A research team from the Department of Construction Systems Engineering at our university’s Graduate School—including Ph.D. candidate Park Ji-yeon and master’s students Jang Sang-beom, Kim Ji-hwan, Kim Seo-young, and Moon Kyung-min (Advisor: Shin Ju-Young)—received the Best Paper Award at the 2026 Annual Conference of the Korean Society of Water Resources (May 20–22, 2026).

Ph.D. candidate Park Ji-yeon presented “Development of a Sewer Water Level Prediction Technique Using Cross-Domain Artificial Intelligence” in the Rainfall Runoff session. This research introduced an AI technique that combines data from different domains to improve the accuracy of sewer water level predictions, aiming to contribute to the advancement of real-time prediction systems for urban flooding response.

Master’s student Jang Sang-beom presented “Development of Scale-Aware Multimodal AI for Rainfall Nowcasting” in the Student Competition Session. This research proposed a multimodal AI model capable of integrally processing meteorological information across various spatial scales, focusing on improving the accuracy of short-term rainfall prediction (nowcasting).

Master’s student Kim Ji-hwan presented “Derivation of Probabilistic Precipitation-Area-Duration (DAF) Curves Using Area Covariate Parameters of the GEV Distribution” in the student competition session. This study presented a methodology for calculating probabilistic DAF curves that account for watershed scale by introducing area as a covariate in the parameters of the GEV distribution, and is expected to contribute to the advancement of design standards for hydraulic structures.

Master’s student Kim Seo-young presented “A Study on a Model for Predicting Performance Degradation of River Structures Over Their Lifecycle Using Survival Analysis” at the student competition session. This study applied survival analysis techniques, primarily used in the medical and insurance fields, to river facilities, quantitatively modeling performance degradation patterns and demonstrating the potential for life-cycle-based maintenance decision support.

Master’s student Moon Kyung-min presented “Development of the Qwen-RAG Framework for Q&A Support in Dam Operation Regulations: Analysis of Document Segmentation and Search Strategy Effects” in the student competition session. This study applied a large language model (LLM)-based RAG (Retrieval-Augmented Generation) framework to dam operation regulations to explore the feasibility of building an intelligent question-answering system for complex water infrastructure operation guidelines.

This award serves as a testament to the laboratory’s broad research capabilities across various fields, including rainfall-runoff modeling, short-term rainfall forecasting, probabilistic hydrology, lifecycle management of river structures, and water infrastructure management. It once again highlights the research achievements of our university’s Convergence Water Resources Engineering Laboratory, which contributes to solving key issues in the water resources sector in the era of climate change.