Catálogo de publicaciones - libros

This paper elaborates on new ideas of a scatter search algorithm for solving multiobjective problems. Our approach adapts the well-known scatter search template for single objective optimization to the multiobjective field. The result is a simple and new metaheuristic called SSMO, which incorporates typical concepts from the multiobjective optimization domain such as Pareto dominance, crowding, and Pareto ranking. We evaluate SSMO with both constrained and unconstrained problems and compare it against NSGA-II. Preliminary results indicate that scatter search is a promising approach for multiobjective optimization.

In extending the Particle Swarm Optimisation methodology to multi-objective problems it is unclear how global guides for particles should be selected. Previous work has relied on metric information in objective space, although this is at variance with the notion of dominance which is used to assess the quality of solutions. Here we propose methods based exclusively on dominance for selecting guides from a non-dominated archive. The methods are evaluated on standard test problems and we find that probabilistic selection favouring archival particles that dominate few particles provides good convergence towards and coverage of the Pareto front. We demonstrate that the scheme is robust to changes in objective scaling. We propose and evaluate methods for confining particles to the feasible region, and find that allowing particles to explore regions close to the constraint boundaries is important to ensure convergence to the Pareto front.

Based on the concept of Immunodominance and Antibody Clonal Selection Theory, we propose a new artificial immune system algorithm, Immune Dominance Clonal Multiobjective Algorithm (IDCMA). The influences of main parameters are analyzed empirically. The simulation comparisons among IDCMA, the Random-Weight Genetic Algorithm and the Strength Pareto Evolutionary Algorithm show that when low-dimensional multiobjective problems are concerned, IDCMA has the best performance in metrics such as Spacing and Coverage of Two Sets.

Real-world engineering optimisation problems are typically multi-objective and highly constrained, and constraints may be both costly to evaluate and binary in nature. In addition, objective functions may be computationally expensive and, in the commercial design cycle, there is a premium placed on rapid initial progress in the optimisation run. In these circumstances, evolutionary algorithms may not be the best choice; we have developed a multi-objective Tabu Search algorithm, designed to perform well under these conditions. Here we present the algorithm along with the constraint handling approach, and test it on a number of benchmark constrained test problems. In addition, we perform a parametric study on a variety of unconstrained test problems in order to determine the optimal parameter settings. Our algorithm performs well compared to a leading multi-objective Genetic Algorithm, and we find that its performance is robust to parameter settings.

In this paper, we propose a new Multi-Objective Particle Swarm Optimizer, which is based on Pareto dominance and the use of a crowding factor to filter out the list of available leaders. We also propose the use of different mutation (or turbulence ) operators which act on different subdivisions of the swarm. Finally, the proposed approach also incorporates the ∈-dominance concept to fix the size of the set of final solutions produced by the algorithm. Our approach is compared against five state-of-the-art algorithms, including three PSO-based approaches recently proposed. The results indicate that the proposed approach is highly competitive, being able to approximate the front even in cases where all the other PSO-based approaches fail.

Differential Evolution (DE) is a simple but powerful evolutionary optimization algorithm with many successful applications. In this paper we propose Differential Evolution for Multiobjective Optimization (DEMO) – a new approach to multiobjective optimization based on DE. DEMO combines the advantages of DE with the mechanisms of Pareto-based ranking and crowding distance sorting, used by state-of-the-art evolutionary algorithms for multiobjective optimization. DEMO is implemented in three variants that achieve competitive results on five ZDT test problems.

In this article, model selection for support vector machines is viewed as a multi-objective optimization problem, where model complexity and training accuracy define two conflicting objectives. Different optimization criteria are evaluated: Split modified radius margin bounds, which allow for comparing existing model selection criteria, and the training error in conjunction with the number of support vectors for designing sparse solutions.

In previous work, we have proposed a novel approach to data clustering based on the explicit optimization of a partitioning with respect to two complementary clustering objectives [6]. Here, we extend this idea by describing an advanced multiobjective clustering algorithm, MOCK, with the capacity to identify good solutions from the Pareto front, and to automatically determine the number of clusters in a data set. The algorithm has been subject to a thorough comparison with alternative clustering techniques and we briefly summarize these results. We then present investigations into the mechanisms at the heart of MOCK: we discuss a simple example demonstrating the synergistic effects at work in multiobjective clustering, which explain its superiority to single-objective clustering techniques, and we analyse how MOCK’s Pareto fronts compare to the performance curves obtained by single-objective algorithms run with a range of different numbers of clusters specified.

An artificial satellite design requires severe design objectives such as performance, reliability, weight, robustness, cost, and so on. To solve the conflicted requirements at the same time, multiobjective optimization is getting more popular in the design. Using the optimization, it becomes ordinary to get many solutions, such as Pareto solutions, quasi-Pareto solutions, and feasible solutions. The alternative solutions, however, are very difficult to be adopted to practical engineering decision directly. Therefore, to make the decision, proper information about the solutions in a function, parameter and real design space should be provided. In this paper, a new approach for the interpretation of Pareto solutions is proposed based on multidimensional visualization and clustering. The proposed method is applied to a thermal robustness and mass optimization problem of heat pipe shape design for an artificial satellite. The information gleaned from the propose approach can support the engineering decision for the design of artificial satellite heat pipe.

A comparative study of the use of Gray coding in multicriteria evolutionary optimisation is performed using the SPEA2 and NSGAII algorithms and applied to a frame structural optimisation problem. A double minimization is handled: constrained mass and number of different cross-section types. Influence of various mutation rates is considered. The comparative statistical results of the test case cover a convergence study during evolution by means of certain metrics that measure front amplitude and distance to the optimal front. Results in a 55 bar-sized frame test case show that the use of the Standard Binary Reflected Gray code compared versus Binary code allows to obtain fast and more accurate solutions, more coverage of non-dominated fronts; both with improved robustness in frame structural multiobjective optimum design.