Recent research led by Rosa Martínez from the Centro Tecnológico Naval y del Mar has introduced an innovative AI-based method to enhance maritime logistics, particularly in the context of vessel sailing times and underwater acoustic pressure. This study, published in ‘Engineering Proceedings,’ focuses on the Port of Cartagena, Spain, and utilizes historical vessel data to create a machine learning model that helps predict sailing times more accurately.
The maritime industry, which accounts for about 80% of global trade, faces significant challenges with delays and inefficiencies that disrupt market chains and contribute to environmental concerns, such as greenhouse gas emissions and underwater noise pollution. Martínez highlights the importance of addressing these issues, stating, “Delays contribute to poor port optimization, disruptions in the market chain, and increased pollution.” The research aligns with the International Maritime Organization’s (IMO) strategy for reducing greenhouse gas emissions from shipping, aiming for a 50% reduction by 2050 compared to 2008 levels.
By leveraging machine learning techniques, the study employs a Random Forest model that achieved an impressive R2 score of 0.85 and a mean squared error (MSE) of 0.145. This accuracy is crucial for shipping companies looking to optimize their operations, as it allows for better planning and reduced idle times in ports. The implications are significant: by implementing such AI-driven tools, ports can improve efficiency, reduce costs, and enhance their environmental footprint.
Moreover, the research includes an underwater acoustic propagation model to assess noise pressure generated by vessels, addressing environmental regulations and public concerns about marine ecosystems. Martínez notes, “This study aligns with the MSFD, particularly regarding Descriptor 11, searching for a balance between optimizing economic marine activities with good environmental status.”
The findings present commercial opportunities for maritime stakeholders, including shipping companies and port authorities. By adopting these AI models, they can improve their operational efficiency and compliance with environmental standards, ultimately leading to a more sustainable maritime industry. Future developments in this area may incorporate more advanced models for even greater accuracy in both sailing time predictions and noise impact assessments.
The research not only contributes to the academic field but also serves as a practical guide for maritime professionals aiming to leverage technology for better logistics and environmental stewardship. The potential for AI to transform maritime logistics is clear, and as Martínez and her team continue to refine their models, the industry may see significant advancements in efficiency and ecological responsibility.
