Pohang University of Science and Technology

This project focuses on developing a large-scale foundation model tailored for smart manufacturing environments. By integrating multimodal industrial data—including process logs, sensor data, and textual production knowledge—the research aims to construct an AI-driven framework that supports adaptive decision-making, predictive maintenance, and intelligent process optimization. The project combines expertise in industrial engineering, artificial intelligence, and data science to build a robust foundation model that generalizes across various manufacturing domains. It also explores collaborative human–AI interaction for sustainable and efficient industrial operations.

This project applies reinforcement learning, specifically the Bandit Algorithm, to build a dynamic fashion item recommendation system. By learning adaptive strategies to maximize user engagement, the system continually refines its suggestions in real time, enhancing personalization and customer satisfaction in online retail environments.

This study explores process modeling, simulation, and optimization using object-centric process mining. By leveraging object-centric event logs, the research aims to derive more realistic process models, identify inefficiencies, and develop simulation frameworks that reflect complex process dynamics. The project integrates industrial engineering principles with artificial intelligence to generate advanced optimization techniques and actionable insights for process improvement.