[pt] As técnicas de programação genética permitem flexibilidade no processo de otimização, possibilitando sua aplicação em diferentes áreas do conhecimento e fornecendo novas maneiras para que especialistas avancem em suas áreas com mais rapidez. Parameter mapping approach é um método de otimização numérica que utiliza a programação genética para mapear valores iniciais em parâmetros ótimos para um sistema. Embora esta abordagem produza bons resultados para problemas com soluções triviais, o uso de grandes equações/árvores pode ser necessário para tornar este mapeamento apropriado em sistemas mais complexos.A fim de aumentar a flexibilidade e aplicabilidade do método a sistemas de diferentes níveis de complexidade, este trabalho introduz uma generalização utilizando a programação genética multigênica, para realizar um mapeamento multivariado, evitando grandes estruturas complexas. Foram considerados três conjuntos de funções de benchmark, variando em complexidade e dimensionalidade. Análises estatísticas foram realizadas, sugerindo que este novo método é mais flexível e mais eficiente (em média), considerando funções de benchmark complexas e de grande dimensionalidade. Esta tese também apresenta uma abordagem do novo algoritmo para otimização numérica multimodal.Este segundo algoritmo utiliza algumas técnicas de niching, baseadas no procedimento chamado de clearing, para manter a diversidade da população. Um conjunto benchmark de funções multimodais, com diferentes características e níveis de dificuldade,foi utilizado para avaliar esse novo algoritmo. A análise estatística sugeriu que esse novo método multimodal, que também utiliza programação genética multigênica,pode ser aplicado para problemas que requerem mais do que uma única solução. Como forma de testar esses métodos em problemas do mundo real, uma aplicação em nanotecnologia é proposta nesta tese: ao timização estrutural de fotodetectores de infravermelho de poços quânticos a partir de uma energia desejada. Os resultados apresentam novas estruturas melhores do que as conhecidas na literatura (melhoria de 59,09 por cento). / [en] Genetic programming techniques allow flexibility in the optimization process, making it possible to use them in different areas of knowledge and providing new ways for specialists to advance in their areas more quickly and more accurately.Parameter mapping approach is a numerical optimization method that uses genetic programming to find an appropriate mapping scheme among initial guesses to optimal parameters for a system. Although this approach yields good results for problems with trivial solutions, the use of large equations/trees may be required to make this mapping appropriate for more complex systems.In order to increase the flexibility and applicability of the method to systems of different levels of complexity, this thesis introduces a generalization by thus using multi-gene genetic programming to perform a multivariate mapping, avoiding large complex structures.Three sets of benchmark functions, varying in complexity and dimensionality, were considered. Statistical analyses carried out suggest that this new method is more flexible and performs better on average, considering challenging benchmark functions of increasing dimensionality.This thesis also presents an improvement of this new method for multimodal numerical optimization.This second algorithm uses some niching techniques based on the clearing procedure to maintain the population diversity. A multimodal benchmark set with different characteristics and difficulty levels to evaluate this new algorithm is used. Statistical analysis suggested that this new multimodal method using multi-gene genetic programming can be used for problems that requires more than a single solution. As a way of testing real-world problems for these methods, one application in nanotechnology is proposed in this thesis: the structural optimization of quantum well infrared photodetector from a desired energy.The results present new structures better than those known in the literature with improvement of 59.09 percent.

La simulation est de plus en plus utilisée dans les études de conception et d’organisation des systèmes complexes. Une étude par optimisation via simulation permet d’optimiser les paramètres d’un système afin d’obtenir les meilleures performances, estimés par la simulation. Toutefois, dans de nombreux systèmes complexes, certaines données sont incertaines (par exemple, les conditions opératoires du système ou le comportement des décideurs). En conséquence, même lorsque l’étude d’optimisation via simulation est réalisée avec le plus grand soin, les solutions obtenues peuvent se révéler inadaptées. Dans ce contexte, notre objectif est d’étudier comment optimiser, via simulation, un système afin qu’il continue d’être performant et robuste. L’étude bibliographique approfondie que nous avons menée montre que très peu de travaux en optimisation via simulation intègrent l’incertain et qu’ils peuvent être très limités dans leur capacité à fournir des solutions robustes en un temps de calcul raisonnable en particulier lorsque des métaheuristiques sont employées. Par ailleurs, la plupart des travaux existants délivrent une solution unique de conception performante du système et ne sont pas adaptés pour prendre en compte les aspects collaboratifs (groupe de décideurs). C’est pourquoi, nous avons proposé une approche originale connectant une recherche des solutions par optimisation évolutionniste multimodale et une évaluation des performances du système via simulation. Notre approche va permettre de fournir plusieurs alternatives performantes de conception d’un système et assez diversifiées pour acquérir aux décideurs une flexibilité dans le choix de la solution à implanter. De plus, nous avons exploité cette flexibilité pour intégrer, d’une part, les préférences individuelles des membres d’une équipe décisionnelle et, d’autre part, la présence de plusieurs environnements pour étudier la robustesse des solutions en un temps de traitement raisonnable par rapport à d’autres approches utilisant des méta heuristiques. Les approches proposées sont illustrées par l’optimisation d’une maille de supply chain. Grâce à cette application, nous avons montré qu’en plus de fournir un choix de solutions performantes pour dimensionner le système, nous pouvons proposer des solutions « collectivement acceptable » pour l’équipe décisionnelle et déterminer des solutions de conception robustes du système. Ces approches fournissent ainsi une flexibilité pour la phase de décision et contribuent à la prise en compte de l’incertitude dans l’optimisation via simulation d’un système. / Simulation is more and more used in studies of design and organization of complex systems. A simulation optimization study search for the system parameters that yield the best performance. However, in many complex systems, data can be uncertain (e.g., the operating conditions of the system or the behavior of decision makers). Therefore, even when the simulation optimization study is performed with the greatest care, the solutions may be inadequate. In this context, our goal is to study how to optimize, via simulation, a robust system. The extensive literature review we conducted shows that few simulation optimization approaches incorporate uncertainty and they can be very limited in their ability to provide robust solutions in a reasonable processing time, especially when metaheuristics are used. In addition, most existing approaches provide a single solution to the design problem and are not adapted to take into account the collaborative aspects (decision maker’s team). Therefore, we propose a novel approach connecting a search for solutions by evolutionary multimodal optimization and the evaluation of the system performance by simulation. Our approach allows to obtain a diverse set of designs that can be considered as efficient in terms of their performance and to provide decision-Makers with flexibility in the choice of the solution to implement. In addition, we use this flexibility to integrate first, the individual preferences of the members of decision maker’s team and secondly, the presence of multiple environments For studying the robustness of solutions in a reasonable processing time compared to other approaches based on metaheuristics. The proposed approaches are illustrated with an example of supply chain. With this application, we have shown that in addition to providing a choice of efficient solutions for sizing the system, we propose "collectively acceptable" solutions to the decision-Making team and we identify robust solutions. Then, these approaches provide flexibility to the decision phase and contribute to the consideration of uncertainty in the simulation optimization of the system.