Aldo Napoli

The automatic identification system of maritime accident risk using rule-based reasoning

Current maritime traffic monitoring systems are not sufficiently adapted to the identification of maritime accident risk. It is very difficult for operators responsible for monitoring traffic to identify which vessels are at risk among all the shipping traffic displayed on their screen. They are overwhelmed by huge amount of kinematic ship data to be decoded. To improve this situation, this paper proposes a system for the automatic identification of maritime accident risk. The system consists of two modules. The first automates expert knowledge acquisition through the computerized exploration of historical maritime data, and the second provides a rule-based reasoning mechanism.

An enhanced spatial reasoning ontology for maritime anomaly detection

Although originally conceived as a conceptual object for modelling knowledge, current ontologies do not make it possible to manipulate spatial knowledge. However, spatial knowledge is an essential component of any modelling specification. This problem provided the motivation for the creation of an expert system driven by an ontology. The system enables experts in the maritime domain to characterise abnormal ship behaviour based on formal semantic properties. Users are able to specify and execute spatial rules that are directly integrated into the ontology and a map interface linked to the ontology displays the results of the inferences obtained.

Detection of false AIS messages for the improvement of maritime situational awareness

The Automatic Identification System (AIS) was initially designed for safety and security of navigation purposes. However it was progressively also used for other objectives, such as surveillance, and thus led to the discovery of behaviors such as the falsification of the AIS messages by people that have been carrying out illegal activities and will to keep their activities up in an hidden way. In addition, the messages contain erroneous data and undergo spoofing attacks. The paper introduces the quality dimensions of data that shall be used in a quality assessment of AIS messages, in order to point out the dubious ones. The principles of a methodological approach for the detection of such data errors and falsifications are introduced.

From Movement Data to Objects Behavior Using Semantic Trajectory and Semantic Events

With the increasing availability of movement data, movement pattern analysis has recently received much attention. While existing movement pattern analyses typically rely only on mobile object positions, recent studies expressed the need to enrich trajectories with semantic information to improve the understanding of movement behaviors. This article extends the idea of semantic trajectories by incorporating the concept of semantic events. The main contribution of this article is the design of a framework where semantic trajectories and semantic events are linked to provide an enriched description of a situation. To test this approach, a prototype was developed where experts can model and then analyze vessels’ trajectories. Results show that using such semantic approach allows for a richer description of both trajectories and events.

The contribution of bayesian networks to manage risks of maritime piracy against oil offshore fields

In recent years pirate attacks against shipping and oil fields have continued to increase in quantity and severity. For example, the attack against the Exxon Mobil oil rig in 2010 off the coast of Nigeria ended in the kidnap of 19 crew members and a reduction in daily oil production of 45,000 barrels, which resulted in an international rise in the price of oil. This example is a perfect illustration of current weaknesses in existing anti-piracy systems. The SARGOS project proposes an innovative system to address this problem. It takes into account the entire threat treatment process; from the detection of a potential threat to implementation of the response. The response to an attack must take into account all of the many parameters related to the threat, the potential target, the available protection resources, environmental constraints, etc. To manage these parameters, the power of Bayesian networks is harnessed to identify potential countermeasures and the means to manage them.

Integration of a Bayesian network for response planning in a maritime piracy risk management system

This article describes an innovative system to protect offshore oil infrastructure against maritime piracy. To detect and respond efficiently to this threat, many factors must be taken into account, including the potential target, the protection methods already in place and operational and environmental constraints, etc. To improve the handling of this complex issue, we have designed a system to manage the entire processing chain; from threat identification to implementation of the response. The system implements Bayesian networks in order to capture the multitude of parameters and their inherent uncertainties, and to identify and manage potential responses. This article describes the system architecture, the integrated Bayesian network and its contribution to response planning.

On the Interest of Data Mining for an Integrity Assessment of AIS Messages

Put in place by the International Maritime Organization, the Automatic Identification System is a worldwide maritime electronic system that sends radio broadcasted messages at a high rate between the stations, either on board the vessels or on shores. However, some misuses of the system such as identity theft, localization spoofing or disappearances have been demonstrated. The high rate of transmission implies a considerable amount of data to process in order to point out those irregularities. This paper proposes a method based on data mining and clustering methods combined to an integrity assessment of AIS messages for anomaly detection, with a proposition of software architecture for a data processing done both on-the-fly and with archived data. The computation of confidence coefficients and the use of data mining techniques will lead to behaviour characterization with the purpose of enhance the maritime situational awareness.

Bayesian Networks in the Management of Oil Field Piracy Risk

In recent years, pirate attacks against shipping and oil field installations have become more frequent and more serious. The SARGOS system provides an innovative solution that addresses the problem from the perspective of the entire processing chain; from the detection of a potential threat to the implementation of a response. The response to an attack must take into account multiple variables: the characteristics of the threat and the potential target, existing protection tools, environmental constraints, etc. The potential of Bayesian networks is used to manage this large number of parameters and identify appropriate counter-measures.

Sense, enrich and classify: The ScanMaris workshop for assessment of vessel's abnormal behavior in the EEZ

Constant monitoring of the Exclusive Economic Zone cannot be performed only using high performance sensors. On the one hand, all available information on the observed area as juridical history of vessels or delineation of fishing zone is not necessarily measurable. On the other hand, even if the large amount of available information could be caught out they would be useless if none thorough sorting and analysis are carried on. So, we propose to sense vessel trail in Exclusive Economic Zone, to enrich the trail with the relevant non-measurable information and to use this material to classify vessel behaviors. The main issues of this approach are the selection of the relevant details which point to uncommon behaviors and the definition of the criteria which allow differentiating uncommon behavior and fraudulent one. Thus, the challenge is to deal with multidisciplinary information (i.e. bathymetry, radar tracks, insurance database, weather, etc.) and multi-domain criteria (i.e. juridical, behavioral, zonal, temporal, etc.). The architecture presented here aims to overcome both the issues.

A methodology to improve the assessment of vulnerability on the maritime supply chain of energy

The globalization of trade is due to the transportation possibilities and the standardization (containerization of freight). The dependency of the economy to the sea and to the merchant navy has increase this last decade. This process forms a worldwide maritime network between the different locations of production and consumption. This network, representing between 80 % and 90% of world traffic is a major economic concern, including freight distribution, raw materials or energy. Rodrigue demonstrates[1] the economic dependency of energy is increasing in the industrialized countries (North America, Europe, East Asia). The inter-regional trade of oil was 31 million bbl/day in 2002 and is expected to grow up to 57 bbl/day in 2030 [2]. Most of the international traffic use a maritime way, where may occur disruptions. For example, the Suez crisis (1956-1957) caused a closure of the canal, reducing the throughput capacity of transportation. This disruption cost a 2 millions of barrels lost per day. This article focuses on vulnerability of the energy supply, and proposes a methodology to formalize and assess the vulnerability of the network by taking into account the spatial structure of maritime territories.