In this study, an optimization model, taken into consideration to the irrigation water as parametric variable, is developed for a water resources system with multi-objective and multiple reservoirs. This model has two stages. In the first stage of the model, the maximized firm power values are maximized by using monthly mean inflow data of the drought period. In the second stage, the maximization of the total energy of the system is realized by locating firm power values as parametric constant and by using data of the monthly average inflow in the model. A dynamic programming with successive approximations is used as optimization technique. The model is applied to a water resources system with multiple reservoirs and multi-objective in the Munzur River Basin, Turkey. Results obtained are evaluated to the energy production, irrigation water and total benefits. As a result, the irrigation water has obtained as 60x10(6) m(3) where the total benefit of the system is maximized.