Experiments are conducted to determine the time to reach a set-point temperature for aluminum finned heat sinks filled with the phase change material n-eicosane. Results thus obtained are integrated with a feed-forward back-propagation artificial neural network to predict operating times. The artificial neural network prediction is then used to determine the optimum configuration that maximizes thermal performance. Four different plate-fin heat sink geometries filled with phase change materials giving rise to different volume fractions of the aluminum were experimentally investigated.