Detecting Anomalous Vessel Dynamics with Functional Data Analysis

Advances in location-acquisition technology open up new areas of applications in maritime monitoring and security. Automatic identification system (AIS) data provide dynamic information on vessel movements. This research proposes a new method for detecting anomalous vessel dynamics using functional data analysis. Empirical investigations of this approach demonstrate the effective detection of outlier flows in terms of ship traffic volume. However, alternative methods such as the 3-sigma rule and the MAD-Median rule fail to detect anomalous vessel traffic. This investigation suggests that the method proposed can improve the safety and operation of ship-to-shore vessel traffic management.