Meet Tom Melamed, Head of Data Science

As the maritime industry navigates increasing fuel costs, regulatory changes, and sustainability goals, data science has emerged as a powerful tool for improving operational efficiency. With a background in physics, computer science, and industrial R&D, Tom brings a research-driven approach to maritime analytics. His work focuses on helping ship owners and charterers optimise fleet performance, reduce costs, and drive sustainability - all through the power of data. 

What ignited your interest in data science, and how did your career path lead you to 90POE?

My journey into data science began with a deep-rooted love for research and experimentation, this started during my time studying physics and computer science at Bristol University. After graduation, I spent several years in industrial research and development, later co-founding and co-managing a successful mobile app development agency. However, I found myself wanting to return to my scientific roots—driven by a passion for solving challenging problems and exploring empirical methods. 

This led me back to academia, where I pursued a master's degree in Data Science at the University of Bath. Following graduation, I led a data science team focused on TV recommendation systems, including projects for one of the UK's largest broadband TV providers. Although rewarding, I wanted to apply my skills to an area with greater practical and social impact. Joining 90POE as Head of Data Science fulfilled that ambition perfectly, allowing me to help our customers significantly reduce costs while simultaneously making substantial contributions to environmental sustainability by reducing atmospheric pollution and CO2 emissions 

How does data play a role in maritime technology, and what impact does it have on the industry? 

Data is hugely important in the maritime sector, especially as the industry grapples with rising fuel costs and increased pressure to decarbonise. Leveraging data effectively provides ship owners and charterers with clear, measurable improvements in operational performance and efficiency. The maritime industry is currently experiencing a pivotal transition: moving from early attempts at harnessing data to a more mature, informed application across the entire sector. 

We are also beginning to see an evolution where the insights gained from detailed, high-frequency sensor data are now being successfully adapted to vessels reliant on traditional noon reports. This shift represents a broader trend—from an initial scepticism about data's reliability to a growing appreciation of its transformative impact among forward-thinking maritime companies. Companies that embrace this shift and partner with capable analytics providers are already experiencing significant and measurable benefits in their fleet performance. 

What innovations in data science do you think will have the biggest impact in maritime in 2025 and beyond? 

Looking ahead, the most impactful innovations will centre around advanced data-driven insights into hull conditions and precise route optimization tailored to the unique characteristics of individual vessels. These advancements promise tangible, repeatable savings in fuel consumption and pollution levels by enabling timely and accurate hull cleaning and highly optimised routing decisions. 

Beyond 2025, I anticipate these innovations will lead to significantly improved operational predictability, making "just-in-time" arrivals increasingly feasible. While full industry-wide adoption of such precise coordination may take longer, primarily due to the high levels of trust, coordination and collaboration required among operators, charterers, and ports, the science is being developed now. The future of maritime data science is extremely promising, and these incremental innovations will cumulatively deliver significant efficiencies. 

If you could give one piece of advice to companies looking to leverage data more effectively, what would it be? 

I am afraid I can not give you just one, I must give you two. First, companies must recognise the immense value of their own data and invest in capturing it accurately and consistently. This doesn’t necessarily mean costly sensor installations on every vessel, but it does require training captains and engineers to appreciate the importance of detailed, timely, and accurate record-keeping. 

Secondly, companies should set high expectations for their analytics providers. With just a few months of quality historical data, skilled providers can derive valuable insights into ship performance. Once these insights are verified and trusted, companies should actively use them to enhance their operations. Done correctly, the returns on this investment can be swift, substantial, and deeply impactful.