Anyone thinking of oil tankers moving across the world’s oceans would probably conclude that these massive vessels follow strictly predetermined routes and schedules to minimize risks to such a valuable cargo. As Kevin Teo has discovered, however, there is a surprisingly high degree of randomness in these ship movements and few studies that investigate the factors driving the ships’ behavior.
A PhD student at NetSI London, Kevin started studying the tanker shipping network a year ago, under the mentorship of Prof. István Kiss, soon after joining the institute’s London hub as its first doctoral student. With a physics background, Kevin remembers listening to online talks by Prof. Sean Carroll about the idea of complexity and emergence. “Learning about ideas of complexity and network theory felt extremely refreshing, compared to my other physics modules.” recalls Kevin, who ended up doing his master’s project on networks and directed acyclic graphs. When NetSI launched its doctoral program in network science, he decided to embrace the opportunity at NetSI’s London hub, where Prof. Kiss and his team had just obtained a highly detailed and timestamped dataset of oil tanker shipping movements spanning 4 years.
The global shipping network accounts for over 80% of the world’s trade by volume, and while the container shipping industry responsible for transporting packaged goods has been widely examined, the same cannot be said about the shipping of crude oil, despite it dominating 30% of the market. The partnership with a maritime analytics company, however, made it possible to analyze detailed data on oil tanker movements, allowing Kevin and other contributors to the study, including Prof. Kiss and Prof. Mauricio Santillana, to gain a more comprehensive understanding of the shipping network dynamics.
Interestingly, ship owners of oil tankers have a considerable flexibility in deciding, even at relatively short notice, where to direct their ships for loading and selling cargo, resulting in a high degree of unpredictability in these ships' movements. A question emerged: would it be possible to predict these movements and transform the tanker shipping industry, known for being a heavy polluter, into a more efficient and environmentally sustainable sector, all while preserving economic profitability?
The study yielded surprising results which will be detailed in an upcoming publication and provide insights into shipping patterns at both the individual ship level and the regional fleet level, with particular focus on the random movement of certain vessels. It was, indeed, this apparently chaotic behavior that was the most intriguing to study, according to Kevin who was drawn to network science “by its allure of unification and the idea that many phenomenologically different systems can share the same underlying mathematical principles”. Being able to model a tanker shipping network that includes these unpredictable actors would greatly benefit the shipping industry and its supporting infrastructure – says Kevin, as it would provide actionable insights for both ship owners and maritime authorities to mitigate ships' reliance on well-known chokepoints in maritime routes, enhance the system's resilience to sudden shocks, and improve regional planning strategies. Increased efficiency of the shipping network would ultimately reduce the adverse effects of the industry on the environment, leading to more sustainable navigation practices.
Working on this research project was valuable for Kevin from a practical perspective as well, particularly in terms of mechanisms for funding research in the UK, where the government allocates funds to research institutions based, among other factors, on their research output with real-world impact on society and economy. Because the study on global shipping networks of oil tankers was conducted in collaboration with an industry partner, and the research findings can be translated in an effective improvement of the shipping industry with implications for wider transportation networks, it aligns with UK’s Research Excellence Framework (REF29) and helps Northeastern University in London to compete for UK government funding.
