Ships are prone to accidents when approaching in a berthing velocity greater than that allowed when determining the risk range corresponding to a port. Therefore, this study develops a machine learning strategy to predict the risk range of an unsafe berthing velocity when the ship approaches in port. To perform analysis, the input parameters were based on the factors affecting the berthing velocity, and the output parameter, i.e., the berthing velocity, was measured at a tanker terminal in the Republic of Korea. Nine machine learning classification algorithms were used to analyze each model, and the top four optimal models were selected through evaluation methods based on the confusion matrix. As a result of the analysis, extra trees, random forest, bagging, and gradient boosting classifiers were identified as good models. As a result of testing using the receiving operator characteristic curve, it was confirmed that the area under the curve of the most dangerous range of berthing velocity was the highest, thus, the risk range was appropriately classified. As such, the derived models can classify and predict the risk range of unsafe berthing velocity before approaching a port; therefore, it is possible to safely berth a ship.

A maritime route is used by sea transportation vessels to access the trading ports, and route design standards for the safety of maritime traffic have been established in various countries and organizations. However, no quantitative safety verification method related to route design currently exists. In this study, a novel maritime route was created and compared with the original route in Incheon, the Republic of Korea, based on the relevant automatic identification system (AIS) data. The attendant traffic density was revealed via kernel density estimation analysis of the AIS data, with the results used to create the boundary of the novel route through an image processing technique. The boundary and the centerline of the maritime route were determined using a line smoothing technique. For safety verification, the centerline of the original route and that of the novel maritime route were compared in terms of sinuosity, intersection angle, and route change envelope (RCE). The sinuosity analysis demonstrated that the route was stable in terms of the outer harbor limit, while the intersection angle analysis demonstrated that the novel maritime route intersection angle was stable. The RCE was used to objectively compare the absolute values of the distance change in the centerline.

The recent emergence of futuristic ships is the result of advances in information and communication technology, big data, and artificial intelligence. They are generally autonomous, which has the potential to significantly improve safety and drastically reduce operating costs. However, the commercialization of Maritime Autonomous Surface Ships requires the development of appropriate technologies, including intelligent navigation systems, which involves the identification of the current maritime traffic conditions and the prediction of future maritime traffic conditions. This study aims to develop an algorithm that predicts future maritime traffic conditions using historical data, with the goal of enhancing the performance of autonomous ships. Using several datasets, we trained and validated an artificial intelligence model using long short-term memory and evaluated the performance by considering several features such as the maritime traffic volume, maritime traffic congestion fluctuation range, fluctuation rate, etc. The algorithm was able to identify features for predicting maritime traffic conditions. The obtained results indicated that the highest performance of the model with a valid loss of 0.0835 was observed under the scenario with all trends and predictions. The maximum values for 3, 6, 12, and 24 days and the congestion of the gate lines around the analysis point showed a significant effect on performance. The results of this study can be used to improve the performance of situation recognition systems in autonomous ships and can be applied to maritime traffic condition recognition technology for coastal ships that navigate more complex sea routes compared to ships navigating the ocean.

For ship passage safety, a bridge across a waterway is a risk for accidents. However, no standard for the safe distance between a bridge and a ship is available in Korea. The UK MCA considers the 90% confidence interval of traffic distribution as the acceptable passage range, using it for measuring the separation between offshore wind farms. In this study, an optimal ship safety distance is proposed by evaluating traffic distribution at the Incheon and Busan harbor bridges and analyzing the confidence intervals. The results, based on the Z-score, reveal that at the Incheon bridge, all but one ship for 2-way departure were in circulation within the 95% confidence interval range, whereas at the Busan harbor bridge, six ships for arrival and two ships for departure were outside the 95% range. Based on the results of this study, the design of bridges across waterways can incorporate traffic distribution corresponding to each port.

The size of ships is increasing rapidly, and over 400 m length overall mega-sized container ships are predicted to appear in the near future. Consequently, studies on large container ships have been conducted; however, based on the 30,000-TEU (twenty-foot equivalent unit) class container ship’s length overall, the deviation of the predicted range showed an 83.9 m difference from 453.0 m to 536.9 m. This is because simple linear regression analysis does not consider trends according to the type of cargo or the size of the cargo. In this study, 5497 container ships up to 20 years of age with an International Maritime Organization (IMO) number registered were clustered according to the change in ship dimensions by k-means clustering algorithm. Based on the clustered data, deadweight tonnage, TEU, length overall, length between perpendiculars, breadth, and maximum draft of container ships with a coverage rate of 75% were analyzed to predict the change in the main dimensions. The results indicated that for a 30,000-TEU container ship, the predicted length overall is 428.4 m, breadth is 67.6 m, and draft is 17.0 m. This study can help minimize the social costs of designing ports in consideration of future mega-sized container ships.