Optimal operation of reservoir systems is the most important issue in water resources management. It presents a large variety of multi-objective problems that require powerful optimization tools in order to fully characterize the existing trade-offs. Many optimization methods have been applied based on mathematical programming and evolutionary computation (especially heuristic methods) with various degrees of success more recently. This paper presents an implementation and comparison of multi-objective particle swarm optimization (MOPSO) and non-dominated sorting genetic algorithm II (NSGA-II) for the optimal operation of two reservoirs constructed on Ozan River catchment in order to maximize income from power generation and flood control capacity using MATLAB software. The alternative solutions were based on Pareto dominance. The results demonstrated superior capacity of the NSGA-II to optimize the operation of the reservoir system, and it provides better coverage of the true Pareto front than MOPSO.