Dr Adam Sobey

Adam Sobey


Group Lead for Marine and Maritime, Data-Centric Engineering

Partner Institution


Adam Sobey is an Associate Professor in the Maritime Engineering group at the University of Southampton, Group Lead for Marine and Maritime in the Data-Centric Engineering Programme of The Alan Turing Institute and Non-Executive Director of Theyr Ltd, a MetOcean and Voyage Optimisation Software provider.

He completed his degree in Astronautics at the University of Southampton in 2006 and his PhD in the Maritime Engineering group in 2010 on the topic of automating the design of leisure boats. From 2009 he managed the Lloyd’s Register/Ministry of Defence Centre of Excellence in Marine Structures at Southampton, developing new techniques to model damaged ships. This research was incorporated into Lloyd’s Register’s design guidance and in 2015 he was awarded the Royal Institution of Naval Architect’s Jeom Paik Award for his investigations into using soft computing approaches to improve the reliability of ship structures. During this time he also worked on the ISO 12215-5 for small craft structural design and MAP-01-074, probabilistic design guidance for the UK submarines. In 2014 he became a Lloyd’s Register Educational Trust funded Research Fellow with 50% of his time seconded to the Institute of High Performance Computing in Singapore, determining how Evolutionary Computation and Monte Carlo Simulations could be used to improve the safety of composite risers. He returned full time to the University and in 2018 was made a lecturer in the Maritime Engineering group. He is an Executive Editor for the Data-Centric Engineering journal by Cambridge University Press.

Research interests

Adam is interested in unsupervised learning, machine learning and evolutionary computation approaches which have been applied in a range of industrial applications with a focus on reducing emissions. In particular:

  1. The genetic algorithm we have developed inspired by multi-level selection theory and our vision to extend it using epigenetics.
  2. An approach to machine learning that fits to the ground truth when we don’t know what the ground truth is, allowing stronger generalisation off dataset.
  3. The potential for multi-policy learning that allows improved transfer learning with reduced data and memory.