Solving optimization problems with multiple (often conflicting) objectives is, generally, a very difficult goal. Evolutionary algorithms (EAs) were initially extended and applied during the mid-eighties in an attempt to stochastically solve problems of this generic class. During the past decade, a variety of multiobjective EA (MOEA) techniques have been proposed and applied to many scientific and engineering applications. Our discussion's intent is to rigorously define multiobjective optimization problems and certain related concepts, present an MOEA classification scheme, and evaluate the variety of contemporary MOEAs. Current MOEA theoretical developments are evaluated; specific topics addressed include fitness functions, Pareto ranking, niching, fitness sharing, mating restriction, and secondary populations. Since the development and application of MOEAs is a dynamic and rapidly growing activity, we focus on key analytical insights based upon critical MOEA evaluation of c...

Abstract. This work proposes a quantitative, non-parametric interpretation of statistical performance of stochastic multiobjective optimizers, including, but not limited to, genetic algorithms. It is shown that, according to this interpretation, typical performance can be defined in terms analogous to the notion of median for ordinal data, as can other measures analogous to other quantiles. Non-parametric statistical test procedures are then shown to be useful in deciding the relative performance of different multiobjective optimizers on a given problem. Illustrative experimental results are provided to support the discussion. 1

s: Jan. 1992 -- Dec. 1994 ffl CTI: Current Technology Index Jan./Feb. 1993 -- Jan./Feb. 1994 ffl DAI: Dissertation Abstracts International: Vol. 53 No. 1 -- Vol. 55 No. 4 (1994) ffl EEA: Electrical & Electronics Abstracts: Jan. 1991 -- Dec. 1994 ffl P: Index to Scientific & Technical Proceedings: Jan. 1986 -- Feb. 1995 (except Nov. 1994) ffl EI A: The Engineering Index Annual: 1987 -- 1992 ffl EI M: The Engineering Index Monthly: Jan. 1993 -- Dec. 1994 The following GA researchers have already kindly supplied their complete autobibliographies and/or proofread references to their papers: Dan Adler, Patrick Argos, Jarmo T. Alander, James E. Baker, Wolfgang Banzhaf, Ralf Bruns, I. L. Bukatova, Thomas Back, Yuval Davidor, Dipankar Dasgupta, Marco Dorigo, Bogdan Filipic, Terence C. Fogarty, David B. Fogel, Toshio Fukuda, Hugo de Garis, Robert C. Glen, David E. Goldberg, Martina Gorges-Schleuter, Jeffrey Horn, Aristides T. Hatjimihail, Mark J. Jakiela, Richard S. Judson, Akihiko Konaga...

Research into solving multiobjective optimization problems (MOP) has as one of its an overall goals that of developing and defining foundations of an Evolutionary Computation (EC)-based MOP theory. In this paper, we introduce relevant MOP concepts, and the notion of Pareto optimality, in particular. Specific notation is defined and theorems are presented ensuring Paretobased Evolutionary Algorithm (EA) implementations are clearly understood. Then, a specific experiment investigating the convergence of an arbitrary EA to a Pareto front is presented. This experiment gives a basis for a theorem showing a specific multiobjective EA statistically converges to the Pareto front. We conclude by using this work to justify further exploration into the theoretical foundations of EC-based MOP solution methods. 1 Introduction Our research focuses on solving scientific and engineering multiobjective optimization problems (MOPs), contributing to the overall goal of developing and defining foundatio...

This paper presents an evolutionary algorithm with a new goal-sequence domination scheme for better decision support in multi-objective optimization. The approach allows the inclusion of advanced hard/soft priority and constraint information on each objective component, and is capable of incorporating multiple specifications with overlapping or non-overlapping objective functions via logical "OR" and "AND" connectives to drive the search towards multiple regions of trade-off. In addition, we propose a dynamic sharing scheme that is simple and adaptively estimated according to the on-line population distribution without needing any a priori parameter setting. Each feature in the proposed algorithm is examined to show its respective contribution, and the performance of the algorithm is compared with other evolutionary optimization methods. It is shown that the proposed algorithm has performed well in the diversity of evolutionary search and uniform distribution of non-dominated individuals along the final trade-offs, without significant computational effort. The algorithm is also applied to the design optimization of a practical servo control system for hard disk drives with a single voice-coil-motor actuator. Results of the evolutionary designed servo control system show a superior closed-loop performance compared to classical PID or RPT approaches.

Cet article pr'esente un 'etat de l'art des m'etaheuristiques appliqu'ees `a la r'esolution de probl`emes d'optimisation combinatoire multi-objectif, suivant une classification propos'ee. L'objectif principal de telles m'ethodes est de g'en'erer une vari'et'e de solutions Pareto-optimales diversifi'ees dans l'espace de recherche. Une analyse critique de chaque classe de m'ethode est pr'esent'ee. Des r'eponses `a des questions ouvertes comme l"evaluation des performance et la comparaison d'algorithmes d'optimisation multiobjectif, et l"etude des paysages des fronti`eres Pareto sont abord'ees. Certains axes de recherche future dans ce domaine tel que la conception d'algorithmes parall`eles et hybrides sont aussi identifi'es. Keywords Optimisation combinatoire multi-objectif, Optimalit'e Pareto, M'etaheuristiques, Paysages de recherche, Evaluation de performances, Algorithmes g'en'etiques, Recherche tabou, Recuit simul'e. Ce travail a b'en'efici'e du soutien du CNET au titre du m...

Abstract. We dwell in largely non-technical terms on the essential differences between single-objective optimization and multiple-objective optimization. We argue in particular that single-objective approaches to real-world problems are almost invariably simplifications of the real-problem which make many ideal solutions unreachable to the optimization method. We promote the use of multi-objective optimization methods, particularly those arising from the evolutionary computation community. We point out that the state of the art in the field of evolutionary multi-objective optimization is such that fast and effective techniques are now available which are capable of finding a well-distributed set of diverse trade-off solutions, with little or no more computational effort than sophisticated single-objective optimizers would have taken to find a single one. The resulting diversity of ideas available through a multi-objective approach leads both to the problem-solver being furnished with a better understanding of the space of possible solutions, and consequently to a better final solution to the problem at hand. We end by very briefly charting the history of the field and hinting at the range of published applications and ongoing research issues. 1

Dynamic functional – link neural networks genetically evolved applied to system identification

This paper presents an evolutionary grey-box model identification methodology that makes the best use of a-priori knowledge on a clear-box model with a global structural representation of the physical system under study, whilst incorporating accurate black-box models for immeasurable and local nonlinearities of a practical system. The evolutionary technique is applied to building dominant structural identification with local parametric tuning without the need of a differentiable performance index in the presence of noisy data. It is shown that the evolutionary technique provides an excellent fitting performance and is capable of accommodating multiple objectives such as to examine the relationships between model complexity and fitting accuracy during the model building process. Validation results show that the proposed method offers robust, uncluttered and accurate models for two practical systems. It is expected that this type of grey-box models will accommodate many practical engineering systems for a better modelling accuracy.

Sizing a pump stacking used in an aircraft lubricationsystem is a challenging task.ThecomThe.#F# of several pumal in parallel and in a single casing,msi deliver specified oil flow rates, on a variablenumab of circuits, and under given flight conditions. Furthermore,theoptim- assem-y has tom.##FyF overall dimall.#T3 weight and cost.Thisoptimis.z3F problem involves a large space search, continuous and discrete variables and m.3T3z#.STyyGj.mGj.m. Algorithm (GA)---stochastic search march. that mat. theme.#GyT of natural biological evolution---seem well suited to solve that kind ofproblemSz new GA is proposed.The efficiency of this GA is #rst dem3G#yT.Sz in solving variousmious.TT3yF test-cases and then applied to the industrial problem.