The University of Essex and The Corporation of Trinity House of Deptford Strond KTP 24_25 R5

A lighthouse keeper's maintenance logs are about to be fed into a machine learning model. The University of Essex and Trinity House, the charity that maintains England's navigation aids, are collaborating to extract new insights from decades of data on buoys, beacons, and lights. These aids to navigation require regular servicing, and their performance records contain patterns that human analysis has not yet revealed. The machine learning techniques will identify which components fail most often, how weather affects equipment lifespan, and where maintenance schedules could be more efficient. If successful, the project could help Trinity House predict failures before they happen, reduce the time vessels spend waiting for repairs, and lower the cost of keeping shipping lanes safe. For the general public, the impact is invisible but real: fewer disruptions to the supply chains that move goods by sea, and more reliable navigation for ferries, fishing boats, and leisure craft. This is applied data science for a specific operational challenge, not fundamental research.