Optimization of the arrangement of fuel assemblies and burnable poisons when reloading pressurized water reactors has, in the past, been performed with many di erent algorithms in an attempt to make reactors more economic and fuel effi cient. The use of the tabu search algorithm in tackling reload core design problems is investigated further here after limited, but promising, previous investigations. The performance of the tabu search implementation developed was compared with established genetic algorithm and simulated annealing optimization routines. Tabu search outperformed these existing programs for a number of dffi erent objective functions on two di erent representative core geometries.