Supervisors

Suitable for

Abstract

The project focuses on the application of data science in maritime operations, specifically in enhancing the detection of ship-to-ship (STS) transfer activities. STS operations entail the exchange of cargo between two vessels at sea, serving both legitimate purposes – such as draught optimisation to prevent grounding in shallow waters or cargo ownership transfer – as well as illegal purposes, such as circumventing international sanctions. Detection hinges on analysing GPS-based patterns in Automatic Identification System (AIS) data emitted by ships.

Vortexa is recognised as an industry leader in real-time global STS operation detection. However, our existing detection model has become outdated. Developing a new model using the latest data and techniques is crucial for enhancing detection accuracy and range, providing substantial value to our clients.

Research challenges:

● Modern neural architectures: explore the integration of state-of-the-art neural networks. This new model must be scalable to detect STS operations worldwide in real-time.

● Enhanced geographic coverage: the new model’s aim is to expand its surveillance capabilities beyond predefined regions of interest. By employing advanced data analytics and machine learning techniques, we seek to automatically identify and monitor potential STS activity across the globe, eliminating dependency on manual inputs and uncovering new areas of interest.

● Increased accuracy and reduced false positives: refine detection algorithms to minimise false positives, particularly in busy maritime areas with frequent anchorages. Achieving superior accuracy will enhance the reliability of alerts, boosting client confidence and data usability.

● Detection of "dark" STS events: one of the most ambitious advances is to develop methodologies to detect STS transfers involving "dark" ships, which disable their AIS transponders to evade sanctions. This requires integrating alternative data sources and advanced pattern recognition technologies, achieving breakthroughs in illicit activity tracking and compliance monitoring.

Achieving these objectives promises substantial advancement in our detection capabilities, reinforcing our industry leadership while providing clients with unmatched insights and reliability.

Expected outcomes:

● Development of a new STS detection model utilising modern neural architecture.

● Enhanced detection accuracy with reduced false positives and broader geographic monitoring.

● Breakthrough methodologies for identifying “dark” STS activities, setting new standards in maritime monitoring. Skills and Experience Required:

● Driven by working in an intellectually engaging environment with the top minds in the industry, where constructive and friendly challenges and debates are encouraged, not avoided

● Strong foundation in software engineering and machine learning, with coursework in advanced machine learning or data science preferred.

● Proficiency in Python, especially in machine learning libraries and geospatial data processing.

● Interest in applying machine learning to real-world maritime challenges and developing cutting-edge detection methods.