Yi Yuan , Hetian Guo , Zipei Fan , Yingzhi Peng , Jiaqi Zhang , Xuan Song , Ryosuke Shibasaki
{"title":"A decomposable multi-fusion spatio-temporal marine traffic flow forecasting algorithm: Taking the North Sea and Baltic Sea region as an example","authors":"Yi Yuan , Hetian Guo , Zipei Fan , Yingzhi Peng , Jiaqi Zhang , Xuan Song , Ryosuke Shibasaki","doi":"10.1016/j.tre.2025.104340","DOIUrl":null,"url":null,"abstract":"<div><div>Maritime transportation, which is responsible for handling over 80% of global trade, is integral to the international supply chain. As global trade continues to expand, the management of marine traffic has become crucial for improving logistics efficiency. Precise and accurate marine traffic flow (MTF) prediction is essential for optimizing shipping routes, reducing transit times, and improving overall supply chain effectiveness. AIS equipment is now mandatory on vessels, transmitting static data such as the vessel’s identification code and flag state, along with dynamic data including latitude, longitude, speed and heading. High-frequency and comprehensive AIS data enables more accurate predictions of maritime traffic flow. In this paper, we propose an end-to-end MTF prediction framework utilizing AIS data as the primary source. The algorithm encompasses three critical steps: data preprocessing, marine traffic network extraction, and maritime traffic flow prediction. This comprehensive approach ensures more accurate predictions of the maritime traffic flow. In the marine traffic network extraction process, we distinguish nodes by categorizing AIS data feature points into mooring points and waypoints to obtain a more accurate identification. We then introduce a Decomposable Multi-fusion Spatio-temporal Network (DMFSTN) to enhance the accuracy of maritime traffic flow predictions. Existing approaches do not solve the relationship between static and dynamic features well. At the same time, the often-used serialized extraction of spatial and temporal features tends to overlook fine-grained spatio-temporal dependence features. To address these issues, our DMFSTN model integrates temporal and spatial features and leverages dynamic and static spatial relationships for more precise MTF predictions. Additionally, our model also decomposes marine flow data into trend and seasonal components, offering insights into underlying patterns. As a case study, we apply our model to analyze MTF of the North Sea and Baltic Sea region using data provided by the Danish Maritime Administration. Extensive comparative and ablation experiments in this dataset demonstrate the effectiveness of our model.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"203 ","pages":"Article 104340"},"PeriodicalIF":8.8000,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transportation Research Part E-Logistics and Transportation Review","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1366554525003813","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ECONOMICS","Score":null,"Total":0}
引用次数: 0
Abstract
Maritime transportation, which is responsible for handling over 80% of global trade, is integral to the international supply chain. As global trade continues to expand, the management of marine traffic has become crucial for improving logistics efficiency. Precise and accurate marine traffic flow (MTF) prediction is essential for optimizing shipping routes, reducing transit times, and improving overall supply chain effectiveness. AIS equipment is now mandatory on vessels, transmitting static data such as the vessel’s identification code and flag state, along with dynamic data including latitude, longitude, speed and heading. High-frequency and comprehensive AIS data enables more accurate predictions of maritime traffic flow. In this paper, we propose an end-to-end MTF prediction framework utilizing AIS data as the primary source. The algorithm encompasses three critical steps: data preprocessing, marine traffic network extraction, and maritime traffic flow prediction. This comprehensive approach ensures more accurate predictions of the maritime traffic flow. In the marine traffic network extraction process, we distinguish nodes by categorizing AIS data feature points into mooring points and waypoints to obtain a more accurate identification. We then introduce a Decomposable Multi-fusion Spatio-temporal Network (DMFSTN) to enhance the accuracy of maritime traffic flow predictions. Existing approaches do not solve the relationship between static and dynamic features well. At the same time, the often-used serialized extraction of spatial and temporal features tends to overlook fine-grained spatio-temporal dependence features. To address these issues, our DMFSTN model integrates temporal and spatial features and leverages dynamic and static spatial relationships for more precise MTF predictions. Additionally, our model also decomposes marine flow data into trend and seasonal components, offering insights into underlying patterns. As a case study, we apply our model to analyze MTF of the North Sea and Baltic Sea region using data provided by the Danish Maritime Administration. Extensive comparative and ablation experiments in this dataset demonstrate the effectiveness of our model.
期刊介绍:
Transportation Research Part E: Logistics and Transportation Review is a reputable journal that publishes high-quality articles covering a wide range of topics in the field of logistics and transportation research. The journal welcomes submissions on various subjects, including transport economics, transport infrastructure and investment appraisal, evaluation of public policies related to transportation, empirical and analytical studies of logistics management practices and performance, logistics and operations models, and logistics and supply chain management.
Part E aims to provide informative and well-researched articles that contribute to the understanding and advancement of the field. The content of the journal is complementary to other prestigious journals in transportation research, such as Transportation Research Part A: Policy and Practice, Part B: Methodological, Part C: Emerging Technologies, Part D: Transport and Environment, and Part F: Traffic Psychology and Behaviour. Together, these journals form a comprehensive and cohesive reference for current research in transportation science.