Doctoral Thesis: 'Multiobjective Optimization of Metallic Frames using Evolutionary Algorithms' by David Greiner Supervisors: Gabriel Winter, Jose Maria Emperador Abstract In this work, it has been developed an efficient method for the multiobjective optimization of metallic frame structures, minimizing simultaneously the constrained mass and the number of different cross-section types. Both objectives are selected due to economic and constructive requirements, respectively. The resolution methodology has been in the context of evolutionary multiobjective algorithms (stochastic methods inspired in the natural evolution mechanisms). They allow to optimize considering several conflicting criteria and to solve problems that are previously forbidden to the classical methods. They are suitable due to the requirements of the problem: global, discrete and multiobjective optimization. After a review of the state of the art of multiobjective optimization methods, specially centred in evolutionary ones and its applications in structural evolutionary optimization, it has been performed an analysis using the most efficient algorithms (thirteen) over several structural test cases considering also some optimization aspects (codification, elitism, mutation rate and population strategy). This study will offer the main directions of an efficient optimisation and it includes the proposal of a new algorithm (DENSEA), which is very competitive. Moreover, in relation with the structural problem of constrained mass, two improving suggestions are handled: the multiobjectivization and the auto-adaptive rebirth operator. Finally, two multicriteria optimization problems that are of interest in industrial engineering are solved: the safety systems design (the containment spray injection system of a nuclear power plant, considering the minimization of the cost and unavailability) and the load dispatch in electric power systems generation (considering the minimization of the cost and the atmospheric pollutant emissions). In both problems some of the previous improving suggestions were applied.