FleetMon in Research: Avoiding Collision at Sea With CADMUSS

in Research, Partnering by

Although the maritime industry has evolved to make sea transport efficient, fast, and reliable, the occurrence of collisions at sea is still high. Considering the ships registered in the EU alone, between 2014 and 2021, there were 22,532 collision occurrences. Globally over 876 ships were lost between 2011-2020, with 49 ships being lost in 2020 alone [1][2]. 

What are the significant factors contributing to such accidents? 

The most important factor is human error. According to the report published by the European Maritime Safety Agency, there were 966 investigations into the accidents relating to the EU vessels between 2014 to 2020. 89.5% of these incidents could be attributed to human action. Navigators today are loaded with far more responsibilities than those a decade ago. Frequent port calls, quick cargo operations, numerous audits and surveys, inclement weather, and high traffic are some factors that push the navigator to his limits. 

Supplementing Intuition With Algorithms


The CADMUSS project has been launched to help navigators with safe navigation [1]. CADMUSS stands for Collision Avoidance Domain-Method Used by Ships and aShore. It is a project being developed with the funding of ERA-NET Cofund MarTERA Call 2019.

The overarching goal of the CADMUSS project is to advise the crew on potential hazards with high accuracy and recommend the evasive maneuver to be employed to minimize the risk. To this end, the CADMUSS project begins with the definition of the ship’s domain as a foundation upon which more of the features are built. 

A vessel’s domain can be understood as an area around one’s vessel that the navigator wants to keep free from other objects and ships during daily navigation. To achieve a distance from such objects so that they do not violate the domain, the collision-avoiding maneuver must be planned and executed in time. Therefore, a region around one’s ship is called the preference-based ship domain, defined by the preferred distance from other ships in which navigators typically perform a successful collision avoidance maneuver. However, the shape and size of this region are not yet standardized in the scientific literature.

Given the vessel under consideration, this area would be of different shapes and sizes. One of the central goals of CADMUSS is to define this area and then use it as a benchmark criterion for assessing the ship’s safety. A collision warning system and a risk assessment system are then to be built on top of the system, which will give a sense of security to the ship in a particular set of circumstances. 

The Threefold Objectives of the CADMUSS Project

1. To develop a safety criterion for the safe passage of ships should a vessel come in the vicinity of another, on the high seas, or in coastal areas.

2. The second objective is implementing the criterion developed above to be used in Decision Support Tools (DSTs) which will help the crew make decisions even under high-pressure situations. The assistance is to be provided not only by a system on board but also by institutions such as Vessel Traffic Services (VTSs) and Fleet Operation Centers (FOCs) that can give more informed advice to the vessel on how to avoid the collision.

3. The third objective is to provide solid grounds for the transferability of this system to the field of risk assessment. So the safety criterion can be used to evaluate the level of risk in a particular area, for a specific vessel, or for a particular voyage.

Are You a Deck Officer? Your Experience Matters to Us!

Fleetmon, one of the consortium partners working with German and Polish marine companies and maritime universities, studies this topic and aims to develop a preference-based ship domain for professional navigators by taking a Ro-Pax vessel as a reference ship type. For this purpose, the knowledge of the experts is collected through an online survey.

The survey is based on an anonymous questionnaire seeking professionals’ preferences on collision avoidance under favorable conditions.

Your participation is greatly appreciated if you hold at least an Officer of the Watch license. It will contribute to a better understanding of maritime safety and evaluate evidence-based and knowledge-based safety measures.

You can find more information about the survey with direct access to the questionnaire here.

To understand the CADMUSS project in detail, please find the details below.

Vessel density between Fehmarn (Germany) and Lolland (Denmark). ©FleetMon

Details of the Features

The system’s foundation lies in accurately defining the ship’s domain. The defined area is composed of two zones, the outer zone and the inner zone. 

The outer zone is based on the comfort zone of the navigator. The distance at which the navigator is confident of avoiding collision with objects. The demarcation of this area is based on data obtained from surveys, interviews, and sessions with experts on a bridge simulator. The Polish Project partners are responsible for this end of the project.

The inner zone is the boundary beyond which, if an object enters, there is no way of avoiding collision with it. The inner zone is determined by models and simulations taking into account parameters such as the hydrodynamics of the ship, the geometry of the encounter, and the weather and ocean conditions. Since some evasive maneuvers require less space to execute, the inner boundary is not so sharply defined. It can be envisioned as hazy and gradual. 

The 2D and 3D Components

A top-down, bird’s eye view model is excellent for evaluating the ship’s domain and the risk from oncoming vessels. But in addition to this, there is an added component of a third dimension to the model being developed. The third dimension considers the vessel’s draft, the ocean’s depth profile, and the vessel’s rolling and pitching while evaluating risk. This will also make the system especially useful in areas where depth can be problematic, such as navigating channels. If the model’s detracting depth factor is absent, the risk will consistently be underestimated.

The final goal is a system that can warn the vessel of oncoming threats. By combining the knowledge of experts with simulations and models, we hope to achieve a system that can send an unambiguous message ensuring the vessel’s safety. 

Risk Analysis and Vessel Traffic Simulation

More than 150 billion AIS position messages from FleetMon are being used for this project. Since AIS can only give a 15-minute resolution to what is going on, parametric modeling is used to interpolate the data to bring down the intervals to 1 minute. Objects representing their maritime counterparts are modeled in simulations using the AIS data. The result is analogous to a war room map, with pieces representing the vessel under consideration and all other maritime objects in its vicinity. Risk assessments and analyses can be made on past data by creating a dynamic model of these objects. 

The Close Range Risk Assessment Module (CRRAM) is one of the project’s ends goals. It is a fully web-based demonstrator module that will make the examination of all AIS-A class ships for the last year possible in 15-minute intervals. This demonstrator will combine the results of both the Polish and the German partners.

This makes the analytical evaluation of the past behavior of a vessel possible, which would be particularly lucrative for the insurance industry. The safety maneuvers performed in the past can be rated, and insight gained into how well the crew handled themselves and how well they were trained. In essence, gaining another metric to accurately assess the risk involved in a voyage.

Research & Development at FleetMon

At FleetMon, we have not only created the world’s first vessel database, but we are also very passionate about research and development. We motivate scientists, institutes, and universities to use our powerful and customized API solutions and provide access to our historical vessel position database.

List of References


[1] https://emsa.europa.eu/accident-investigation-publications/annual-overview.html

[2] https://www.iii.org/fact-statistic/facts-statistics-marine-accidents