Two contrasting forms of advice for decision-makers are to either follow one’s heart (emotions) or one’s head (reason). This is a false dichotomy – but how should decision-makers combine heart and head? Decisions can be fruitfully analysed as a set of components: a decision-problem embedded within an on-going situation, with values-at-stake, possible options-with-consequences, choice, action and review. Structured decision-making models (head theories) approach this multifaceted nature of decisions by a divide-and-conquer strategy with thinking tasks provided to help decision-makers clarify the decision-problem, identify important values-at-stake, find credible options, choose the most credible option, act effectively and fairly review the outcomes of the choice. Emotions are complex and can also fruitfully be analysed as a set of components: an appraisal of a situation’s implication for the actor’s goals and values, bodily and cognitive changes, phenomenological experience and desires. Emotions can both help and hinder decision making, so wise decision-makers should neither ignore nor rely upon emotions, but instead treat emotions as fallible resources. The complex nature of emotions implies that different emotion-enabled tasks might assist decision-makers for different components of the decision. On the basis of this analysis an emotion-enabled, structured, decision-making procedure was developed and investigated by taking ten participants with decision dilemmas through the procedure. This investigation, based on repeated use of the Hermeneutic Single-Case Efficacy Design, provided some initial support for the effectiveness of the model: participants found the procedure generally helpful (p < 0.005), had increased confidence in their final choice (p < 0.005), which at follow-up they were satisfied with (p < 0.005). The use of emotions as fallible resources was also investigated through tracing emotion-enabled changes in participants’ decision making. Suggestions for further development and investigation of integrating emotions into structured approaches are offered.

Nous cherchons à équiper un processus d’aide à la décision d’outils permettantde répondre aux exigences de redevabilité. Un décideur fournit de l’information quant à ses préférences au sujet de la façon d’arbitrer entre des points de vue conflictuels. Un analyste, chargé d’éclairer la prise de décision, fait l’hypothèse d’un modèle de raisonnement, et l’ajuste aux informations fournies par le décideur. Nous faisons l’hypothèse d’un processus d’élicitation robuste, dont les recommandations sont déduites des éléments dialectiques. Nous nous sommes donc intéressés à la résolution d’un problème inverse concernant le modèle, ainsi qu’à la production d’explications, si possible correctes, complètes, facile à calculer et à comprendre. Nous avons considéré deux formes de représentation du raisonnement: l’une ayant trait à la comparaison de paires d’alternatives fondée sur un modèle de valeur additive, l’autre ayant trait au tri des alternatives dans des catégories ordonnées fondé sur un raisonnement non-compensatoire. / We consider providing a decision aiding process with tools aiming at complying to the demands of accountability. Decision makers, seeking support, provide preference information in the form of reference cases, that illustrates their views on the way of taking into account conflicting points of view. The analyst, who provides the support, assumes a generic representation of the reasoning with preferences, and fits the aggregation procedure to the preference information. We assume a robust elicitation process, where the recommendations stemming from the fitted procedure can be deduced from dialectical elements. Therefore, we are interested in solving an inverse problem concerning the model, and in deriving explanations, if possible sound, complete, easy to compute and to understand. We address two distinct forms of reasoning: one aimed at comparing pairs of alternatives with an additive value model, the other aimed at sorting alternatives into ordered categories with a noncompensatory model.

Les travaux de recherche menés dans cette thèse s’inscrivent dans deux champs disciplinaires que sont l’évaluation du confort et l’aide multicritère à la décision.L’objectif de la thèse est de construire un modèle pour évaluer des trains sur le point de vue du confort tel qu’il est perçu par les voyageurs. L’approche utilisée pour cela repose sur trois étapes principales de construction d’un modèle d’aide multicritère à la décision. La première consiste à définir et à formaliser les critères de confort du problème. Dans la deuxième étape, il s’agit de construire les échelles afin de pouvoir évaluer les trains sur chaque critère de confort considéré.La troisième étape consiste à agréger les critères de confort en utilisant des méthodes d’agrégation multicritère. Cette étape nécessite l’élicitation des préférences des décideurs afin de mettre en oeuvre les méthodes d’agrégation.Notre contribution est de formaliser une approche pour la construction d’un modèle d’évaluation du confort dans les trains. Cette approche peut être appliquée à d’autres problématiques que l’évaluation du confort. Elle présente deux particularités principales. La première est d’intégrer dans la construction du modèle des facteurs importants traduisant la perception du confort. Nous avons choisi pour cela un modèle hiérarchique comportant plusieurs niveaux. La deuxième particularité de l’approche est d’utiliser des méthodes d’agrégation pouvant être différentes d’un noeud à un autre du modèle. Elle présente également d’autres aspects plus spécifiques, notamment lors de l’élicitation des préférences où nous construisons des exemples d’apprentissage informatifs pour accélérer le processus d’élicitation / This PhD thesis falls within two scientific areas, which are comfort evaluation and multiple criteria decision aiding. The main purpose is to develop a model in order to evaluate trains on the comfort point of view, as percieved by passengers. The developed approach is based on three main steps of developing a multiple criteria decision aiding model. The first one consists on defining and formalizing the criteria of comfort. In the second step, the scales of each considered criterion must be built in order to evaluate the trains on these last. The third step aims at aggregating the criteria, using multiple criteria aggregation methods, in order to obtain an overall comfort evaluation of trains. For this purpose, the decision maker’s preferences must be elicited

Assessment of the myriad of historic attempts to manage and/or restore degraded drylands offers a rich opportunity to learn from the past, particularly if conducted with full stakeholder engagement. Participatory environmental assessment of past land management and restoration actions would contribute to the improvement of future management techniques in a way relevant to the concerns of people involved with or impacted by these actions. This can also help to deal with the often scant information available, conflicting values and perceptions among stakeholders, and the uncertainties inherent to complex dryland systems. In this study I applied and evaluated a participatory protocol that incorporated multi-criteria decision analysis (MCDA) tools to assess five actions in the San Simon watershed, one of the most extreme examples of degradation and human intervention in southeastern Arizona (U.S.). The participatory assessment process included a semi-structured interview, elicitation of local-based assessment criteria, prioritization of the assessment criteria, estimation of data, a MCDA-based integration and group evaluation of final results. The process was used to evaluate five combinations of grazing management, vegetation management and hydraulic structures implemented between the 1940s and 1980s. The application of this process allowed me to not only evaluate these actions in a participatory way, but also to identify and compare values and perceptions connected to the historic, cultural and scientific narratives used by three different categories of stakeholders (researchers, practitioners and land users). The revised Simos' procedure used to elicit assessment criteria weights proved useful to expose values and perceptions, source of the individual criteria priorities, while revealing conflictive points of views among the stakeholders. The outranking-facilitated participatory assessment, when compared to the unaided baseline assessment, proved useful in making stakeholder preferences explicit in the form of evaluation criteria and weights, while incorporating data and uncertainty. The specific MCDA outranking integration model used, ELECTRE IS, proved to be simple and systematically synthetic, helping stakeholders structure and re-evaluate their unaided assessments. The results of this study provide insights in how stakeholders' knowledge and views can be elicited, explored and effectively incorporated to assess and learn from past land management and restoration actions implemented in drylands.

Cette thèse s'appuie sur deux axes. L'un appliqué traite de la création d'un indicateur dont le but est d'évaluer la gravité attendue des conséquences d'un scénario de pollution accidentelle. J'ai choisi d'utiliser des outils méthodologiques appartenant au domaine de l'aide multi-critères à la décision pour traiter ce premier sujet. Ce problème impliquant plusieurs disciplines scientifiques, j'ai choisi de le diviser en plusieurs sous-problèmes à travers une arborescence de critères. J'ai également impliqué plusieurs experts, notamment en toxicologie et en écologie afin de mieux prendre en compte les aspects liés à ces deux disciplines dans la création de cet indicateur.L'étude des méthodes de tri multicritère effectuée lors des recherches sur le premier axe m'a amené à en proposer une nouvelle que j'ai nommé algorithme du Dominance Based Monte Carlo (DBMC). Cet algorithme a comme particularités de n'être pas fondé sur un modèle et de fonctionner de manière stochastique. Nous avons étudié ses propriétés théoriques, en particulier nous avons démontré qu'en dépit de sa nature stochastique, le résultat de l'algorithme Dominance Based Monte Carlo converge presque sûrement. Nous avons également étudié son comportement et ses performances pratiques à travers un test nommé k-fold cross validation et les avons comparés aux performances d'autres algorithmes d'élicitation des préférences pour le tri multi-critères. / This thesis is based on two main axes. The first one deals with the creation of an indicator that aims at evaluating the expected severity of the consequences of a scenario of accidental pollution. In order to create this methodology ofevaluation, I chose to use methodological tools from multi-criteria decision aiding. So as to deal with the complexity of this problem, i decided to split it into several sub-problems using a hierarchy of criteria, being mainly inspired by the "value focused thinking approach". In this work, I interacted with several experts in toxicology and in ecology in order to betterdeal with every aspect of this problem.While studying several elicitation methods for the multi-criteria sorting problem, I proposed a new one that I named Dominance Based Monte Carlo algorithm (DBMC), which brings me to the secons axis of this thesis. This elicitation algorithm has two main specificities: being model free and a stochastic functionning. In this thesis, we study its theoretical properties. In particular, we prove that despite its stochastic nature, the result of the Dominance Based Monte Carlo algorithm converges almost surely. We also study its practical performances through a test named k-fold validation and we compared these performances to those of other elicitation algorithms for the sorting problem.

Sous nos latitudes, l’usage des bâtiments existants et les consommations énergétiques associées (chauffage, climatisation, ventilation, eau chaude sanitaire, éclairage et autres usages) sont responsables d’impacts considérables sur l’Environnement. De plus, le renouvellement du parc existant étant inférieur à 1% par an, dans la plupart des pays développés, la réhabilitation des bâtiments constitue un levier majeur de réduction des consommations d’énergie et des émissions de gaz à effet de serre. Cependant, l’identification de stratégies optimales de réhabilitation énergétique, incluant la planification des actions dans le temps, demeure une problématique complexe pour les acteurs de la Construction. Ces travaux de thèse visent à produire des connaissances afin de contribuer à l’aide à la décision pour l’identification de programmes efficaces de réhabilitation énergétique, à partir de méthodes d’optimisation multicritères. Les solutions (programmes séquentiels de réhabilitation énergétique) sont optimisées en termes de composition et de phasage. La composition est définie par la combinaison de mesures de réhabilitation mise en oeuvre. Celles-ci concernent l’enveloppe des bâtiments (isolation thermique, remplacement des fenêtres, surfaces de fenêtres) et le remplacement des équipements de chauffage, ventilation et production d’ECS. Pour chacune des mesures, plusieurs alternatives sont envisagées. Le phasage correspond à la permutation de ces mesures, définissant la séquence de mise en oeuvre. Les solutions sont évaluées sur une base multicritère et sur le cycle de vie. Les fonctions objectifs ciblent les impacts environnementaux de l’ACV (Analyse de Cycle de Vie), des indicateurs économiques, le bien-être des occupants par le confort thermique adaptatif en été. Des modèles d’ACV et d’analyse du coût du cycle de vie, utilisant la simulation thermique dynamique pour le calcul des besoins de chauffage et des températures intérieures, ont été développés pour l’évaluation des performances des solutions. Etant donnée la nature mathématique du problème (multicritère, combinatoire, à variables discrètes et à fonctions objectifs implicites non-linéaires), deux méthodes d’optimisation multicritères sont étudiées : les algorithmes génétiques (NSGA II) et la programmation dynamique. Dans l’approche génétique, la modélisation des solutions, sous la forme d’un couple de chromosomes, permet d’identifier des programmes séquentiels efficaces de réhabilitation énergétique et d’analyser les surfaces de compromis, en termes de définition et performances des solutions, de compromis entre les critères de décision. A partir de la représentation du problème par un graphe séquentiel, la programmation dynamique permet alors de comparer les solutions approchées issues de l’algorithme génétique, ou d’approches court-termistes, au front de Pareto exact. L’optimisation exacte a également été exploitée pour analyser la sensibilité des solutions à différents paramètres de modélisation dont le comportement des occupants, l’évolution des prix de l’énergie, la durée de vie des composants de réhabilitation. Les contraintes budgétaires s’appliquant au projet de réhabilitation ont été ensuite intégrées dans un algorithme génétique multicritère sous contraintes, adapté à l’étude des stratégies de réhabilitation sous la contrainte d’un plan de financement. Enfin, l’approche génétique a été étendue depuis l’échelle du bâtiment à celle du par cet l’optimisation exacte a été utilisée pour caractériser les typologies de bâtiment en réhabilitation. L’intérêt des différentes méthodes est illustré sur une étude de cas (…) / Under our latitudes, existing buildings energy consumptions, related to heating, cooling, ventilation, domestic hot water (DHW), lighting and other uses, are responsible for significant environmental burdens. Moreover, existing buildings annual replacement rate being lower than 1%, in most developed countries, existing stock retrofit represents a major lever to reach commitments on climate change and non-renewable energy consumption mitigation. However, the identification of optimal sustainable retrofit programs, including actions planning over a time period, is still a difficult task for professionals.This thesis aims at producing knowledge in order to contribute to decision support for efficient energy retrofit programs identification, through the application of different multi-criteria optimization techniques. The solutions (sequential building energy retrofit programs) are optimized both on their content and planning. The content refers to the combination of retrofit measures considered. These address holistically building envelopes (thermal insulation, windows replacement, window to wall ratios), and the replacement of equipment for ventilation, heating and DHW production. For each of these measures, various options are considered. The planning corresponds to the permutation of these measures, defining a time sequence for implementation. The solutions are evaluated on a multi-criteria and life cycle basis. The objective functions considered target environmental impacts evaluated using LCA (Life Cycle Assessment), some financial indicators and occupants' well-being through thermal comfort in summer. Life cycle assessment and life cycle cost models, using building dynamic thermal simulation for heating load and thermal comfort evaluation, are implemented to assess solutions performances.Considering the problem mathematical nature (multi-criteria, combinatorial, discrete variables, implicit non-linear objective functions), two suitable multi-criteria optimization techniques have been studied: multi-criteria genetic algorithm (NSGA-II) and dynamic programming. In the genetic approach, the modelling of each solution by a pair of chromosomes allowed to identify efficient sequential energy retrofit programs and analyse Pareto compromise surfaces, in terms of solutions features, performances and relationships in between criteria. Then, the representation of the problem on a sequential graph enabled us to apply dynamic programming, to compare both the genetic approximate solutions, and the results of some short- term approaches to the exact Pareto frontier. The search for exact solutions also been exploited to perform sensitivity analysis on different modelling parameters such as heating temperature setting, energy prices evolution or materials lifespan. Real life budget constraints have been incorporated to build a constrained multi-criteria genetic optimisation method, suitable to study retrofit strategies under financing plans. At the end, the genetic approach has been extended from building scale to stock scale and exact optimization has been used to characterize building types in terms of energy retrofit.The benefits of these methods have been illustrated on case studies. Knowledge has been produced in terms of multi criteria optimization methodology, applied to sequential energy retrofit, and understanding of building stocks evolution. These developments contribute to decision aiding; providing decision makers with efficient energy retrofit strategies and a description of the comprise surface, at the building or building stock scale, on a multi- criteria basis, over life cycle

La prise en compte accrue de l’environnement et des nuisances dues à l’usage massif de l’automobile influence les politiques de mobilité des agglomérations françaises. De plus en plus, celles-ci se dotent d’un réseau de transport en commun en site propre ; l’idée est de rendre les transports collectifs plus attractifs pour les usagers et donc plus compétitifs par rapport à la voiture particulière. Pour toutes ces villes se pose alors la question de l’insertion de ces lignes de transport en commun à deux échelles spatiales : au niveau de la ville avec la détermination du tracé de la ligne de transport et au niveau de la voirie avec la position du site propre au sein de l’espace public et la répartition des autres usagers au sein de cet espace. Les projets de transport suscitent de l’intérêt de la part de la population locale et nombreux sont les acteurs qui souhaitent y participer. La multiplicité des acteurs concernés par le projet, et surtout la différence de points de vue qu’ils défendent, peuvent alors engendrer des blocages dans le déroulement du projet, allant parfois jusqu’à sa suspension. L’objectif de notre recherche est d’élaborer un processus permettant de déterminer d'une part le tracé de la ligne de transport et d'autre part l’implantation de la plateforme de transport et la répartition des usagers au sein de l’espace public. Ce processus doit prendre en compte les aspects liés au transport, ceux liés à l’urbanisme de l’agglomération ainsi que les points de vue des différents acteurs. La méthodologie alors développée repose sur l’utilisation de la méthode d’aide multicritère à la décision ELECTRE III comme support de la concertation. À partir de l’analyse historique des projets de transport qui se sont déroulés en France, les critères permettant de choisir l’implantation du tracé de la ligne et la répartition des différents usagers de l’espace public ont été déterminés. Dès lors, l’utilisation d’ELECTRE III constitue un support de débats avec les différents acteurs pour mettre en exergue les éléments du projet sur lesquels les points de vue divergent. Pour animer ces débats, un nouvel acteur, le facilitateur, est introduit au sein des projets de transport. À partir des recommandations issues de la phase de concertation, les élus, véritables décideurs du projet, peuvent prendre les mesures qui s’imposent, notamment en termes de communication, afin que le projet subisse le moins de retard possible et aboutisse à une variante acceptée par le plus grand nombre d’acteurs / Taking into account the environmental aspects and the nuisances due to the massive use of private cars use has an influence on the French agglomerations mobility policies. More and more, these agglomerations develop their own public transport network; the idea is to make public transport more attractive for the users than today and therefore more competitive than private cars. For all cities, the question of the public transport lines implementation is posed at two levels: that of the city to determinate the public transport layout and that of the street in order to choose the position of the transport platform and the distribution of users in the public space. Public transport projects arouse the interest from the local population, and a lot of stakeholders are thus encouraged to participate in them. The multiplicity of stakeholders concerned by the project, and especially the different points of view they defend, can also block the project progress, even leading to its end. Our research is aimed to develop a process to determine, on the one hand, the public transport line layout and on the other, the platform implementation and the public space users distribution. This process has to take into account the aspects linked to transport, those connected to the agglomeration urban design and the different stakeholders’ points of view. The methodology we have created is based on the use of ELECTRE III, a multicriteria decision aiding tool, as a support for "concertation"¹. From the historical analysis of the French public transport projects, the criteria used to choose the line layout and the public space users distribution have been determined. Then, the use of ELECTRE III is a support for the debates with the various stakeholders, to highlight the parts of the projects that are problematic because the different viewpoints are divergent. To conduct the debates, a new actor, the facilitator, is introduced in the transport project. From the recommendations stemmed from the concertation stage, the elected representatives, the real project decision makers, can take the necessary steps, notably in terms of communication, so that the project suffers from as fewer delays as possible and attains the acceptance of most stakeholders ¹ The French word “concertation” does not have any real equivalent in English. It designates a participative decision-making process in which the various stakeholders are able to express their point of view and have it taken into account in the decision-making process

Submitted by Maria Medeiros (maria.dilva1@ufcg.edu.br) on 2018-09-11T12:41:27Z No. of bitstreams: 1 VICTOR VIDAL NEGREIROS BEZERRA - DISSERTAÇÃO (PROFIAP-CCJS) 2016.pdf: 3966281 bytes, checksum: 545852b43a64e3cdd977c00aa9db9066 (MD5) / Made available in DSpace on 2018-09-11T12:41:27Z (GMT). No. of bitstreams: 1 VICTOR VIDAL NEGREIROS BEZERRA - DISSERTAÇÃO (PROFIAP-CCJS) 2016.pdf: 3966281 bytes, checksum: 545852b43a64e3cdd977c00aa9db9066 (MD5) Previous issue date: 2016-04-19 / A avaliação de desempenho no contexto municipal tem se apresentado como uma importante ferramenta para o desenvolvimento e melhoria da gestão pública nesse nível federativo. Apesar disso, as avaliações do desempenho municipal têm se fundamentado em metodologias onde somente poucos critérios avaliativos são considerados e processados. Nesse sentido, o presente estudo buscou cooperar no desenvolvimento e aprimoramento de avalições de desempenho municipal através das técnicas multicritério de apoio a decisão. Tomando como base os indicadores de gasto per capita nas 28 funções de governo, este estudo teve como objetivo aplicar uma metodologia multicritério de avaliação de desempenho para medir a performance de 12 municípios do estado da Paraíba. Através da aplicação do método multicritério PROMETHEE II, foi possível identificar a aplicabilidade desse tipo de metodologia no contexto da avaliação de desempenho de municípios. A análise estatística dos dados obtidos para os anos de 2014 e 2015 mostram que algumas áreas de atuação governamental, a exemplo de Saneamento, Segurança Pública e Gestão Ambiental tem recebido menor ênfase na aplicação de recursos públicos, enquanto que áreas como Administração, Saúde e Educação tem recebido maior destaque na alocação desses recursos. Já na aplicação do método multicritério PROMETHEE II, foi possível identificar rankings de desempenho capazes de verificar e comparar de forma global a performance dos municípios estudados. / The performance measurement in the municipal context has emerged as an important tool for the development and improvement of public administration in a federal level. Nevertheless, assessments of municipal performance have been based on methodologies where only a few evaluative criteria are considered and processed. In this sense, this study sought to cooperate in the development and improvement of municipal performance measurement through Multi-Criteria Decision Aiding Techniques. Based on the per capita spending indicators in 28 functions of government, this study aimed to apply a multi-criteria methodology of performance evaluation to measure the performance of 12 municipalities in the state of Paraiba. By applying the multi-criteria method PROMETHEE II, it was possible to identify the applicability of this type of methodology in assessing the performance of municipalities. The statistical analysis of the data for the years 2014 and 2015 show that some areas of government action, such as Sanitation, Public Safety and Environmental Management have received less emphasis in the use of public funds, while areas such as administration, health and education have received the most attention in the allocation of these resources. In the application of the multi-criteria method PROMETHEE II, it was possible to identify performance rankings that were able to check and compare globally the performance of the studied municipalities.

Submitted by Marcia Silva (marcia@latec.uff.br) on 2016-10-07T19:57:13Z No. of bitstreams: 1 Disssert Leandro Peçanha de Souza.pdf: 11466487 bytes, checksum: 099df51af11e59887ad31dab009228a8 (MD5) / Approved for entry into archive by Biblioteca da Escola de Engenharia (bee@ndc.uff.br) on 2017-01-26T12:26:11Z (GMT) No. of bitstreams: 1 Disssert Leandro Peçanha de Souza.pdf: 11466487 bytes, checksum: 099df51af11e59887ad31dab009228a8 (MD5) / Made available in DSpace on 2017-01-26T12:26:11Z (GMT). No. of bitstreams: 1 Disssert Leandro Peçanha de Souza.pdf: 11466487 bytes, checksum: 099df51af11e59887ad31dab009228a8 (MD5) Previous issue date: 2016-04-05 / A alta concorrência e a necessidade crescente de melhores resultados com recursos cada vez mais reduzidos impõem que as empresas selecionem e invistam em projetos que tenham maior potencial para gerar vantagem competitiva. O Comitê de TI (Tecnologia da Informação) é uma das estruturas de Governança de TI mais elementares e utilizadas nas organizações para potencializar o desempenho da TI, sendo que uma das suas principais atribuições consiste em priorizar os projetos de TI. Definir com clareza um portfólio alinhado aos objetivos estratégicos, não levando em conta somente fatores financeiros ou exclusivamente técnicos, é um problema de decisão multicritério complexo. Este estudo pretende demonstrar a utilização de uma metodologia de Apoio Multicritério à Decisão (AMD) através da aplicação de um método híbrido composto pelo AHP (Analytic Hierarchy Process) e TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution), que permitirá ordenar (ranking) os projetos de investimento do portfólio da TI / The high competition and growing need for the best results with increasingly limited resources require companies to select and invest in projects that have the greatest potential to create a competitive advantage. The IT Committee (Information Technology) is one of the most basic IT Governance structures and is used in organizations to enhance IT performance, since one of its main attributes is prioritizing IT projects. Clearly defining a portfolio aligned to strategic objectives, taking into account not only financial or purely technical factors, is a complex multi-criteria problem for decision-making. This study aims to demonstrate the use of a Multi-criteria Decision Aid (MCDA) methodology, by applying a hybrid method composed of the AHP (Analytic Hierarchy Process) and TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution), which will allow for sorting (ranking) IT portfolio investment projects

Depuis le 19ème siècle, de nombreux dispositifs de protection ont été implantés en montagne pour réduire le risque torrentiel. L'évaluation multi-échelle de leur efficacité est nécessaire pour comparer et choisir les stratégies de maintenance et de gestion de l’occupation des sols. Le chapitre I met en évidence les ambiguïtés du concept d’efficacité évalué selon plusieurs critères dans un contexte d’information imparfaite, issue de sources multiples, inégalement fiables. Pour aider l’évaluation multidimensionnelle, le problème de décision est décomposé en plusieurs sous-problèmes formulés, modélisés, traités puis intégrés. L’état de l’art du chapitre II permet de retenir les méthodes de la sûreté de fonctionnement (SdF) et d'aide à la décision (AD) basées sur le Raisonnement Évidentiel, associant les ensembles flous, les possibilités, les fonctions de croyance et la fusion d’information pour traiter chaque problème. Le chapitre III explicite le concept d'efficacité puis propose une adaptation des méthodes de la SdF à la formulation multi-échelle des problèmes. Le chapitre IV adapte les méthodes d'AD basées sur le Raisonnement Évidentiel, les améliore pour mieux représenter la connaissance disponible et évaluer la qualité d’une décision, et les associe. Le chapitre V propose une méthodologie séquentielle qui intègre ces méthodes pour évaluer les mesures de protection, de l’échelle de l’ouvrage à celle du bassin d’enjeux. Enfin, le chapitre VI discute les résultats, identifie les apports respectifs de l'intégration et des nouvelles méthodes, puis décrit les perspectives de déploiement opérationnel de la méthodologie développée. / Many torrential protective check dams have been built in French mountainous areas since the 19th century. They aim to reduce risk in torrential watersheds. Analysis of their effectiveness is needed to assess residual risk and to choose the best maintenance strategy in a context of reduced budgets. Chapter I introduces main issues due both to definition of effectiveness concept and also the several criteria and system scales that have to be assessed. Moreover, required expertise is based on imperfect information, provided by several sources with different reliability levels. To help such decisions, this thesis proposes to break down general assessment into several smaller decision-making problems. In chapter II, dependability analysis tools and Evidential Reasoning based-decision-aiding methods (ER-DAMs), using fuzzy sets and possibility theories, are chosen to help individual decisions. Chapters III to V then introduce new developments. Effectiveness concept is first defined. A multi-scale dependability analysis helps to describe each decision-making problem. ER-DAMs are then adapted to help effectiveness assessment at each system’s scale. Technical improvements are provided to these methods for a better knowledge representation and final decision. Improved methods are then associated. Finally, individual problems and associated methods are integrated in a generic methodology to help torrential protective measures’ effectiveness assessment at watershed scale. The last chapter VI focuses on discussion of integrative aspects and of ER-DAMs’ specific developments. It gives some elements to improve methods but also to help operational implementation of the generic methodology.