Supervisors

Suitable for

Abstract

The "Destination Sequence Model" project seeks to advance our capacity to forecast global oil and gas tanker destinations. At Vortexa, we've already developed an industry-leading model that provides accurate, real-time predictions of vessel destinations. This information is valuable to our clients, helping them optimise logistics, understand supply and demand trends, take advantage of market opportunities, and make informed decisions about their investments.

This project aims to push the boundaries of our predictive capabilities by leveraging advanced machine learning architectures, specifically Transformers, for vessel movement analysis. By treating historical vessel movements as sequences, we aim to refine prediction precision and obtain deeper insights into behavioural patterns in tanker routes. Despite the success of our current models, incorporation of elements like geohashing remains unexplored, which holds potential for addressing significant prediction challenges.

Research challenges:

● Dynamic data landscape: the energy market is highly volatile, influenced by factors such as supply and demand, geopolitical events, regulatory changes, and weather. Accurately predicting vessel destinations requires robust modelling to adapt to these fluctuations.

● Data incompleteness and noise: shipping schedules and routes often remain confidential, leaving gaps in our data. Furthermore, the Automatic Identification System (AIS), a critical data source, is susceptible to noise, data manipulation, and coverage gaps, thus complicating prediction efforts.

● Erratic vessel behaviour: tankers can alter their destinations mid-journey in response to market conditions, price fluctuations, or new orders, introducing additional complexity into forecasting models.

● Complex logistics: the process of transporting energy often involves multiple stops, transfers, and storage points, necessitating sophisticated modelling techniques surpassing traditional methods.

Expected outcomes:

● Development and evaluation of an enhanced prediction model integrating geohashing.

● Identification of key factors influencing tanker routes through advanced sequence modelling.

● Generation of insights into the market factors driving vessel diversions.

 

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 online machine-learning algorithms and data streams.

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