ABSTRACT
A method is proposed to automatically produce synthetic maritime traffic representations from historical Automatic Identification System data. The method consists of a two-phase analysis of maritime traffic data, which includes ship traffic motion pattern identification as well as ship route characterization. Given the location of a port, ship trajectories and ship characteristics are extracted enabling to estimate ship traffic volume, composition and origin-destination topology. In addition, each ship route is individually analysed in terms of ship type distribution and daily and hourly patterns, which allows the characterization of the typical behaviour of a group of similar ships along a particular route. Furthermore, the speed distribution at the study area in terms of time of day is analysed to capture the traffic state changes over time. The method is applied to obtain a light and structured representation of the maritime traffic off the Azores Archipelago.
