{"id":20323,"date":"2026-10-21T13:37:57","date_gmt":"2026-10-21T13:37:57","guid":{"rendered":"https:\/\/sominnport.cat\/?post_type=cas&p=20323"},"modified":"2026-10-21T13:59:25","modified_gmt":"2026-10-21T13:59:25","slug":"ai-to-optimise-maritime-routes-and-minimise-the-environmental-impact-of-navigation","status":"publish","type":"cas","link":"https:\/\/sominnport.cat\/en\/observatori\/ai-to-optimise-maritime-routes-and-minimise-the-environmental-impact-of-navigation\/","title":{"rendered":"AI to Optimise Maritime Routes and Minimise the Environmental Impact of Navigation"},"content":{"rendered":"\n

Project Leaders<\/strong><\/h2>\n\n\n\n

The project has been developed by researchers at Pusan National University (South Korea). The study was led by Assistant Professor Dowon Kim, together with doctoral student Seongbeom Park and Professor Jinhyeok Yun.<\/p>\n\n\n\n

Project Description<\/strong><\/h2>\n\n\n\n

This innovative project develops an artificial intelligence system to optimise vessel routes and speeds<\/strong>, simultaneously taking into account both fuel consumption levels and environmental conditions in coastal areas.<\/p>\n\n\n\n

The proposal addresses a specific challenge: the impact of navigation operations on the environment depends not only on the emissions generated but also on meteorological conditions, which influence their dispersion. Strategies based solely on general reductions in vessel speed may increase navigation time without having a positive effect on the emissions that may reach the coast. Therefore, local meteorological conditions must be taken into account.<\/p>\n\n\n\n

To address this challenge, the system combines artificial intelligence techniques based on physical principles (physics-informed deep learning<\/em>) with multi-objective optimisation methods. It uses environmental data collected through sensors to reconstruct airflow fields in near real time and, based on this information, predict the behaviour of emissions present in the atmosphere under different meteorological conditions.<\/p>\n\n\n\n

Based on these predictions, a multi-objective Bayesian optimisation system (a method that compares different alternatives to identify the most suitable ones according to various factors) determines route and speed combinations that seek to balance fuel consumption in order to minimise the accumulation of emissions according to meteorological conditions.<\/p>\n\n\n\n

The researchers refer to this approach as temporal navigation<\/em>, as it allows the vessel’s route and speed to be adapted to current meteorological conditions rather than applying a uniform speed reduction.<\/p>\n\n\n\n

The system has been tested in different navigation scenarios, including simulations in the area surrounding the Port of Busan, South Korea. In these scenarios, the AI-based approach achieved a 20-35% improvement in route optimisation performance and a 34-78% reduction in maximum exposure to the pollutant emissions studied<\/strong>, compared with conventional navigation strategies or the corresponding reference scenarios.<\/p>\n\n\n\n

The project’s main innovation lies in incorporating local environmental conditions into navigation decisions, rather than focusing optimisation solely on fuel consumption or total emissions. In this way, the methodology takes into account both the characteristics of navigation operations and meteorological conditions, which influence the behaviour and distribution of emissions present in the atmosphere.<\/p>\n\n\n\n

This case demonstrates how the combination of artificial intelligence, environmental data and navigation optimisation techniques can contribute to improving operational efficiency by incorporating environmental criteria into maritime route planning.<\/p>\n\n\n\n

Project Objectives<\/strong><\/h2>\n\n\n\n

The main objective is to develop an artificial intelligence system capable of optimising vessel routes and speed profiles to reduce fuel consumption and minimise exposure to emissions during maritime operations.<\/p>\n\n\n\n

The project also aims to incorporate local meteorological conditions into navigation decisions, facilitating strategies that are better adapted to environmental conditions and contributing to more efficient maritime operations that are more respectful of the coastal environment.<\/p>\n\n\n\n

Publication Date<\/strong><\/h2>\n\n\n\n

21 October 2026<\/p>\n\n\n\n

Links of Interest<\/strong><\/h2>\n\n\n\n