Abstarct: Optimization is an engineering tool to maximize the performance with the lowest cost to benefit ratio thus, many engineers and scientists seek new and reliable optimization techniques. Drinking water as a basic and vital need of all societies is distributed through pipe networks which are expensive structures. As a consequence of development, improvement of the existing networks is a major task in most expanding cities. Optimization of construction and improvement of water distribution networks may reduce the cost of construction projects without deficiency in performance. The present study dealt with the optimization of water distribution network using genetic algorithm (GA). GA techniques successfully employed to optimize some standard examples as well as some new networks. In this research, a code in Visual Basic (ver. 6) was developed to apply the GA. EPANET 2.0 was employed by the code to find the water velocity in pipes and pressure in nodes of the proposed network. Two fitness functions were used to comply with the pressure (Case1) and pressure- velocity (Case 2) constricts. Various GA operators were employed to study their performance in the process. The results shows that the GA is capable of lowering the cost of construction to the level found by other optimization techniques. As an improvement, the proposed method needs a much smaller number of generations to reach the optimum answer. Two fitness functions (dynamic and static functions), two operators (Roulette Wheel and Tournament), and three types of crossovers (One point, Two point, Uniform) for two cases (total of 24 combinations) were studied for each example with 100 runs. The number of minimum cost obtained by each combination was recorded. Some combinations returned as many as 94 minimum answers. As a conclusion, GA can be used to design the water distribution networks less expensively.