Maritime Informatics

Introduction

Lind states that “maritime informatics is the application of information systems to [increase] the efficiency, safety, and ecological sustainability of the world’s shipping industry” (Lind et al, 2020, p. vii). He further states that “maritime informatics is an applied science. Developed by data scientists to meet the need of practice and applied by [the shipping industry] and data scientists cooperatively.”

The emergent discipline of maritime informatics was caused by the proliferation of onboard transponders called automated identification systems (AIS) in the late 20th century. Automated identification systems transmit and receive ship positions from other ships using radio waves and are used for the purpose of navigational safety. However, shore stations and satellites are also able to receive AIS information and store ship particulars in a database.

Once the particulars are committed to a database, the data can then be combined with spatial objects (IHO, 2000) such as anchorage polygons (Andrienko & Andrienko, 2021), berth coordinates, and demarcation lines such as emissions zones. Proper modeling of ship positions combined with static and dynamic objects can produce outcomes of operational and commercial interest.

Method

Maritime informatics can be broken down into the following elements:

• geographic information systems (GIS)
• automated identification systems (AIS)
• relational databases
• static data
• geospatial data
• geocapture
• data analytics and modeling

Geographic Information Systems (GIS)

Visual analytics is defined as the “science of analytical reasoning facilitated by interactive visual interfaces” (Thomas & Cook, 2005). Due to the fact that humans reason using information available to them, visual representations of information are best suited for this task (Andrienko & Andrienko, 2021).

Graphical information systems (GIS) allow maritime-related objects to be represented on a map rendered on a computer screen. The first use of GIS in 1963 was to manage the allocation and overland movement of bulk resources by the Canadian Government (ESRI, 2023).

Natalia Andreinko and Gennady Ardrienko published Visual Analytics of Vessel Movement that succinctly describes the value of a user interface in relation to “visual analytics techniques and procedures for analyzing automatic identification system (AIS) data” (Artikis & Zissis, 2021). Environmental Systems Research Institute (ESRI) is the largest purveyor of GIS frameworks with a 43% market share (ESRI, 2021).

Automated Identification Systems (AIS)

Automated identification systems transmit and receive ship positions from other ships using radio waves for the purpose of navigational safety. Additional data is transmitted as well. Examples of additional data is the vessel’s name, where the vessel is proceeding to, and when it will get there.

Due to the sensitive nature of automated identification systems, operational use is managed by the International Maritime Organization (IMO) which is a subsidiary of the United Nations. The latest IMO guidelines, Guidelines for the Onboard Operational Use of Shipborne Automated Identification Systems (AIS) was published in 2015 and is a resource for shipping and data professionals to understand how the AIS is used operationally aboard merchant vessels.

Use of AIS data in the field of maritime informatics can be aligned with the rise of cloud computing and an increase in network speeds. Today, shore-based or satellite-based stations are able to receive, collect, and commit the data to remote databases for future analysis.

The use of AIS in the field of maritime informatics is a new and fragmented subject. Currently many data professionals and subject-matter experts are exploring different AIS data models with varying degrees of efficacy. In 2020 and 2021, Springer published Maritime Informatics (1st ed.) and Guide to Maritime Informatics (1st ed.) respectively. In these first-edition volumes various applications of AIS data are explored. The importance of these volumes — the first of their kind — can not be overemphasized. The volumes describe efforts in this emergent discipline that underpin this shift within the midstream supply chain within the context of Third Wave Economics (Salinas, 2021).

Relational Databases

For the uninitiated, a conventional relational database for use with maritime data, namely AIS data, could be visualized as a data table. The leftmost column in the table contains an identifier that is used to access the row which contains different types of descriptive data siloed in successive columns. The different types of AIS data that may be committed to a relational database can be explicitly found in the IMO’s New and Amended Performance Standards (Resolution MSC.74(69)). Automated information system data is usually stored off site and managed by a third party, such as Amazon Web Services (AWS) or Digital Ocean. Automated information systems produce a high volume of data and thus can be considered an exercise in big data (Oracle, 2023).

Information on a relational database can be selected, updated, deleted, and altered. The language used to describe the ways the data can be worked with is called structured query language (SQL). Modern data management efforts are conducted using PostgreSQL (Etienne et al., 2021). NoSQL databases such as MongoDB are mentioned in literature however as of this writing — though there is ample literature — no applied use case was found.

Once AIS data is collected, usually based on some sort of filtering criteria, disparate databases containing static and dynamic data can be merged with the ship position database to fulfill various modeling requirements. Outcomes may be density maps, trade flows, or support for search and rescue.

A primer for working with AIS data in a relational database can be found in Maritime Data Processing in Relational Databases (2021) by Laurent Etienne, Cyril Ray Ekuba Camossi, and Clement Iphar.

Static Object

A static object exhibits a single, immutable, coordinate. The International Hydrographic Organization (IHO) publishes a document called IHO Transfer Standard for Digital Hydrographic Data (2000). Within this text an explicit definition of geospatial data as it relates to maritime objects is rendered. Examples of static data that are important in a maritime context are listed.

Dynamic Object

A dynamic object moves and may have a vector and mutable coordinates.

Geospatial Data

Geospatial data is a data object that contains position information and can be found in IHO Transfer Standard for Digital Hydrographic Data (2000). Within this text an explicit definition of geospatial data as it relates to maritime objects is rendered. Examples of geospatial data that are important in a maritime context are listed. It should be noted static and dynamic objects are children of geospatial data.

Data Analytics & Modeling

“[Hydrographic models are] designed to permit the transfer of data describing the real world. The real word is far too complex for a complete description to be practical, therefore a simplified, highly-specific, view of the real world must be used. This is achieved by modeling [reality].

[Schemas are] specifically concerned with those entities in the real world that are of relevance to hydrography. [Hydrographic regimes are] considered to be geo-spatial. As a result, [models define] real world entities as a combination of descriptive and spatial characteristics. Within the model these sets of characteristics are defined in terms of [static] objects and [geo]spatial objects. An object is defined as an identifiable set of information. An object may have attributes and may be related to other objects” (International Hydrographic Organization, 2000, p.2.1).

Data modeling and outcomes in the context of maritime informatics involves identifying a problem to be solved and devising a model to address that problem using dynamic and static objects. An simple example would be identification of whether or not a vessel possesses the necessary protection and indemnity (P&I) insurance to enter California and using AIS data to understand whether the vessel is within California’s economic exclusion zone (EEZ) using AIS position information. There will only be two outcomes, either the vessel is in regulatory compliance or it is not.

Conclusion

One of the current challenges of maritime informatics is the requirement of a multidisciplinary approach (Lind et al, 2020, p. vii). Data scientists and shipping professionals must work in concert to produce outcomes. Thus, a proper project management approach is critical when analyzing and modeling data related to marine transportation.