Back to top

Guide to Maritime Informatics

About the book

Alexander Artikis and Dimitris Zissis

In the last 25 years, information systems have had a disruptive effect on society and business. Up until recently though, the majority of passengers and goods were transported by sea in many ways similar to the way they were at the turn of the previous century. Gradually, advanced information technologies are being introduced, in an attempt to make shipping safer, greener, more efficient and transparent. The emerging field of Maritime Informatics studies the application of Information Technology and Information Systems to maritime transportation. Maritime informatics can be considered as both a field of study and domain of application. As an application domain, it is the outlet of innovations originating from Data Science and Artificial Intelligence, while as a field of study, it sits on the fence between Computer Science and Marine Engineering. Its complexity lies within this duality, as it is faced with disciplinary barriers while demanding a systemic transdisciplinary approach.

At present, there is a growing body of knowledge developing, which remains undocumented in a single source or textbook designed to ease students and practitioners into this new field. The objective of this book is to collect the material required for an undergraduate or postgraduate student to develop the core knowledge of this domain, in an analytical approach through real-world examples and case studies. The aim is to present our audience with an overview of the main technological innovations which are having a disruptive effect on the maritime industry, describe their principal ideas, methods of operation and applications, and discuss future developments. The book is designed in such a way as to first introduce required knowledge, algorithmic approaches and technical details, before presenting real-world applications.

Read More…


Part I: Maritime Data

1. Maritime Reporting Systems

Konstantina Bereta, Konstantinos Chatzikokolakis and Dimitris Zissis

In recent years, numerous maritime systems track vessels while travelling across the oceans. Ship reporting systems are used to provide, gather or exchange information through radio reports. This information is used to provide data for multiple purposes including search and rescue, vessel traffic services, prevention of marine pollution and many more. In reality though researchers and scientists are finding out that these data sets provide a new set of possibilities for improving our understanding of what is happening or might be happening at sea. This chapter provides an introduction to the main vessel reporting systems available today, while discussing some of their shortcomings and strong points. In this context, several applications and potential uses are described.

2. Navigating the Ocean of Publicly Available Maritime Data

Theodoros Tzouramanis

In the past, the data-capturing tasks involved in researching and developing maritime projects were often time-consuming and costly. It was not unusual that different project ventures, unaware that other projects were similarly engaged, were unknowingly simultaneously gathering data of a similar nature or even the very same data. As part of an initiative to prevent such a waste of resources, a number of international organizations decided to make high-quality maritime data, which had been painstakingly collected over many years, freely accessible to the public. The aim of this chapter is to draw up a compilation of maritime datasets that hundreds of core and strategic data sources across the world make freely available online at the time of writing. These datasets provide, in their large majority, and each one in its own field, a complete and detailed view of the entire aquatic environment of the earth. Ultimately, this chapter intends to contribute a useful working tool to all who operate as experts or as users of maritime geographic information systems. If this chapter manages in a helpful way to set readers on the path to successfully finding and selecting the data that fit their needs and to navigating effortlessly the sea of online information, then it will have achieved its goal.

Part II: Off-line Maritime Data Processing

3. Maritime Data Processing in Relational Databases

Laurent Etienne, Cyril Ray, Elena Camossi and Clément Iphar

Maritime data processing research has long used spatio-temporal relational databases. This model suits well the requirements of off-line applications dealing with average-size and known in advance geographic data that can be represented in tabular form. This chapter explores off-line maritime data processing in such relational databases and provides a step-by-step guide to build a maritime database for investigating maritime traffic and vessel behaviour. Along the chapter, examples and exercises are proposed to build a maritime database using the data available in the open, heterogeneous, integrated dataset for maritime intelligence, surveillance, and reconnaissance that is described in the work of Cyril Ray et al. (2019). The dataset exemplifies the variety of data that are nowadays available for monitoring the activities at sea, mainly the Automatic Identification System (AIS), which is openly broadcast and provides worldwide information on the maritime traffic. All the examples and the exercises refer to the syntax of the widespread relational database management system PostgreSQL and its spatial extension PostGIS, which are an established and standard-based combination for spatial data representation and querying. Along the chapter, the reader is guided to experience with the spatio-temporal features offered by the database management system, including spatial and temporal data types, indexes, queries and functions, to incrementally investigate vessel behaviours and the resulting maritime traffic.

4. Maritime Data Analytics

Panagiotis Tampakis, Stylianos Sideridis, Panagiotis Nikitopoulos, Nikos Pelekis and Yannis Theodoridis

The goal of mobility data analytics is to extract valuable knowledge out of a plethora of data sources that produce immense volumes of data. Focusing on the maritime domain, this relates to several challenging use-case scenarios, such as discovering valuable behavioural patterns of moving objects, identifying different types of activities in a region of interest, estimating fishing pressure or environmental fingerprint, etc. In this chapter, we focus on the exploration, preparation of data and application of several offline maritime data analytics techniques. Initially, we present several methods that assist an analyst to explore and gain insight of the data under analysis. Subsequently, we study several preprocessing techniques that aim to clean, transform, compress and partition long GPS traces into meaningful portions of movement. Finally, we overview some representative maritime knowledge discovery techniques, such as trajectory clustering, group behaviour identification, hot-spot analysis, frequent route or network discovery and data-driven predictive analytics methods.

5. Visual Analytics of Vessel Movement

Natalia Andrienko and Gennady Andrienko

Visual analytics techniques support the process of data analysis, reasoning, and knowledge building performed by a human analyst. The techniques combine interactive, human-controllable visual displays with interactive operations for data querying and filtering, data transformations, calculation of derived data, and application of computational techniques for analysis and modelling. We demonstrate the use of visual analytics techniques and procedures for analyzing Automatic Identification System (AIS) data. We begin with showing how visual analytics approaches can help in exploring properties of the data, detecting problems, and finding ways to clean and improve the data. Then we describe two analysis scenarios focusing on the events of vessel stopping and on the vessel traffic through the strait between the bay of Brest, France, and the outer sea. Thereby we show how different techniques are applied and combined.

The analysis in this chapter has been performed using V-Analytics.

Part III: On-line Maritime Data Processing

6. Online Mobility Tracking against Evolving Maritime Trajectories

Kostas Patroumpas

We examine techniques concerning mobility tracking over trajectories of vessels monitored over a large maritime area. We focus particularly on maintaining summarized representations of such trajectories in online fashion based on surveillance data streams of positions relayed from a fleet of numerous vessels using the Automatic Identification System (AIS). First, we review generic, state-of-the-art simplification algorithms that can offer concise summaries of each trajectory as it evolves. Instead of retaining every incoming position, such methods drop any predictable positions along trajectory segments of ''normal'' motion characteristics with minimal loss in accuracy. We then discuss online filters that can reduce much of the noise inherent in the reported vessel positions. Furthermore, we present a method for deriving trajectory synopses designed specifically for the maritime domain. With suitable parametrization, this technique incrementally annotates streaming positions that convey salient trajectory events (stop, change in speed or heading, slow motion, etc.) detected when the motion pattern of a given vessel changes significantly. Finally, we discuss a qualitative comparison of maritime-specific synopses along with trajectory approximations obtained from generic simplification algorithms and highlight their pros and cons in terms of approximation error and compression ratio.

7. Link Discovery for Maritime Monitoring

Georgios M. Santipantakis, Christos Doulkeridis and George A. Vouros

Link discovery in the maritime domain is the process of identifying relations - usually of spatial or spatio-temporal nature - between entities that originate from different data sources. Essentially, link discovery is a step towards data integration, which enables interlinking data from disparate sources. As a typical example, vessel trajectories need to be enriched with various types of information: weather conditions, events, contextual data. In turn, this provides enriched data descriptions to data analysis operations, which may lead to the identification of hidden or complex patterns, which would otherwise not be discovered, as they rely on data originating from disparate data sources. This chapter presents the fundamental concepts of link discovery relevant to the maritime domain, focusing on spatial and spatio-temporal data. Due to the processing-intensive nature of the link discovery task over voluminous data, several techniques for efficient processing are presented together with examples on real-world data from the maritime domain.

8. Composite Maritime Event Recognition

Manolis Pitsikalis and Alexander Artikis

Composite maritime event recognition systems support maritime situational awareness as they allow for the real-time detection of dangerous, suspicious and illegal vessel activities. To illustrate the use of such systems, we motivate and present a series of composite maritime event patterns in a formal language. For effective recognition, the presented maritime patterns have been developed in close collaboration with domain experts, and evaluated with the use of Automatic Identification System (AIS) datasets.

Part IV: Applications

9. Uncertainty Handling for Maritime Route Deviation

Anne-Laure Jousselme, Clement Iphar and Giuliana Pallotta

Detecting and classifying anomalies generally contributes to Maritime Situation Awareness and highly benefits from the combination of multiple sources, as correlating their output allows detecting inconsistencies in vessels' behaviour. In particular, detecting the route followed by vessels, and identifying off-route vessels is a challenging problem which requires a first characterisation of maritime routes (the normalcy) and secondly the association of a vessel track to an existing route. In this context, adequate uncertainty representation and processing is crucial for this higher-level task where the operator analyses information in conjunction with background knowledge. The maritime anomaly detection solution is framed into a mathematical uncertainty theory, encoding thus semantics in both uncertainty representation and reasoning. The choice of the theory together with the associated calculus defines then directly the output of the algorithm and the result presented to the user. In this chapter, we dissect six classical Uncertainty Representation and Reasoning Techniques (URRTs), each solving the problem of track to route association. In their basic form, the URRTs are framed into the three uncertainty theories of probabilities, belief functions and fuzzy sets, which capture different feature of information deficiencies, of uncertainty, imprecision, graduality. The different URRTs are qualitatively evaluated according to their expressiveness along these deficiencies and quantitatively evaluated according their trueness, precision and certainty when processing real AIS data with route labels.

10. Maritime Network Analysis: Connectivity and Spatial Distribution

César Ducruet, Justin Berli, Dimitris Zissis and Giannis Spiliopoulos

In this chapter we apply conventional graph-theory and complex network methods to a sample of port and inter-port shipping flows at and amongst the top 50 European ports in 2017. Such methods help to detect the main topological and geographic structures of this network in order to answer three main questions. First, why are certain port nodes better connected than others? Such a level of hierarchy is best approached by testing the scale-free and rich-club dimension of the network. For this we measure node connectivity in various ways, from local to global indices, all confirming inequality in traffic distribution. Second, what is the influence of cargo specialisation or diversity on the network structure? This relates to the concepts of multiplexity and assortativity, i.e. the ability of nodes to diversify their activity or to specialise. Two principal layers are analysed and compared, namely cargo and bulk, showing that larger ports and links are more diversified. Lastly, what are the substructures or geographic patterns underlying the distribution of maritime flows? To answer this, we examine the influence of physical distance on connectivity and on the emergence of subnetworks.

11. Shipping Economics and Analytics

Roar Adland

In theory, spatial shipping data can be used to track variables that influence the supply or demand for maritime transportation. In real applications, the limitations of using AIS data alone for economic analysis soon become apparent, though it remains a very useful tool. In this chapter we review how AIS data can be used for data-driven market analysis in shipping. In particular, we outline how AIS data can be used to track commodity flows and key variables for fleet efficiency, supply and demand. We also emphasize the limitations of ship tracking data for commercial analysis. The chapter deals solely with the economics and analytics of bulk shipping (i.e. bulk carriers or tankers) as it represents the purest case of a perfectly competitive market where observed changes in supply and demand should have some bearing on the economic outcomes.