Efficient AIS Data Processing for Environmentally Safe Shipping

Abstract

Reducing ship accidents at sea is important to all economic, environmental, and cultural sectors of Greece. Despite an increase in traffic and national monitoring, ships formulate routes according to their best judgment risking an accident. In this study we take a dataset spanning in 3 years from the AIS (Automatic Identification System) network, which is transmitting in public a ship's identity and location with an interval of seconds, and we load it in a trajectory database supported by the Hermes Moving Objects Database (MOD) system. Presented analysis begins by extracting statistics for the dataset, both general (number of ships and position reports) as well as safety related ones. Simple queries on the dataset illustrate the capabilities of Hermes and allow to gain insight on how the ships move in the Greek Seas. Analysis of movement based on an Origin-Destination matrix between interesting areas in the Greek territory is presented. One of the newest challenges that emerged during this process is that the amount of the positioning data is becoming more and more massive. As a conclusion, a preliminary review of possible solutions to this challenge along with others such as dealing with the noise in AIS data is mentioned and we also briefly discuss the need for interdisciplinary cooperation.
JEL Classifications: Z00