Notice

Gyeongho Kim, a Ph.D. student in Industrial Engineering at UNIST (President Park Jong-rae), was selected for an outstanding presentation paper for his research presented at the Spring Conference of the Korean Society for Reliability held from June 18-20, 2025.

The research selected as an outstanding presentation paper is titled ‘Development of a robust tool wear prediction method under novel operating conditions using deep unsupervised domain adaptation and physics-guided adjustment’. This study proposes an accurate and robust tool wear prediction methodology under novel operating conditions. By utilizing deep learning-based unsupervised domain adaptation methodology and physics-informed prediction adjustment techniques to predict tool wear, this research can help improve productivity and process efficiency in manufacturing and machining processes. In particular, it is expected to help determine real-time tool replacement timing in environments where various operating conditions are utilized.

Student Gyeongho Kim stated that based on the novel tool wear prediction methodology under new operating conditions proposed in this research, he plans to develop a foundation model for robust tool prognostics that can withstand various input and environmental changes.