises (SMEs) in the precision manufacturing sector, using air bearings as a case study. The framework was validated through real-world applications in a precision manufacturing SME and features a knowledge graph that was used for staff education and improved operational efficiency. Our approach integrates expert-driven ontology modeling with a modular web crawler and advanced Natural Language Processing (NLP) techniques, specifically the BERT-based ELECTRA model, for entity recognition and relationship extraction. This enables the framework to identify potential customers and competitors from unstructured online data, improving strategic decision-making. Key outcomes include enhanced supply chain transparency, more accurate market positioning, and better identification of long-tail demand patterns. The study demonstrates the practical value of combining knowledge graphs with NLP in industrial applications and suggests future extensions such as integration with large language models and the development of a domain-specific search engine.