Catálogo de publicaciones - libros

Evolutionary algorithms have been quite effective in dealing with single-objective “optimization” while the area of Evolutionary Multiobjective Optimization (EMOO) has extended its efficiency to Multiple Criteria Decision Making (MCDM) as well. The number of technical publications in EMOO is impressive and indicative of a rather explosive growth in recent years. It is fair to say however that most of the progress has been in applying and evolving algorithms and their convergence properties, not in evolving the optimality concept itself, nor in expanding the notions of true optimization. Yet, the conceptual constructs based on evolution and Darwinian selection have probably most to contribute – at least in theory – to the evolution of optimality. They should be least dependent on a priori fixation of anything in problem formulation: constraints, objectives or alternatives. Modern systems and problems are typical for their flexibility , not for their fixation. In this paper we draw attention to the impossibility of optimization when crucial variables are given and present Eight basic concepts of optimality . In the second part of this contribution we choose a more realistic problem of linear programming where constraints are not “given” but flexible and to be optimized and objective functions are multiple: De novo programming .

Evolutionary multicriteria optimization has traditionally concentrated on problems comprising 2 or 3 objectives. While engineering design problems can often be conveniently formulated as multiobjective optimization problems, these often comprise a relatively large number of objectives. Such problems pose new challenges for algorithm design, visualisation and implementation. Each of these three topics is addressed. Progressive articulation of design preferences is demonstrated to assist in reducing the region of interest for the search and, thereby, simplified the problem. Parallel coordinates have proved a useful tool for visualising many objectives in a two-dimensional graph and the computational grid and wireless Personal Digital Assistants offer technological solutions to implementation difficulties arising in complex system design.

After 20 years of development of multiobjective metaheuristics the procedures for solving multiple objective combinatorial optimization problems are generally the result of a blend of evolutionary, neighborhood search, and problem dependent components. Indeed, even though the first procedures were direct adaptations of single objective metaheuristics inspired by evolutionary algorithms or neighborhood search algorithms, hybrid procedures have been introduced very quickly. This paper discusses hybridations found in the literature and mentions recently introduced metaheuristic principles.

Due to the vagaries of optimization problems encountered in practice, users resort to different algorithms for solving different optimization problems. In this paper, we suggest an optimization procedure which specializes in solving multi-objective, multi-global problems. The algorithm is carefully designed so as to degenerate to efficient algorithms for solving other simpler optimization problems, such as single-objective uni-global problems, single-objective multi-global problems and multi-objective uni-global problems. The efficacy of the proposed algorithm in solving various problems is demonstrated on a number of test problems. Because of it’s efficiency in handling different types of problems with equal ease, this algorithm should find increasing use in real-world optimization problems.

The hypervolume measure is one of the most frequently applied measures for comparing the results of evolutionary multiobjective optimization algorithms (EMOA). The idea to use this measure for selection is self-evident. A steady-state EMOA will be devised, that combines concepts of non-dominated sorting with a selection operator based on the hypervolume measure. The algorithm computes a well distributed set of solutions with bounded size thereby focussing on interesting regions of the Pareto front(s). By means of standard benchmark problems the algorithm will be compared to other well established EMOA. The results show that our new algorithm achieves good convergence to the Pareto front and outperforms standard methods in the hypervolume covered. We also studied the applicability of the new approach in the important field of design optimization. In order to reduce the number of time consuming precise function evaluations, the algorithm will be supported by approximate function evaluations based on Kriging metamodels. First results on an airfoil redesign problem indicate a good performance of this approach, especially if the computation of a small, bounded number of well-distributed solutions is desired.

Contemporary evolutionary multiobjective optimisation techniques are becoming increasingly focussed on the notions of archiving, explicit diversity maintenance and population-based Pareto ranking to achieve good approximations of the Pareto front. While it is certainly true that these techniques have been effective, they come at a significant complexity cost that ultimately limits their application to complex problems. This paper proposes a new model that moves away from explicit population-wide Pareto ranking, abandons both complex archiving and diversity measures and incorporates a continuous accretion-based approach that is divergent from the discretely generational nature of traditional evolutionary algorithms. Results indicate that the new approach, the Combative Accretion Model (CAM), achieves markedly better approximations than NSGA across a range of well-recognised test functions. Moreover, CAM is more efficient than NSGAII with respect to the number of comparisons (by an order of magnitude), while achieving comparable, and generally preferable, fronts.

While single-objective Evolutionary Algorithms (EAs) parallelization schemes are both well established and easy to implement, this is not the case for Multi-Objective Evolutionary Algorithms (MOEAs). Nevertheless, the need for parallelizing MOEAs arises in many real-world applications, where fitness evaluations and the optimization process can be very time consuming. In this paper, we test the ‘divide and conquer’ approach to parallelize MOEAs, aimed at improving the speed of convergence beyond a parallel island MOEA with migration. We also suggest a clustering based parallelization scheme for MOEAs and compare it to several alternative MOEA parallelization schemes on multiple standard multi-objective test functions.

An improved orthogonal multi-objective evolutionary algorithm (OMOEA), called OMOEA-II, is proposed in this paper. Two new crossovers used in OMOEA-II are orthogonal crossover and linear crossover. By using these two crossover operators, only small orthogonal array rather than large orthogonal array is needed for exploiting optimal in the global space. Such reduction in orthogonal array can avoid exponential creation of solutions of OMOEA and improve the performance in robusticity without degrading precision and distribution of solutions. Experimental results show that OMOEA-II can solve problems with high dimensions and large number of local Pareto-optimal fronts better than some existing algorithms recently reported in the literatures.

Path relinking algorithms have proved their efficiency in single objective optimization. Here we propose to adapt this concept to Pareto optimization. We combine this original approach to a genetic algorithm. By applying this hybrid approach to a bi-objective permutation flow-shop problem, we show the interest of this approach. In this paper, we present first an Adaptive Genetic Algorithm dedicated to obtain a first well diversified approximation of the Pareto set. Then, we present an original hybridization with Path Relinking algorithm, in order to intensify the search between solutions obtained by the first approach. Results obtained are promising and show that cooperation between these optimization methods could be efficient for Pareto optimization.

This paper proposes a new multiobjective evolutionary approach—the dynamic archive evolution strategy (DAES) to investigate the adaptive balance between proximity and diversity. In DAES, a novel dynamic external archive is proposed to store elitist individuals as well as relatively better individuals through archive increase scheme and archive decrease scheme. Additionally, a combinatorial operator that inherits merits from Gaussian mutation of proximity exploration and Cauchy mutation of diversity preservation is elaborately devised. Meanwhile, a complete nondominance selection ensures maximal pressure of proximity exploitation while a corresponding fitness assignment ensures the similar pressure of diversity preservation. By graphical presentation and performance metrics on three prominent benchmark functions, DAES is found to outperform three state-of-the-art multiobjective evolutionary algorithms to some extent in terms of finding a near-optimal, well-extended and uniformly diversified Pareto optimal front.