Most of the current domestic hot water tanks store heat in sensible form. Latent heat storage systems have become popular recently due to their capacity of storing energy in high density with slight temperature variation. They use phase change materials (PCMs) as the storage medium which have a high heat of fusion and are able to store and release large amount of energy. This paper reports the outcomes of implementing PCM in a standard domestic hot water tank, on energy consumption and the discharge period and suggests a method for optimization of the tanks regarding their discharge time. Geometrical aspects of PCM containers and the amount of PCM inside the tank were design parameters of this study. TRNSYS was used to simulate the behavior of the tank with PCM. Artificial neural network was used to find a correlation between design parameters of the tank and the discharge time. Genetic Algorithm was then used in order to find the optimum amount of PCM needed for a desired peak shift. Finally, a comparison between an optimized tank with PCM and a tank without PCM was performed and it was concluded that an optimized tank is able to provide hot water while consuming less energy compared to the regular one (without PCM). Furthermore, by applying a certain amount of PCM, electrical energy consumption could be shifted completely to off peak periods characterized by less power demand. (C) 2015 Elsevier B.V. All rights reserved.