Cooperation in self-organized heterogeneous swarms is a phenomenon from nature with many applications in autonomous robots. I specifically analyzed the problem of auto-regulated team formation in multi-agent systems and several strategies to learn socially how to make multi-objective decisions. To this end I proposed new multi-objective ranking relations and analyzed their properties theoretically and within multi-objective metaheuristics. The results showed that simple decision mechanism suffice to build effective teams of heterogeneous agents and that diversity in groups is not a problem but can increase the efficiency of multi-agent systems.

[ES] Aquesta tesi presenta els resultats de la feina de recerca dut a terme sobre el modelatge i el disseny de controladors per a micro-aeronaus no tripulades mitjançant tècniques d'optimització multi-objectiu. Dos principals camps d'estudi estan presents al llarg d'ella. D'una banda, l'estudi de com modelar i controlar plataformes aèries de petita envergadura. I, de l'altra, l'estudi sobre l'ús de tècniques heurístiques d'optimització multi-objectiu per aplicar en el procés de parametrització de models i controladors en micro-aeronaus no tripulades. S'obtenen com a resultat principal una sèrie d'eines que permeten prescindir d'experiments en túnels de vent o de sensòrica d'alt cost, passant directament a la utilització de dades de vol experimental a la identificació paramètrica de models dinàmics. A més, es demostra com la utilització d'eines d'optimització multi-objectiu en diferents fases de desenvolupament de controladors ajuda a augmentar el coneixement sobre la plataforma a controlar i augmenta la fiabilitat i robustesa dels controladors desenvolupats, disminuint el risc de passar de les fases prèvies de el disseny a la validació en vol real. / [CA] Esta tesis presenta los resultados del trabajo de investigación llevado a cabo sobre el modelado y el diseño de controladores para micro-aeronaves no tripuladas mediante técnicas de optimización multi-objetivo. Dos principales campos de estudio están presentes a lo largo de ella. Por un lado, el estudio de cómo modelar y controlar plataformas aéreas de pequeña envergadura. Y, por otro, el estudio sobre el empleo de técnicas heurísticas de optimización multi-objetivo para aplicar en el proceso de parametrización de modelos y controladores en micro-aeronaves no tripuladas. Se obtienen como resultado principal una serie de herramientas que permiten prescindir de experimentos en túneles de viento o de sensórica de alto coste, pasando directamente a la utilización de datos de vuelo experimental en la identificación paramétrica de modelos dinámicos. Además, se demuestra como la utilización de herramientas de optimización multi-objetivo en diferentes fases del desarrollo de controladores ayuda a aumentar el conocimiento sobre la plataforma a controlar y aumenta la fiabilidad y robustez de los controladores desarrollados, disminuyendo el riesgo de pasar de las fases previas del diseño a la validación en vuelo real. / [EN] This thesis presents the results of the research work carried out on the modelling and design of controllers for micro-unmanned aerial vehicles by means of multi-objective optimization techniques. Two main fields of study are present throughout it. On one hand, the study of how to model and control small aerial platforms. And, on the other, the study on the use of heuristic multi-objective optimization techniques to apply in the process of models and controllers parameterization in micro-unmanned aerial vehicles. The main result is a series of tools that make it possible manage without wind tunnel experiments or high-cost air-data sensors, going directly to the use of experimental flight data in the parametric identification of dynamic models. In addition, a demonstration is given on how the use of multi-objective optimization tools in different phases of controller development helps to increase knowledge about the platform to be controlled and increases the reliability and robustness of the controllers developed, reducing the risk of hoping from the initial design phases to validation in real flight. / Velasco Carrau, J. (2020). Unmanned Aerial Vehicles Modelling and Control Design. A Multi-Objective Optimization Approach [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/156034 / TESIS

[ES] En esta tesis doctoral se estudia el problema de secuenciación de máquinas paralelas no relacionadas con necesidad de ajustes y recursos adicionales asignados en los ajustes. En este problema, se tiene un grupo de tareas (también llamadas trabajos), donde cada una debe ser procesada en una de las máquinas paralelas disponibles. Para procesar una tarea después de otra en la misma máquina, se debe hacer un ajuste en la máquina. Se asume que estos ajustes deben ser realizados por un recurso adicional limitado (por ejemplo, operarios). En esta tesis doctoral se estudian dos variantes del problema planteado: 1) considerando el problema con el único objetivo de minimizar el tiempo máximo de finalización de todos los trabajos (makespan), y 2) considerando el problema multi-objetivo minimizando simultáneamente el makespan y el consumo máximo de recursos adicionales. Inicialmente, se realiza una completa revisión bibliográfica sobre estudios relacionados con el problema planteado. En esta revisión se detecta que, a pesar de existir numerosos estudios de secuenciación de máquinas paralelas, no muchos de estos estudios tienen en cuenta recursos adicionales. Posteriormente, para introducir el problema a estudiar antes de plantear métodos de resolución, se realiza una breve explicación de los principales problemas de secuenciación de máquinas paralelas. El problema de un solo objetivo está clasificado como NP-Hard. Por ello, para abordar su resolución se han diseñado e implementado heurísticas y metaheurísticas siguiendo dos enfoques diferentes. Para el primer enfoque, que ignora la información sobre el consumo de recursos adicionales en la fase constructiva, se adaptan dos de los mejores algoritmos existentes en la literatura para el problema de máquinas paralelas con ajustes sin necesidad de recursos adicionales. En el segundo enfoque, que sí tiene en cuenta la información sobre el consumo de recursos adicionales en la fase constructiva, se proponen nuevos algoritmos heurísticos y metaheurísticos para resolver el problema. Tras analizar los resultados de los experimentos computacionales realizados, concluimos que hay diferencias entre los dos enfoques, siendo significativamente mejor el enfoque que tiene en cuenta la información sobre los recursos adicionales. Al igual que en el caso de un solo objetivo, la complejidad del problema multi-objetivo obliga a presentar algoritmos heurísticos o metaheurísticos para resolverlo. En esta tesis se presenta un nuevo algoritmo metaheurístico multi-objetivo eficiente para encontrar buenas aproximaciones a la frontera de Pareto del problema. Además, se adaptaron otros tres algoritmos que han mostrado buenos resultados en diferentes estudios de problemas de secuenciación de máquinas multi-objetivo. Después de realizar experimentos computacionales exhaustivos, concluimos que el nuevo algoritmo propuesto en esta tesis es significativamente mejor que los otros tres algoritmos existentes, y que se han adaptado para resolver este problema. / [CAT] En aquesta tesi doctoral s'estudia el problema de seqüenciació de màquines paral·leles no relacionades amb necessitat d'ajustos i recursos addicionals assignats en els ajustos. En aquest problema, es tenen un grup de tasques (també anomenades treballs), on cadascuna ha de ser processada en una de les màquines paral·leles disponibles. Per processar una tasca després d'una altra en la mateixa màquina, s'ha de fer un ajustament en la màquina. S'assumeix que aquests ajustos en les màquines per a processar una tasca després del processament d'una altra, han de ser realitzats per un recurs addicional limitat (per exemple, operaris). En aquesta tesi doctoral s'estudien dos variants al problema plantejat: 1) considerant el problema com l'únic objectiu de minimitzar el temps màxim de finalització de tots els treballs (makespan), i 2) considerant el problema multi-objectiu minimitzant simultàniament el makespan i el consum màxim de recursos addicionals. Inicialment, es realitza una completa revisió bibliogràfica sobre estudis relacionats amb el problema plantejat. En esta revisió es detecta que, tot i existir nombrosos estudis de seqüenciació de màquines paral·leles, hi ha molts pocs que tenen en compte recursos addicionals. Posteriorment, per introduir el problema a estudiar abans de plantejar mètodes de resolució, es realitza una breu explicació dels principals problemes de seqüenciació de màquines paral·leles. El problema d'un sol objectiu està classificat com NP-Hard. Per això, per abordar la seua resolució s'han dissenyat i implementat heurístiques y metaheurístiques seguint dos enfocs diferents. El primer enfoc ignora la informació sobre el consum de recursos en la fase constructiva, adaptant dos dels millors algoritmes existents en la literatura per al problema de seqüenciació de màquines paral·leles amb ajustaments sense necessitat de recursos. Per al segon enfoc si es té en compte la informació sobre el consum de recursos en la fase constructiva. Després d'analitzar els resultats dels experiments computacionals realitzats, concloem que hi ha diferencies entre els dos enfocs, sent significativament millor l'enfoc que té en compte la informació sobre el recursos. De la mateixa manera que en el cas d'un sol objectiu, la complexitat del problema multi-objectiu obliga a presentar algoritmes heurístics o metaheurístics per a resoldre-ho. En aquesta tesi es presenta un nou algoritme metaheurístic multi-objectiu eficient per trobar bones aproximacions a la frontera de Pareto del problema. A més, es van adaptar altres tres algoritmes que han mostrat bons resultats en diferents estudis de problemes de seqüenciació de màquines multi-objectiu. Després de realitzar experiments computacionals exhaustius, concloem que el nou algoritme proposat en aquesta tesi és significativament millor que els altres tres algoritmes existents i que s'han adaptat per resoldre aquest problema. / [EN] In this thesis we study the unrelated parallel machine scheduling problem with setup times and additional limited resources in the setups. In this problem, we have a group of tasks (also called jobs), where each one must be processed on one of the available parallel machines. To process one job after another on the same machine, a setup must be made on the machine. It is assumed that these setups on machines must be made by a limited additional resource (eg, operators). In this thesis two variants of the problem are studied: 1) considering the problem with the objective of minimizing the maximum completion time of all jobs (makespan), and 2) considering the multi-objective problem, minimizing the makespan and the maximum consumption of additional resources. Initially, a complete literature review is carried out on studies related to the problem addressed in this thesis. This review finds that despite numerous parallel machine scheduling studies, there are very few that take into account additional resources. Subsequently, to introduce the problem addressed before proposing resolution methods, a brief explanation of the main parallel machines scheduling problems is made. The problem with a single objective is classified as NP-Hard. Therefore, to solve it, heuristics and metaheuristics have been designed and implemented following two different approaches. For the first approach, which ignores the information on the consumption of resources in the construction phase, two of the best algorithms existing in the literature for the problem of parallel machines with setups without additional resources are adapted. For the second approach, which does take into account information on the consumption of resources in the construction phase, new heuristic and metaheuristic algorithms are proposed to solve the problem. Following the results of the computational experiments, we conclude that there are differences between the two approaches, the approach that takes into account the information on resources being significantly better. As in the case of a single objective, the complexity of the multi-objective problem requires the formulation of heuristic or metaheuristic algorithms to solve it. In this thesis, a new efficient multi-objective metaheuristic algorithm is presented to find good approximations to the Pareto front of the problem. In addition, three other algorithms that have shown good results in different studies of multi-objective machine scheduling problems were adapted. After carrying out exhaustive computational experiments, we concluded that the new algorithm proposed in this thesis is significantly better than the other three adapted algorithms. / Yepes Borrero, JC. (2020). Secuenciación de máquinas con necesidad de ajustes y recursos adicionales [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/158742 / TESIS

Due to the changes and improvements that have occurred over the years, the manufacturing sector has evolved. Companies in the 21st-century face changes in the marketplace that are difficult to predict due to international competition and the rapid emergence of new products. To cope, companies must reinvent themselves and design manufacturing systems that seek to produce quality and low-cost products, and respond to the changes that must be faced. These capabilities are encompassed in reconfigurable manufacturing systems (RMS), capable of dealing with uncertainties quickly and economically. On the other hand, production planning with this type of system presents a significant challenge. Although simulation-based optimization techniques have been applied to address certain RMS challenges, only a few studies have applied simulation-based multi-objective optimization to simultaneously address several conflicting design objectives, as is the case in this project. This project aims to investigate some aspects using SBMO that directly affect the performance of a plant and demonstrate the usefulness of the method. / <p>Det finns övrigt digitalt material (t.ex. film-, bild- eller ljudfiler) eller modeller/artefakter tillhörande examensarbetet som ska skickas till arkivet.</p>

Ces dernières décennies, l'industrie en microélectronique s'est astreinte à suivre la loi de Moore pour améliorer la performance des circuits intégrés (Integrated Circuit, IC). Cependant, il sera sans doute impossible de suivre cette loi dans le futur à cause de limitations physiques apparaissant avec la miniaturisation des transistors en-dessous d'un certain seuil si aucune innovatio n'a lieu. Afin de surmonter ce problème, de nouvelles technologies ont émergées, et parmi elles les circuits 3D (3D-Stacked Integrated Circuit, 3D-SIC) ont été proposés pour maintenir l'évolution de la loi de Moore. Les 3D-SIC peuvent apporter de nombreux avantages dans le design des futurs IC mais au coût d'une complexité de design accrue étant donné leur nature fortement combinatoire, et l'optimisation de plusieurs critères conflictuels. Dans cette thèse, nous présentons une première étude des outils qui pourraient aider dans le design de 3D-SIC, en utilisant l'optimisation multi-objectifs (multiobjective optimization, MOO) et l'aide multicritère à la décision (multi-criteria decision aid, MCDA). Notre étude vise l'une des problématiques principales dans le design de 3D-SIC: le partitionnement avec estimation du floorplanning en tenant compte de plusieurs objectifs. Cette thèse montre que l'utilisation d'un paradigme multicritère peut fournir une analyse pertinente et objective du problème. Cela peut permettre une exploration rapide de l'espace de design et une amélioration des flots de conception actuels étant donné qu'il est possible de fournir des informations qualitatives et quantitatives par rapport à l'espace de design qui ne seraient pas disponibles avec les outils actuels. De même, de par sa flexibilité, la MOO peut tenir compte des multiples degrés de liberté des 3D-SIC, ce qui permet plus de possibilités de design qui ne sont généralement pas prises en compte avec les outils actuels. De plus, les algorithmes développés peuvent montrer des propriétés de robustesse même si le problème est complexe. Enfin, appliquer l'aide multicritère à la décision pourrait permettre aux designers de faire des choix pertinents selon un processus transparent. / In the past decades, the microelectronic industry has been following the Moore's law to improve the performance of integrated circuits (IC). However, it will probably be impossible to follow this law in the future due to physical limitations appearing with the miniaturization of the transistors below a certain threshold without innovation. In order to overcome this problem, new technologies have emerged, and among them the 3D-Stacked Integrated Circuits (3D-SIC) have been proposed to keep the Moore's momentum alive. 3D-SICs can bring numerous advantages in the design of future ICs but at the cost of additional design complexity due to their highly combinatorial nature, and the optimization of several conflicting criteria. In this thesis, we present a first study of tools that can help the design of 3D-SICs, using mutiobjective optimization (MOO) and multi-criteria decision aid (MCDA). Our study has targeted one of the main issues in the design of 3D-SICs: the partitioning with floorplanning estimation under multiple objectives. This thesis shows that the use of a multi-criteria paradigm can provide relevant and objective analysis of the problem. This can allow a quick design space exploration and an improvement of the current design flows as it is possible to provide qualitative and quantitative information about a design space, that would not be available with current tools. Also, with its flexibility, MOO can cope with the multiple degrees of freedom of 3D-SICs, which enables more design possibilities that are usually not taken into account with current tools. In addition, the developed algorithms can show robustness properties even if the problem is complex. Finally, applying multi-criteria decision aid would allow designers to make relevant choices in a transparent process. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

The transport sector is a major cause of emissions both in Sweden and globally. This master thesis aims to develop a model for estimating emissions from heavy transport on a specific route. The emissions can be used in a route planning software and help the driver choose a route that contributes to reduced emissions. The methodology was to investigate attributes, like vehicle-related attributes and topography, and their impact on transport emissions. The carbon dioxide, methane and nitrous oxide emissions were converted into carbon dioxide equivalents, which were incorporated as one cost together with a precalculated driving time as a second cost in a multi objective function used for route planning. Different tests were conducted to investigate the accuracy and the usability of the model. First, a validation test was performed, where the optimized routes were analyzed. The test showed that the model was more likely to choose a shorter route in general. The fuel consumption values largely met expectations towards generic values and measurements, that were gathered from research. A second test of the model made use of the driving time combined with the emissions in a multi objective function. In this test, a weighting coefficient was varied and analyzed to understand the possibility to find a value of the coefficient for the best trade-off. The result showed that the model generates different solutions for different coefficients and that it is possible to find a suitable trade-off between the driving time and emissions. Therefore, this study shows that there is a possibility to combine emission with other objectives such as driving time for route optimization. Finally, a sensitivity analysis was performed, where attribute factors and assumptions were varied to see how sensitive they were and, in turn, how much a change would impact the calculated emissions. The result from the sensitivity analysis showed that the changes in topography-attributes had less impact than changes on vehicle-related attributes. In conclusion, this thesis has built a foundation for route planning, based on the environmental aspect, for heavy transports.

Multi-objective evolutionary algorithms have gained a lot of atten- tion in the recent years. They have proven to be among the best multi-objective optimizers and have been used in many industrial ap- plications. However, their usability is hindered by the large number of evaluations of the objective functions they require. These can be expensive when solving practical tasks. In order to reduce the num- ber of objective function evaluations, surrogate models can be used. These are a simple and fast approximations of the real objectives. In this work we present the results of research made between the years 2009 and 2013. We present a multi-objective evolutionary algo- rithm with aggregate surrogate model, its newer version, which also uses a surrogate model for the pre-selection of individuals. In the next part we discuss the problem of selection of a particular type of model. We show which characteristics of the various models are im- portant and desirable and provide a framework which combines sur- rogate modeling with meta-learning. Finally, in the last part, we ap- ply multi-objective optimization to the problem of hyper-parameters tuning. We show that additional objectives can make finding of good parameters for classifiers faster. 1

Background: Multiple Objective Optimization problems(MOOPs) are common and evident in every field. Container port terminals are one of the fields in which MOOP occurs. In this research, we have taken a case in logistics management and modelled Multi-agent systems to solve the MOOP using Non-dominated Sorting Genetic Algorithm-II (NSGA-II). Objectives: The purpose of this study is to build AI-based models for solving a Multiple Objective Optimization Problem occurred in port terminals. At first, we develop a port agent with an objective function of maximizing throughput and a customer agent with an objective function of maximizing business profit. Then, we solve the problem using the single-objective optimization model and multi-objective optimization model. We then compare the results of both models to assess their performance. Methods: A literature review is conducted to choose the best algorithm among the existing algorithms, which were used previously in solving other Multiple Objective Optimization problems. An experiment is conducted to know how well the models performed to solve the problem so that all the participants are benefited simultaneously. Results: The results show that all three participants that are port, customer one and customer two have gained profits by solving the problem in multi-objective optimization model. Whereas in a single-objective optimization model, a single participant has achieved earnings at a time, leaving the rest of the participants either in loss or with minimal profits. Conclusion: We can conclude that multi-objective optimization model has performed better than the single-objective optimization model because of the impartial results among the participants.

Les infrastructures critiques (CIs) sont essentielles au fonctionnement de la société moderne. Leur sécurité et leur fiabilité sont les principales préoccupations. La complexité des CIs exige des approches d'analyse de système capables de voir le problème de plusieurs points de vue. La présente thèse porte sur l'intégration de la perspective de contrôle dans l'analyse de sécurité et de fiabilité des éléments de configuration. L'intégration est d'abord abordée par examiner les propriétés de contrôle d'un microgrid d'alimentation électrique. Un schéma basé sur la simulation est développé pour l'analyse sous différentes perspectives : le service d'approvisionnement, la contrôlabilité et la topologie. Un cadre basé sur la commande prédictive (MPC) est proposé pour analyser le microrgrid dans divers scenarios de défaillance. Ensuite, un cadre multi-perspectif est développé pour analyser les CIs considérant le service d'approvisionnement, la contrôlabilité et la topologie. Ce cadre permet d'identifier le rôle des éléments de CIs et de quantifier les conséquences de scénarios de défaillances, par rapport aux différents perspectives considérées. Afin de présenter le cadre d'analyse, un réseau de transport de gaz réel à travers plusieurs pays de l'Union européenne est considéré comme une étude de cas. En fin, un cadre d'optimisation a trois objectifs est proposé pour la conception de CI : la topologie du réseau et l'allocation des capacités de liaison sont optimisées minimisant la demande non fournie et la complexité structurelle du système, et en même temps maximisant la contrôlabilité du système. Une investigation approfondie sur les multiples objectifs considérés est effectuée pour tirer des informations utiles pour la conception du système. Les résultats de cette thèse démontrent l'importance de développer du cadre d'analyse des CIs considérant de plusieurs perspectives pertinentes pour la conception, l'opération et la protection des CIs. / Critical infrastructures (CIs) provide essential goods and service for modern society. Their safety and reliability are primary concerns. The complexity of CIs calls for approaches of system analysis capable of viewing the problem from multiple perspectives. The focus of the present thesis is on the integration of the control perspective into the safety and reliability analysis of CIs. The integration is first approached by investigating the control properties of a small network system, i.e., an electric power microgrid. A simulation-based scheme is developed for the analysis from different perspectives: supply service, controllability and topology. An optimization-based model predictive control framework is proposed to analyze the microgrid under various failure scenarios. Then, a multi-perspective framework is developed to analyze CIs with respect to supply service, controllability and topology. This framework enables identifying the role of the CI elements and quantifying the consequences of scenarios of multiple failures, with respect to the different perspectives considered. To demonstrate the analysis framework, a benchmark network representative of a real gas transmission network across several countries of the European Union (EU) is considered as case study. At last, a multi-objective optimization framework is proposed for complex CIs design: design of network topology and allocation of link capacities are performed in an optimal way to minimize the non-supplied demand and the structural complexity of the system, while at the same time to maximize the system controllability. Investigation on the multiple objectives considered is performed to retrieve useful insights for system design. The findings of this thesis demonstrate the importance of developing frameworks of analysis of CIs that allow considering different perspectives relevant for CIs design, operation and protection.

Dans ce travail, nous étudions l’évaluation de la fiabilité, la modélisation et l’optimisation de systèmes à états arbitraires à incertitude épistémique. Tout d'abord, une approche universelle de modélisation à l'état arbitraire est proposée afin d'étudier efficacement les systèmes industriels modernes aux structures, mécanismes de fonctionnement et exigences de fiabilité de plus en plus complexes. De simples implémentations de modèles de fiabilité binaires, continus ou multi-états traditionnels ont montré leurs lacunes en termes de manque de généralité lors de la modélisation de structures, systèmes, réseaux et systèmes de systèmes industriels modernes et complexes. Dans ce travail, nous intéressons aussi particulièrement aux systèmes monotones, non seulement parce que la monotonie est apparue couramment dans la plupart des modèles de fiabilité standard, mais aussi qu’une propriété mathématique aussi simple permet une simplification énorme de nombreux problèmes extrêmement complexes. Ensuite, pour les systèmes de fiabilité monotones à états arbitraires, nous essayons de résoudre les problèmes suivants, qui sont apparus dans les principes mêmes de la modélisation mathématique: 1. L’évaluation de la fiabilité dans un environnement incertain épistémique avec des structures hiérarchiques être exploitées par toute approche de programmation 2; l'optimisation de la fiabilité / maintenance pour les systèmes à grande fiabilité avec incertitude épistémique. / In this work, we study the reliability assessment, modeling and optimization of arbitrary-state systems with epistemic uncertainty. Firstly, a universal arbitrary-state modelling approach is proposed, in order to effectively study the modern industrial systems with increasingly complicated structures, operation mechanisms and reliability demands. Simple implementations of traditional binary, continuous or multi-state reliability models have been showing their deficiencies in lack of generality, when modelling such complex modern industrial structures, systems, networks and systems-of-systems. In this work, we are also particularly interested in monotone systems, not only because monotonicity commonly appeared in most of the standard reliability models, but also that such a simple mathematical property allows a huge simplification to many extremely complex problems. Then, for the arbitrary-state monotone reliability systems, we try to solve the following challenges that appeared in its very fundamentals of mathematical modeling: 1. The reliability assessment under epistemic uncertain environment with hierarchy structures; 2. The reliability/maintenance optimization for large reliability systems under epistemic uncertainty.