Abstract
Container shipping plays a pivotal role in global trade, and understanding the duration that vessels spend in ports is crucial for efficient voyage planning by shipping companies. However, these companies often rely solely on one-way communication for required arrival times provided by terminals. This reliance on fixed schedules can lead to vessels arriving punctually, only to face berths that are still occupied, resulting in unnecessary waiting times. Regrettably, limited attention has been given to these issues from the perspective of shipping companies. This study addresses this gap by focusing on the estimation of dwell times for container vessels at a terminal in the Port of Busan using various machine learning techniques. The estimations were compared against the terminal’s operational reference. To compile the dataset, a 41-month history of terminal berth schedules and vessel particulars data were utilized and preprocessed for effective training. Outliers were removed, and dimensions were reduced. Six regression machine learning algorithms, namely adaptive learning, gradient boosting, light gradient boosting, extreme gradient boosting, categorical boosting and random forest, were employed, and their parameters were fine-tuned for optimal performance on the validation dataset. The results indicated that all models exhibited superior performance compared to the terminal’s operating reference model.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.