Tokyo University of Marine Science and Technology

[Awards and Honors] Mr. Kim Jun-seong (alumnus of this university) and Professor Daisuke Watanabe received the Japan Institute of Navigation Paper Award.

School of Marine Technology Companies / researchers Current students

2026Years6Month4On that day, Mr. Kim Jun-seong, a graduate of the doctoral program at this university, and Professor Daisuke Watanabe,2025I received the Japan Institute of Navigation Paper Award for the year. This award is given to outstanding papers published in the Japan Institute of Navigation's Journal and English-language journal, which are recognized as making a significant contribution to the field of navigational science.

【Winner】
Kim Jun-seong (At the time of posting: 3rd year doctoral student in Applied Environmental Systems Engineering, Graduate School of Marine Science and Technology; Current affiliation: Incheon Maritime High School, South Korea)
Daisuke Watanabe (Department of Distribution and Information Engineering, Graduate School of Science and Technology)

[Award-winning paper]
"LNG"Application of an AI-based hybrid model to ship spot fare prediction," Transactions of the Japan Institute of Navigation, No.153issue,2025Years

[Summary of the Award-winning Paper]
 In recent years, in addition to seasonal factors such as climate change due to global warming, geopolitical risks, changes in international affairs, policy changes, and disruptions to supply chains due to the COVID-19 pandemic have also contributed to the situation.LNGShip spot fares are becoming more volatile.
 Therefore, this research uses deep learning, a type of artificial intelligence,LNGThe purpose is to predict spot freight rates for ships. Specifically, it uses long-term memory, which is commonly used in time-series data forecasting.Long Short-Term Memory (LSTM)) and gated recurrent units (Gated Recurrent Unit (GRU)) and, in previous studies, demonstrated high performance in predicting various shipping indices.CRANEand attention mechanism (Attention Mechanism, ATTWe constructed a hybrid deep learning model that combines the following:
 ThisLNGWe aimed to improve the accuracy of predicting spot freight rates for ships and identified the optimal prediction method through performance comparisons of multiple models.

Japan Institute of Navigation Award_Watanabe2026_1.jpg

Professor Watanabe 3.jpg
(From left to right in the photo) Professor Watanabe, Ms. Kim, and Takahiro Takemoto, President of the Japan Institute of Navigation.
Photo courtesy of the Japan Institute of Navigation

<Related links>
Japanese Society of Navigation

BACK TO NEWSLIST