Prediction of vessel arrival time to port: a review of current studies

Accurate prediction of vessel arrival time is crucial for the efficiency of port operations and international trade; however, a systematic review of the research on this topic has not yet been conducted. This paper provides the first systematic review of the prediction of vessel arrival time to port, encompassing 29 academic studies published since 2011. By reviewing the literature, it first identifies and categorizes six key factors affecting vessel arrival time prediction: vessel static information, dynamic information, route conditions, environmental conditions, human factors, and external unexpected factors. The review highlights the challenges of precise vessel arrival time prediction. After closely examining the existing research studies, we found that two frameworks, namely non-trajectory-based prediction of vessel’s ETA to port and prediction of vessel’s ETA to port by path finding, and four categories of prediction models are commonly used: statistical models, machine learning models, deep learning models, and reinforcement learning models. This review also explores potential future research directions and serves as a critical resource for feature usage, model selection, the current research state, and future development directions for researchers, industry practitioners, and policymakers, advancing the understanding and application of data-driven prediction methodologies for improved maritime operational efficiency and digitalization.