A description logic approach to discover suspicious itineraries from maritime container trajectories

Abstract

About 90% of the world's cargo is transported in maritime containers, but less than 2% is physically inspected by custom authorities. The standard method to handle this problem consists in document-based risk analysis and route-based risk indicators to target anomalies. In this paper, we exploit a logic based approach to identify suspicious patterns in container itineraries. Specifically, we present an ontology to explicitly formalize the knowledge of the maritime container domain and a formalisation of two suspicious movement patterns to enable their discovery in a knowledge base. The formalisation can be extended to support the discovery of other itinerary patterns. Furthermore, the approach we present can be the basis for future development towards the formalisation and search of patterns in itineraries. © 2011 Springer-Verlag.