A novel quantitative analysis employing the Principal Component Analysis (PCA) of containership traffic in the Black Sea from 2018 to 2021 is performed. The study uses a matrix covering five ship size classes from A to E for four years of operation, from 2018 to 2021, accounting for ship traffic, CO2, fuel consumption (FC), shipping intensity, and eco and traffic efficiency. Only the first two principal factors are analysed because of their total variation weight. Shipping intensity, FC intensity, and CO2 intensity plays a significant role in the first factor, while Eco efficiency, FC efficiency, and Traffic efficiency are considered for the second factor. Notably, the set of parameters pertains to time and is strongly associated with DWT. Two principal components were identified, F1 and F2, where F1 integrates efficiency and intensity. At the same time, F2 separates the intensity from the efficiency conditional on the ship size and the year of operations. In the principal component F1 the activities of ships A and C differ from B, D and E, separating more efficiently from less efficiently used ships, and in F2, the activities of class sizes of ships C and D and E contrast A and B ships, distinguishing the big-size class ships from small ones. It was concluded that the most intensively used ships are the ship size classes C and D, and the most efficient are ship size classes A and B. The most intensive use of the ships was in 2020, followed by 2021, and the most efficient were in 2018, 2019. Based on the ship activities and using the Within-class variance, ships are grouped into two clusters of similar activities, where the first one, with lower variance and more homogeneous, includes only the ship size class A. The second one with a relatively large variance consists of the rest size of the ships. Linear relationships considering the intensity and efficiency are derived as a function of the main variables, where the factor loading represents the variable’s coefficient, given as a relative weight to any factor.
Read full abstract