Research Topic
LLM-Driven Process Safety Intelligence
With the rapid advancement of artificial intelligence and large language models, industrial systems are entering a new era of intelligent decision support and automated knowledge generation. Process safety, however, remains heavily dependent on expert experience, manual hazard analysis, and fragmented safety information. In recent years, researchers have begun exploring the integration of AI into safety critical engineering applications, particularly for hazard identification, accident prediction, and operational decision making. This research seeks to develop LLM driven frameworks for process safety that combine process knowledge, graph based system representations, and multimodal operational data to generate, evaluate, and explain potential accident scenarios. Such an approach would support both proactive risk assessment and adaptive human AI collaboration for safer and more sustainable industrial operations.
