Intermittent sources of energy represent a challenge for electrical networks, particularly regarding demand satisfaction at peak times. Energy management tools such as load shaving or storage systems can be used to mitigate abrupt variations in the network.The value of different mechanisms to move energy through time is determined by a multi-objective programming approach, that aims at minimizing operating costs as well as carbon emissions. By combining the output with sensitivity theory, the new technique builds a three-dimensional Pareto front, in which the usual information on indifference costs is accompanied by a third dimension that provides the decision maker with a quantitative measure to evaluate the relative impact of the two conflicting objectives. The interest of the methodology is assessed on three instances representing typical configurations in Brazil, Germany and France, respectively corresponding to a system that is hydro-dominated, thermo-dominated, and with a balanced mix of hydro and thermal power.