Container Ship Fleet Route Evaluation and Similarity Measurement between Two Shipping Line Ports

Abstract

The characterization of ship routes and route similarity measurement based on Automatic Identification System (AIS) data are topics of various scientific interests. Common route research approaches use available AIS identifiers of ship types. However, assessing route and similarity profiles for individual fleets requires collecting data from secondary sources, dedicated software libraries or the creation of specific methods. Using an open-source approach, public AIS and ship data, we evaluate route characteristics for the container ships of a single fleet in a six-month period, calling on two selected ports of the shipping line on the USA East Coast. We evaluate the routes in terms of length, duration and speed, whereas for the similarity measurement we employ the discrete Fréchet distance (DFD). The voyage length, duration and average speed distributions were observed to be moderately positive (0.77), negative (−0.62), and highly positively skewed based on the adjusted Fisher–Pearson coefficient of skewness (1.23). The most similar voyages were from the same ships, with the lowest discrete Fréchet distance similarity value (0.9 NM), whereas 2 different ships had the most dissimilar voyages, with the highest DFD value (14.1 NM). The proposed methodology enables assessment of similarities between individual ships, or between fleets.

Vessel Trajectory Similarity Computation Based on Heterogeneous Graph Neural Network

“…Similarly, for Fréchet distance, Cao et al [10] used discrete Fréchet distance to solve the maximum distance between vessel trajectories and obtained a distance matrix, which was then decomposed using PCA to determine trajectory cluster. Šakan et al [26] studied route characteristics by using discrete Fréchet distance to measure the similarity of individual container fleets from a statistical perspective. Considering that DTW and discrete Fréchet distance are the most commonly used classical vessel trajectory distance measures and they are based on different similarity calculation ideas (warping and shape), this article chooses them as experimental distance measures for approximation.…”