Дисертації з теми "Prise de Décision Multicritère"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Prise de Décision Multicritère".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Glaize, Annabelle. "Prise de décision en santé : une approche de décision multicritère." Thesis, Lille 1, 2019. http://www.theses.fr/2019LIL1A006.
Повний текст джерелаDecisions in healthcare are often complex and difficult to make and justify. Multi-criteria decision analysis (MCDA) is a decision-making tool that has been proven to be useful in numerous applications in healthcare contexts. Specifically, this operations research tool enables the integration of multiple conflicting criteria and encourages stakeholders to participate in the decision-making process. The purpose of this PhD dissertation is to contribute to the scientific literature on MCDA methods and how they should be applied in healthcare contexts, which are characterised by complex decision-making, by expanding these methods’ possible applications. This research is composed of three essays, each of which answers a specific research question related to decision-making in healthcare. The first essay maps the literature and assesses how the steps of the MCDA process are followed in different healthcare contexts. The second essay combines the business process improvement (BPI) methodology and lean methods to assess a chemotherapy outpatient service that suffers from difficulties in the patient flow process and propose improvement opportunities. The third essay builds on the findings of the second and applies the ELECTRE III method to define which actions could help improve the quality of care and patient satisfaction of the outpatient service
Almeida, Dias Juscelino. "Multiple criteria decision aiding for sorting problems : concepts, methodologies, and applications." Paris 9, 2011. https://bu.dauphine.psl.eu/fileviewer/index.php?doc=2011PA090003.
Повний текст джерелаOur thesis deals with sorting problems following a constructive approach. The aim is to assign objects of a decision, evaluated on multiple criteria, to a set of categories. Each category is pre-defined to receive these objects, which will be or might be processed in the same way. Our research provides a taxonomy framework, in which ten different types of sorting problems is defined with a practical usefulness. Two sorting methods are proposed within this taxonomy, called ELECTRE TRI-C and ELECTRE TRI-NC. These methods deal with decision aiding contexts, where the set of categories is completely ordered. Each category is characterized by a single typical reference action and several ones, respectively. The assignment procedure is based on a descending rule and an ascending rule, which must be used conjointly. Our research also provides a segmenting description algorithm for analyzing the relationship between the assigning conditions of decision aiding sorting models (e. G. ELECTRE TRI-C) and the preferences of decision makers, including an analysis of incoherencies and incompatibilities, without making use of an optimization model. The main research results are validated by two real-world applications (assisted reproduction and agro-environmental risk), which are modeled with medical and environmental experts, respectively
A nossa investigação centra-se em problemas de classificação seguindo uma abordagem construtivista. O objectivo é afectar objectos de decisão, avaliados com critérios múltiplos, em categorias. Cada categoria é pré-definida para agrupar estes objectos, que irão ser ou deverão ser sujeitos ao mesmo tratamento. A nossa pesquisa fornece uma taxonomia, onde definimos dez tipos de problemas de classificação com interesse prático. Dois métodos de classificação são propostos no âmbito desta taxonomia, denominados ELECTRE TRI-C e ELECTRE TRI-NC. Estes métodos enquadram-se em contextos de apoio à decisão, onde as categorias são completamente ordenadas. Cada categoria é caracterizada, respectivamente, por uma acção de referência típica ou múltiplas acções típicas. O procedimento de afectação é composto por uma regra descendente e uma regra ascendente, que devem ser usadas conjuntamente. A nossa pesquisa fornece também um algoritmo de descrição segmentada para analisar a relação entre as condições de afectação de modelos de classificação em apoio à decisão (e. G. ELECTRE TRI-C) e as preferências dos decisores, incluindo uma análise de incoerências e incompatibilidades, sem auxílio de um modelo de optimização. Os principais resultados da investigação são validados por duas aplicações reais (reprodução assistida e riscos agro-ambientais), que foram modelizadas, respectivamente, com especialistas médicas e agro-ambientais
Dubus, Jean-Philippe. "Prise de décision multiattribut avec le modèle GAI." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00812558.
Повний текст джерелаBouchet, Philippe. "Approche axiomatique en décision multicritère : cas des variables mixtes." Université Joseph Fourier (Grenoble ; 1971-2015), 1994. http://www.theses.fr/1994GRE10105.
Повний текст джерелаAldiab, Ali. "Contribution à la définition d'une approche concertée et multicritère du choix d'un ensemble fondation-système porte de bâtiment." Artois, 2004. http://www.theses.fr/2004ARTO0202.
Повний текст джерелаThe goal of our work is to propose a concerted and multicriterion approach to choose both carried system and foundations system of building. The created tool can help the client to work out his program and the architect to define the more adapted orientations of both carried system and foundations system. For its creation, we develop the following phases: problem analysis: the analysis of both thesis context and lack of decision-making tools underlines the need for a rational and progressive approach, composed of two levels: the elimination of solutions not satisfying the limits: technical, economic, time, etc. The multicriterion analysis for defining the best solution according to the particular project criteria. Information collection and analysis which allow: to describe the various project aspects and to define the data necessary to operation of choice mechanisms, to define the feasibility limits and the selection criteria, to define the solutions answers to these limits and these criteria. Approach test and validation: they make it possible to illustrate, on an example, the approach contributions in quality and cost terms
Benabbou, Loubna. "Contributions à la classificaton supervisée multi-classes et multicritère en aide à la décision." Doctoral thesis, Université Laval, 2009. http://hdl.handle.net/20.500.11794/20646.
Повний текст джерелаQureshi, Aiman Mazhar. "Modélisation et aide à la décision multicritère du confort thermique en milieu urbain." Electronic Thesis or Diss., Amiens, 2022. http://www.theses.fr/2022AMIE0081.
Повний текст джерелаUrban areas are the prevalent places of residence for people and are vulnerable to exasperating weather conditions such as heat stress. Periods of heat waves are increasingly reoccurring in the current atmosphere, and they are known to pose a serious and major threat to the health of human beings all over the world. Urban heat islands and heat waves increase thermal risks in urban areas and the vulnerability of the urban population. The increase in the number of heat episodes in urban areas has become a significant concern due to its adverse effects on human health and economic activities. The objective of this work is to identify the sensitivity of thermal comfort and their action variables, the modeling of thermal stress using the most influential meteorological variables, the identification of risk factors and highlight the correlation of meteorological trends and influencing parameters, solutions for mitigating heat stress and mathematical support for decision-making. Several machine and deep learning techniques were used for the system dynamic modeling of the thermal comfort. Optimized results are obtained from the Gated Recurrent Unit (GRU) model which is used for the development of a web simulation tool allowing the inhabitants to evaluate their level of comfort according to the weather conditions. A heat vulnerability index map has been developed to indicate the vulnerability of occupants considering different aspects in a medium-sized city such as planning, green space, density, energy, quality air, water bodies and extreme heat events. The obtained results highlighted that poor air quality and heat events are interrelated, which draws the attention for decision-makers to intervene the additional measures in high-risk places. Field monitoring is carried out using sensors and a thermal camera to measure relevant variables and take action to minimize the effects of heat stress. In Last, multi-criteria decision-making methods were applied for the initial development of a decision support tool for the selection of urban heat resilience interventions that allows flexible, dynamic, and predictive use for designers and the users
Karnib, Ali. "Approche multicritère pour l'aide au choix d'une solution de réseau technique urbain : application au réseau d'assainissement pluvial." Artois, 1996. http://www.theses.fr/1996ARTOA001.
Повний текст джерелаAttouri, Emna. "Contribution dans l’industrialisation de la construction : Mise au point d’un outil d’aide à la décision appliqué au gros œuvre." Thesis, Ecole centrale de Lille, 2019. http://www.theses.fr/2019ECLI0016.
Повний текст джерелаOffsite Construction (OSC), Modern Methods of Construction (MMC), Offsite Manufacturing (OSM), and Offsite Production (OSP) are all terms cited as being possible solutions for addressing time, quality and cost concerns associated with traditional construction method. Successful implementation of these methods on a particular project requires systematic analysis and early decision making based on specific factors of the project. There is a lack of an efficient systematic approach that is able to match the changing needs, deal with the growing complexity of building projects and take into consideration recently developed innovations, technologies and regulations. In this respect these issues have been addressed in this research study. The objective of this dissertation is the development of an early- phase decision tool to support the general contractor in selecting construction methods. We propose a multi-level decision framework. The first level conducts a feasibility study and evaluates the applicability of offsite construction. The second level proposes an integrated approach for the decision-making problem that combines the Analytical Hierarchy Process (AHP) and the Preference Ranking Organization Method for Enrichment Evaluations (PROMETHEE). The combination of both approaches enables a careful evaluation of different construction methods and scenarios for the same project. The research methodology adopted is a mixed approach, combining quantitative and qualitative data collection methods, including three case studies, 10 semi-structured interviews, questionnaires with different construction stakeholders and focus groups with 9 construction engineers. The study first identifies the advantages and limitation of offsite construction. Second, it highlights the factors influencing decisions on the use of prework. Then, it presents and describes a computerized tool to aid project teams in the decision-making process. Finally, the decision-making tool was tested and with construction practitioners on three different case studies. They validate that the developed framework and tool are useful and effective in the decision-making process
Durand, Sylvain. "Sur quelques paradoxes en théorie du choix social et en décision multicritère." Phd thesis, Université Joseph Fourier (Grenoble), 2000. http://tel.archives-ouvertes.fr/tel-00006743.
Повний текст джерелаBeck, Amina. "Choix de projets pharmaceutiques : exemple d'application des méthodes Electre IV et Prométhée III et comparaison avec une méthode multicritère d'aide à la décision, méthode des toiles." Dijon, 1993. http://www.theses.fr/1993DIJOE019.
Повний текст джерелаBlanquart, Stéphanie. "Role de l'aide multicritère à la décision (AMCD) dans la promotion d'une agriculture durable : Application à la régulation des pesticides au sein du secteur horticole de la Région Provence-Côte d'Azur." Nice, 2006. http://www.theses.fr/2006NICE0042.
Повний текст джерелаNatural resources management suffers from multiple dysfunctions: ecological, social-economic, territorial, but also informational. Facing these dysfunctions, a recent literature is trying to adapt various means to go beyond them, with the sustainable development as a common purpose. As the first user of soil, water for irrigation, ecosystems and biological diversity, agriculture is directly concerned by this concept. Indeed, there is a need for considering both agriculture and environment within a sustainable agriculture. In that respect, this study is proposing a local process (LP) based on modern decision methods such as Multi-Criteria Decision-Aid (MCDA), in order to promote a sustainable agriculture at a local level. Applied to pesticide control within the horticultural sector and used to answer a request from the INRA (National Institute of Agronomic Research), the promotion of a sustainable horticulture needs adopting an integrated biological protection (IBP) that would use in priority biological protection against chemicals. The final purpose of this study is to help local authorities determine their own agricultural sector requirements and propose an adapted environmental policy. Results from implementing the AMCD-LP are showing an incentive based on informational management problems related to the adoption of new agricultural practices and the reality of a “technological lock-in”. In a local policy context, recommendations are proposing an experimental phase with local horticulturists. The latter are being sorted in three criteria groups (technical, behavioural and sustainability) allowing high-quality information broadcast within the area
Deparis, Stéphane. "Etude de l'effet du conflit multicritère sur l'expression des préférenes : Une approche empirique." Phd thesis, Ecole Centrale Paris, 2012. http://tel.archives-ouvertes.fr/tel-00740658.
Повний текст джерелаFathivand, Khalili Fatemeh. "Résolution des problèmes de classement multicritère : modélisation des préférences ; relation de surclassement et présentation des différentes méthodes." Dijon, 1994. http://www.theses.fr/1994DIJOE012.
Повний текст джерелаThis thesis consists of modelizing the preferences of decision makers in order to help them to apprehend as well as possible their problem of choice. It depends on the decision field : the objects, the criteria, the consequences and the weight which represent the relative importance of each criterion. This modelization is based on the preferences and the objectives of decision makers. It is materialized by the multicriterion methods of outranking which are developped from the concepts of real, quasi or pseudo criterion. The introduction of the concept of blurry subsets offer more security to this modelization whose objective is to obtain an arrangement of the objects from the "most appropriate" to the "least ones". We have evoked the problem of modelization of preferences by introducing a new multicriterion method of outranking which leads us to a complete preordering and a classification of the objects by a global criterion. The introduction of the concept of decisive points and the procedure of unifying the scales have also an important utility for treating the quantitative, qualitative, homogeneous and heterogeneous informations, also for accomplishing the behavior of decision makers and economical
Henriet, Laurent. "Systèmes d'évaluation et de classification multicritères pour l'aide à la décision : Construction de modèles et procédures d'affectation." Phd thesis, Université Paris Dauphine - Paris IX, 2000. http://tel.archives-ouvertes.fr/tel-00528799.
Повний текст джерелаServant, Mathieu. "Mécanismes de prise de décision dans des environnements conflictuels : approches comportementales, computationnelles et électrophysiologiques." Thesis, Aix-Marseille, 2015. http://www.theses.fr/2015AIXM4764.
Повний текст джерелаA perceptual decision is a deliberative process that aims to choose a categorical proposition or course of action from a set of alternatives on the basis of available sensory information. Models of perceptual decision-making assume that sensory information is accumulated to some threshold level, whence the decision terminates in a choice. The recent discovery of neural correlates of these theoretical predictions in the non-human primate brain has reinforced their validity. However, neurophysiological studies of perceptual decision-making mechanisms in humans are relatively scarce. This work aims at enhancing our understanding of the computations and neurophysiology underpinning such mechanisms in humans, through the study of decision-making contexts more complex than those used in monkey research
Damart, Sébastien. "Une étude de la contribution des outils d'aide à la décision aux démarches de concertation : Le cas des décisions publiques de transport." Phd thesis, Université Paris Dauphine - Paris IX, 2003. http://tel.archives-ouvertes.fr/tel-00421927.
Повний текст джерелаDe ce fait, les législateurs ont tenté, depuis plusieurs décennies en France, de préciser un contenu et des modalités de mise en œuvre d'une démarche de concertation. Néanmoins, de nombreux auteurs concluent de toute façon à un décalage persistant et croissant entre ce que préconisent les textes de loi et ce qu'est la pratique de la décision publique en matière d'environnement ou de questions liées aux transports. Les réflexions de ces auteurs les conduisent à relever également une évidente inadéquation à des démarches participatives, des outils actuellement mobilisés dans les processus de décision. Dans des contextes participatifs, il faut donc d'une part imaginer d'autres outils capables de supporter différents points de vue et différentes rationalités et d'autre part parvenir à en évaluer l'intérêt dans ce type de contexte. Notre thèse s'inscrit dans cette perspective.
L'objet de la thèse est en effet, d'identifier, dans le cadre des décisions publiques, la nature et la forme des outils d'aide à la décision qui peuvent instrumenter et rendre possible des démarches de concertation. De fait, le travail conduit à préciser ainsi la nature des liens et de la contribution des outils d'aide à la décision aux démarches de concertation.
Pour cela, nous construisons une grille de lecture des démarches de concertation fondée sur les concepts d'intégration et d'identification qui nous permet, dans une perspective d'abduction d'inférer des facteurs contextuels dont dépend la pertinence d'outils d'aide à la décision par rapport à une démarche de concertation.
Marleau, Donais Francis. "Intégrer le transport durable dans les processus décisionnels pour le réaménagement de rues : application de l’aide multicritère à la décision." Doctoral thesis, Université Laval, 2021. http://hdl.handle.net/20.500.11794/69240.
Повний текст джерелаThe numerous negative impacts linked to the democratization of cars in the last century led to rethink transportation systems and transportation planning practices. Transportation systems should be planned to safeguard long-term ecological vitality, provide basic accessibility and ensure equal access to transport services. This is sustainable transportation. Many actions and strategies were proposed in the last decades to implement sustainable transportation. Redesigning streets in favor of active and public transportation is one of these strategies. However, there is currently a gap between the plans, that adopt a sustainable transportation vision, and practices, that still realize unsustainable transportation projects. The different government levels require new decision aid tools to change their decision processes. Multicriteria decision aiding (MCDA) is one method that is more and more discussed in the transportation literature to integrate a holistic perspective to decision processes. In this regard, this thesis studies more specifically the use of MCDA to improve decision processes linked to street redesigns. First of all, a descriptive literature review was conducted to better understand the evaluation methods that are currently used in transportation decision processes. The two main evaluation methods in the field (cost-benefit analysis and MCDA) were examined according to their perceived strengths and weaknesses, to the different ways to combine them and to their abilities to integrate sustainable development principles in the decision processes. To take into account the different conceptualizations of decision-aiding, the results were analyzed according to four decision-aiding approaches based on the concept of communicative action from Habermas (objectivist, conformist, adjustive and reflexive). Subsequently, a method to develop a multicriteria spatial decision support system (MC-SDSS) to prioritize the streets to redesign as Complete Streets was proposed and applied in collaboration with the City of Quebec in Canada. Complete streets is a popular movement in North America for sustainable transportation to design " streets for everyone ". The MC-SDSS was developed during multidisciplinary group workshops (transportation, urban planning, environment, infrastructure, urban design and public participation) gathering Quebec City professionals using the MCDA method MACBETH. The development of the MC-SDSS was split in five phases: (1) structuring the problem, (2) constructing attractiveness scales, (3) deriving scaling constants, (4) validating the model and (5) producing priority maps. The process has been successful. Indeed, Quebec City has been using this MC-SDSS in the elaboration of its Complete Streets strategy since 2017. Following the MC-SDSS developed in Quebec City, a post-project evaluation framework, specific to MCDA, was generated and applied to the MC-SDSS. The evaluation framework is based on five questions: why evaluate?, what to evaluate?, on what is the evaluation based?, how to evaluate? and who is involved in the evaluation?. The evaluation was applied under the form of a series of individual interviews carried with the professionals involved in the development and use of the MC-SDSS. The interviews were about the development, use and future of the MC-SDSS and were analyzed according to the thematic analysis method. Based on the challenges and difficulties identified in the thematic analysis, various recommendations are suggested to improve practices. Finally, to assess the potential to export the development of MC-SDSS to other contexts, the practices and perceptions of professionals from 11 municipalities in the Province of Quebec were documented during multidisciplinary group workshops (infrastructure, transport, urban planning, environment and geomatics). The objectives of these workshops were (1) to create a portrait of current practices for street rehabilitation and redesign to integrate urban planning, transportation and environment and (2) to identify the issues and challenges of developing MC-SDSS in various municipal contexts. The workshops were analyzed by schematizing the current decision processes and by using the causal mapping method. A series of guidelines is proposed to allow the development of MC-SDSS with the municipalities. Those guidelines aim at easing the development of a new type of decision support system for municipalities, but also at allowing the improvement of current municipal decision processes by better integrating the sustainable vision adopted in the politics and plans and the projects realized by the professionals.
Dehouche, Nassim. "Management de portefeuilles de projets : modèles multicritère d'évaluation, de sélection et d'argumentation." Thesis, Paris 9, 2014. http://www.theses.fr/2014PA090017.
Повний текст джерелаProject portfolio management (PPM) involves the use of methods and tools, allowing an organization to plan, evaluate, analyze and screen the execution of a set of projects or project proposals, sharing common resources or aiming at the attainment of common objectives. Multicriteria decision aid models are useful tools to support this process, given their ability to accurately model preferences, and rationally agregate points of view. However, existing models present some lacks that limit their use outside of academic circles : (i) They neglect the non-symetrical nature of the importance of some criteria that are relevant in PPM. (ii) The black box effect makes it hard to use them for the argumentation of decisions and to gain their acceptance by users (iii) They are implicitly fitted for private/for-profit projects, which limits their use in public organizations. In this thesis, our contribution consists in proposing two multicriteria methods for supporting the activities of evaluating, selecting and arguing decision, for project portfolio management. We propose: (i) An analysis of the specific features of public and private projects and their consequences for decision support (ii) A framework that allows an independent modeling of the abilities of a criterion to oppose and convince (iii) Two transparent multicriteria agregation procedures, fitted for different decision contexts. We ensure the theoretical validity of our approaches and illustrate their applicability on real data, with satisfying results
Bellaaj, Salma. "Adaptation aux risques liés aux stratégies d'externalisation : cas des entreprises industrielles tunisiennes." Phd thesis, Université Paris VIII Vincennes-Saint Denis, 2009. http://tel.archives-ouvertes.fr/tel-00638208.
Повний текст джерелаTacnet, Jean-Marc. "Prise en compte de l'incertitude dans l'expertise des risques naturels en montagne par analyse multicritères et fusion d'information." Phd thesis, Ecole Nationale Supérieure des Mines de Saint-Etienne, 2009. http://tel.archives-ouvertes.fr/tel-00771692.
Повний текст джерелаSirilertsuwan, Petchprakai. "Décisions de fabrication et modèle d’aide à la prise en charge à plusieurs niveaux pour améliorer la durabilité dans les chaînes d’approvisionnement en textiles et en vêtements." Thesis, Lille 1, 2020. http://www.theses.fr/2020LIL1I045.
Повний текст джерелаA recent trend towards sustainability has led to increases in various sustainable practices, but sustainability has still not been fully implemented into manufacturing and supply location decisions. A fragmented product supply chain (SC), which has various locations and multi-tier suppliers, leads to difficulties in traceability to ensure the SC has business, environmental, and social/socio-economic sustainability, known as the triple bottom line (TBL). Thus, this thesis aims to reveal which manufacturing decisions and location configurations better contribute to TBL, as well as to develop a location decision-support model for designing or evaluating multi-tier SCs with objective measurements and TBL factor considerations. Mixed methods are employed, including systematic literature reviews, semi-structured interviews, and SC simulations for the model formulation and its viscose t-shirt application. The thesis highlights that TBL benefits of proximity and distant manufacturing are location-dependent with their sources from spatial, cultural, ethnic, and linguistic proximity as well as country-, supplier-, and firm-specific. Spatial proximity benefits can be from proximity manufacturing to markets, materials suppliers, and headquarters as well as proximity between headquarters and market. The propose model has the potential to reveal the lowest or optimized cost and carbon dioxide equivalent (CO2e) SCs. The model is capable of revealing important factors and possible risks from future local and global disruptions, benefiting long-term supply chain planning. The model differentiates itself from the others by incorporating TBL from not only manufacturing and logistics activities but also sustainability assurance activities performed by suppliers and focal firms. The model potentially helps enhance TBL sustainability and supply chain visibility. This thesis has theoretical contributions to location theories, manufacturing decisions, cost and CO2e computational models, and sustainable multi-tier supply chain management
Mhamdi, Samir. "La concurrence entre les places financières, vers une nouvelle géographie des activités financières et de nouveaux centres émergents : applications des Modèles multicritères d'Aide à la décision : ACP, logique floue et MCO." Nice, 2008. http://www.theses.fr/2008NICE0032.
Повний текст джерелаWhat is meant by ‘financial centres’? What part do these play nationally and internationally? How may one analyse competition between financial centres? In this context, what strategies are followed? Our research has aimed to provide answers to these questions by undertaking an in-depth theoretical study followed by an empirical study shedding light on theoretical concepts. Several themes are discussed here: the competition between financial centres and the various strategies adopted by financial markets; the impact of technological innovations, of internationalisation and of competitive strategies on the characteristics of future financial centres; and, the geography of international financial operations
Sobrie, Olivier. "Learning preferences with multiple-criteria models." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLC057/document.
Повний текст джерелаMultiple-criteria decision analysis (MCDA) aims at providing support in order to make a decision. MCDA methods allow to handle choice, ranking and sorting problems. These methods usually involve the elicitation of models. Eliciting the parameters of these models is not trivial. Indirect elicitation methods simplify this task by learning the parameters of the decision model from preference statements issued by the decision maker (DM) such as “alternative a is preferred to alternative b” or “alternative a should be classified in the best category”. The information provided by the decision maker are usually parsimonious. The MCDA model is learned through an interactive process between the DM and the decision analyst. The analyst helps the DM to modify and revise his/her statements if needed. The process ends once a model satisfying the preferences of the DM is found. Preference learning (PL) is a subfield of machine learning which focuses on the elicitation of preferences. Algorithms in this subfield are able to deal with large data sets and are validated withartificial and real data sets. Data sets used in PL are usually collected from different sources and aresubject to noise. Unlike in MCDA, there is little or no interaction with the user in PL. The input data set is considered as a noisy sample of a “ground truth”. Algorithms used in this field have strong statistical properties that allow them to filter noise in the data sets.In this thesis, we develop learning algorithms to infer the parameters of MCDA models. Precisely, we develop a metaheuristic designed for learning the parameters of a MCDA sorting model called majority rule sorting (MR-Sort) model. This metaheuristic is assessed with artificial and real data sets issued from the PL field. We use the algorithm to deal with a real application in the medical domain. Then we modify the metaheuristic to learn the parameters of a more expressive model called the non-compensatory sorting (NCS) model. After that, we develop a new type of veto rule for MR-Sort and NCS models which allows to take criteria coalitions into account. The last part of the thesis introduces semidefinite programming (SDP) in the context of multiple-criteria decision analysis. We use SDP to learn the parameters of an additive value function model
Kilani, Meriam. "Multiple product-project decisions coordination support : application to oil and gas development projects." Electronic Thesis or Diss., université Paris-Saclay, 2023. http://www.theses.fr/2023UPAST027.
Повний текст джерелаThe major challenge addressed in this research concerns the coordination of the multiple interdependent decisions that must be made during the project, either technical, financial, or contractual. Interdependence means that making one decision without considering the impacts for other decisions may imply some underperformance, or even dead ends, iterations, and rework.To overcome this challenge, a more adaptable multi-decision-making process has been proposed, consisting of three blocks: 1/ modeling the decision network and formulating the multi-decision problem; 2/ structuring the problem to propose relevant and plausible scenarios assembled from elementary decision alternatives; 3/ solving the problem by selecting and recommending scenarios.Building the multi-decision-making process is based on multiple possibilities for each block. The decision-maker selects from a set of possible choices to adapt the decision-making process to the precise context.For block #1, we have first articulated the need to build a global decision network that models the decisions under study and the interdependencies they may have with other decisions. We have then argued that graphs and matrices can be used to fulfill this need. Both methods allow to include all decisions and interdependencies of the decision network in one single model, each of them having its advantages and drawbacks, with a kind of complementarity.Then, to formulate the local multi-decision problem, two interactions-based clustering approaches are proposed: the top-down approach (considering decision interdependencies) and the bottom-up approach (with an additional due date-based grouping of decisions). Both help to delineate the focus of decision makers on a specific set of decisions, since considering the whole network of decisions at the same time can be challenging.In block #2, to structure the problem, two matrix-based and one graph-based methods have been proposed. These methods offer the possibility to generate possible scenarios considering compatibility and performance criteria, either sequentially (morphological analysis), simultaneously (QFD), or with a hybrid way (graph exploration). For the two matrix-based methods, an algorithm was proposed to facilitate the identification of plausible scenarios. As for the graph-based method, a lighter heuristic can be applied on live during a decision meeting.Finally, to solve the problem in block #3, several MCDA methods have been listed for evaluating and selecting a recommended scenario: absolute compensatory methods, relative pairwise comparison methods, and relative comparison to reference point methods.According to industrial actors, such a process could improve coordination mechanisms between the major decisions of their projects. Even though decisions were interdependent, they were not often considered as such, and our proposed process permits (according to them) to have a better vision of the decisions to be made together and of the consequences of the choices. A fictitious case study, inspired by real past projects, was used to illustrate the proposed multi-decision coordination process.We are convinced that our research will provide a solid basis for further studies on the coordination of multiple interdependent decisions in complex projects, although there are academic and industrial perspectives that need to be tackled
Lamboray, Claude. "Prudent ranking rules: theoretical contributions and applications." Doctoral thesis, Universite Libre de Bruxelles, 2007. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210662.
Повний текст джерела
Doctorat en Sciences de l'ingénieur
info:eu-repo/semantics/nonPublished
Debord, Bernard. "Axiomatisation de procédures d'agrégation de préférences." Phd thesis, Grenoble 1, 1987. http://tel.archives-ouvertes.fr/tel-00010237.
Повний текст джерелаRakotoarivelo, Jean-Baptiste. "Aide à la décision multi-critère pour la gestion des risques dans le domaine financier." Thesis, Toulouse 3, 2018. http://www.theses.fr/2018TOU30067/document.
Повний текст джерелаWe are working on multicriteria decision analysis. We started with the study of risk typology through a huge review of literature. We have developed a theoretical model grouping four different categories of nineteen financial risk cases. Through this theoretical model, we have applied them to the "Caisse d'Epargne Midi-Pyrénées". In this thesis, we seek to make a contribution to the security management of information systems at the level of the savings bank. These analyzes are based on facts observed, collected and measured with real experiments resulting in its information system security policy and want to offer a pragmatic approach to the presentation of financial risk analysis through methods supporting. multicriteria decision analysis. The development of this model makes it possible to represent certain specific aspects of the financial risks that have often occurred in their activities. Our research led to the achievement of a concrete result in relation to the needs of the information system manager of the savings bank. It is an effective decision support system by constructing relative figures for the values of the criteria assigned by the RSSI
Zheng, Jun. "Elicitation des Préférences pour des Modèles d'Agrégation basés sur des Points de référence : Algorithmes et Procédures." Phd thesis, Ecole Centrale Paris, 2012. http://tel.archives-ouvertes.fr/tel-00740655.
Повний текст джерелаMarchant, Thierry. "Agrégation de relations valuées par la méthode de Borda, en vue d'un rangement: considérations axiomatiques." Doctoral thesis, Universite Libre de Bruxelles, 1996. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/212380.
Повний текст джерелаDepuis 20 à 30 ans, l'aide multicritère à la décision est apparue. L'expansion de cette nouvelle discipline s'est marquée dans la littérature essentiellement par un foisonnement de nouvelles méthodes multicritères d'aide à la décision et par des applications de celles-ci à des problèmes "réels". Pour la plupart de ces méthodes, il n'y pas ou peu de fondements théoriques. Seul le bon sens a guidé les créateurs de ces méthodes.
Depuis une dizaine d'années, le besoin de bases théoriques solides se fait de plus en plus sentir. C'est dans cette perspective que nous avons réalisé le présent travail. Ceci étant dit, nous n'allons pas vraiment nous occuper de méthodes multicritères à la décision dans ce travail, mais seulement de fragments de méthodes. En effet, les méthodes multicritères d'aide à la décision peuvent généralement être décomposées en trois parties (outre la définition de l'ensemble des alternatives et le choix des critères):
- Modélisation des préférences: pendant cette étape, les préférences du décideur sont modélisées le long de chaque critère.
- Agrégation des préférences: un modèle global de préférences est construit au départ des modèles obtenus critère par critère à la fin de la phase précédente.
- Exploitation des résultats de l'agrégation: du modèle global de préférences issu de la phase 2, on déduit un choix, un rangement, une partition, selon les besoins.
Jusqu'à présent, à cause de la difficulté du problème, peu de méthodes ont été axiomatisées de bout en bout; la plupart des travaux ne s'intéressent qu'à une ou deux des trois étapes que nous venons de décrire.
Nous nous sommes intéressés à une méthode bien connue: la méthode de Borda. Elle accepte comme données de départ des relations binaires. Elle intervient donc après la phase de modélisation des préférences. Le résultat de cette méthode est un rangement. Elle effectue donc les opérations correspondant aux étapes 2 et 3. Dans la suite de ce travail nous appellerons méthode de rangement toute méthode effectuant les étapes 2 et 3 pour aboutir à un rangement. Etant donné que les méthodes de rangement, celle de Borda en particulier, sont utilisées également en choix social, nous puiserons abondamment dans le vocabulaire, les outils et les résultats du choix social. Les résultats présentés seront valides en choix social, mais nous nous sommes efforcés de les rendre aussi pertinents que possible en aide multicritère à la décision.
Dans le chapitre II, après quelques définitions et notations, nous présentons quelques méthodes de rangement classiques, y compris la méthode de Borda, et quelques résultats majeurs de la littérature. Nous généralisons une caractérisation des méthodes de scorage due à Myerson (1995).
Nous nous tournons ensuite vers les relations valuées. La raison en est la suivante: elles sont utilisées depuis longtemps dans plusieurs méthodes multicritères et, depuis peu, elles le sont aussi en choix social (p.ex. Banerjec 1994) car elles permettent de modéliser plus finement les préférences des décideurs confrontés à des informations incertaines, imprécises, contradictoires, lacunaires, Nous commençons donc le chapitre III par des notations et définitions relatives aux relations valuées.
Ensuite, nous présentons quelques méthodes de rangement opérant au départ de relations valuées. C'est-à-dire des méthodes de rangement qui agissent non pas sur des relations nettes, mais sur des relations valuées et qui fournissent comme précédemment un rangement des alternatives. N'ayant trouvé dans la littérature aucune méthode de ce type, toutes celles que nous présentons sont neuves ou des généralisations de méthodes existantes; comme par exemple, les méthodes de scorage généralisées, que nous caractérisons en généralisant encore une fois le résultat de Myerson.
Nous présentons enfin ce que nous appelons la méthode de Borda généralisée, qui est une des généralisations possibles de la méthode de Borda au cas valué. Nous basant sur un article de Farkas et Nitzan (1979), nous montrons que contrairement à ce qui se passait dans le cas particulier envisagé par Farkas et Nitzan (agrégation d'ordres totaux), la méthode de Borda généralisée (et sa particularisation au cas net) n'est pas toujours équivalente à la méthode proximité à l'unanimité. Cette dernière méthode classe chaque alternative en fonction de l'importance des changements qu'il faudrait faire subir à un ensemble de relations pour que l’alternative considérée gagne à l'unanimité. Nous identifions quelques cas où l'équivalence est vraie.
Ensuite, nous reprenons un résultat de Debord (1987). Il s'agit d'une caractérisation de la méthode de Borda en tant que méthode de choix appliquée à des préordres totaux. Nous la généralisons de deux façons au cas de la méthode de Borda en tant que méthode de rangement appliquée à des relations valuées. Lorsqu'on applique la méthode de Borda, on est amené à calculer une fonction à valeurs réelles sur l'ensemble des alternatives.
La valeur prise par cette fonction pour une alternative s'appelle le score de Borda de cette alternative. Ensuite, on range les alternatives par ordre décroissant de leur score de Borda. La tentation est grande - et beaucoup y succombent (peut-être avec raison) d'utiliser le score de Borda non seulement pour calculer le rangement mais aussi pour estimer si l'écart entre deux alternatives est important ou non (voir par exemple Brans 1994). Cette approche n'a, à notre connaissance, jamais été étudiée d'un point de vue théorique. Nous présentons deux caractérisations de la méthode de Borda utilisée à cette fin.
Dans la dernière partie du chapitre III, nous abandonnons la démarche qui visait à caractériser une méthode par un ensemble de propriétés le plus petit possible. Nous comparons 12 méthodes sur base d'une vingtaine de propriétés. Les résultats de cette partie sont résumés dans quelques tableaux.
Ce travail aborde donc la méthode de Borda et sa généralisation au cas valué sous différents angles. Il livre une série de résultats qui, espérons-le, devraient permettre de mieux comprendre la méthode de Borda et peut-être de l'utiliser à meilleur escient. Toutefois, quoique notre objectif ait été de présenter des résultats pertinents en aide multicritère à la décision (et nous avons fait des progrès dans ce sens), il reste du chemin à faire. Nous sommes probablement encore trop proche du choix social. Ceci constitue donc une voie de recherche intéressante, de même que l'étude d'autres méthodes de rangement et l'étude de méthodes complètes d'aide multicritère à la décision: modélisation du problème (identification du ou des décideur(s), des alternatives et des critères), modélisation des préférences, agrégation des préférences et exploitation des résultats de l'agrégation.
Doctorat en sciences appliquées
info:eu-repo/semantics/nonPublished
Abichou, Bouthaïna. "Contribution à la formalition de bilans/états de santé multi-niveaux d'un système pour aider à la prise de décision en maintenance : agrégation d'indicateurs par l'intégrale de Choquet." Electronic Thesis or Diss., Université de Lorraine, 2013. http://www.theses.fr/2013LORR0020.
Повний текст джерелаThis work is addressing the health assessment of a multi-component system by means of multi-levels health check-up. Thus scientific Ph. D. objective aims to establish items of a generic health check-up concept. It focuses specifically on the functions of anomaly detection, normalization and aggregation of different indicators to develop a synthetic index representing the overall health status for each element within the system. In that way, it is proposed a new approach for detecting conditional anomalies. This approach has the advantage of quantifying the deviation for each indicator compared to its nominal behavior while taking into account the context in which the system operates. An extension of the Choquet integral used as an operator aggregating indicators is also proposed. This extension regards on the one hand, a process of an unsupervised learning of the capacity coefficients for the lowest level of abstraction, namely components level, and on the other hand, an approach to inference them from one level to another. These contributions are implemented on a ship diesel engine which is the most critical system for the BMCI project of the MER-PACA pole to which this thesis is attached
Staedelin, Marie. "Methodological approaches for the benefit-risk assessment of medicinal products in European regulatory decision-making : a special emphasis on the MultiCriteria Decision Analysis "MCDA” Method a quantitative approach." Thesis, Strasbourg, 2014. http://www.theses.fr/2014STRAJ013.
Повний текст джерелаThe benefit-risk evaluation of new medicines plays a central role in safeguarding public health. Nevertheless, it seems that the benefit-risk evaluation calls for further improvement. In 2010, no review had been performed of how available benefit-risk assessment methods could be applied for a regulatory benefit-risk assessment and how feasible that would be when facing real-life cases. The objective of this thesis has thus been to identify method(s) that could be theoretically used for such an assessment, and then to confront it/them to real-life cases, in order to determine their applicability. The results of the methods evaluation showed that the most suitable methods for a regulatory benefit-risk assessment of medicinal products are the MCDA method and the MCDA based methods. The results of the practical application of the MCDA indicated that the method could be used for medicinal products registered through a common registration scenario in Europe. However it should be noted that this method provides neither a “ready-made” recipe to perform an assessment nor a direct answer
Leitzelman, Mylène. "Mise en place d'un système d'informations stratégiques multicritères facilitant l'intégration des ressources régionales et la prise de décision dans le domaine de l'environnement Application à la ville de Marseille." Aix-Marseille 3, 1998. http://www.theses.fr/1998AIX30078.
Повний текст джерелаChakhchoukh, Mehdi. "Visualization to Support Multi-Criteria Decision-making in Agronomy." Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASG085.
Повний текст джерелаThe increasing complexity of agricultural systems necessitates sophisticated decision-making tools that can handle multiple criteria and accommodate complex trade-off analysis tasks. This thesis develops visualizations that facilitate decision-making processes in agronomy. This work has three main contributions: (i) understanding how provenance can support trade-off analysis, (ii) articulating high-level design and visualization needs to support group comparison in trade-off scenarios, and (iii) understanding how different visualizations can affect comparisons and decision-making in trade-off analysis. After an introductory chapter and a chapter on related work, the thesis details these three main contributions.The 3rd chapter of the thesis investigates how analytic provenance mechanisms can assist experts in recalling and tracking complex trade-off analyses. We developed VisProm, a web-based system integrating in-visualization provenance views to help experts track trade-offs and their objectives when exploring complex simulation results. Observation sessions with groups of experts revealed eight key tasks supported by our designs, highlighting new opportunities for provenance-driven trade-off analysis, such as monitoring trade-off space coverage and tracking alternative scenarios. One key outcome was the need to consider conflicting objectives and compare how different solutions or trade-off spaces fare under these objectives.Building on this, the 4th chapter explores the needs and challenges experts face when comparing trade-off spaces (that are often expressed as groups of data points, e.g., groups of simulation results) that optimize different objectives. Through workshops with domain experts and visualization designers, we identified high-level design and visualization needs to support group comparison in trade-off scenarios. This chapter lays the groundwork for developing effective visualization techniques for comparing groups that represent different trade-offs in terms of what objectives they optimize. They led to the implementation of a visualization prototype that visually encodes a variety of trade-off metrics. These encode and visually communicate experts' priorities in terms of objectives, the notion of ideal solutions, and how far current groups of solutions are from those ideals.The 5th chapter focuses on the evaluation of visualization techniques for comparing groups of points (solutions) when they represent different trade-offs. Motivated by the visualization needs and design requirements of the previous chapter, we selected three promising tabular-based visualization techniques to study. These techniques encode trade-off priorities and ideal solutions in different ways: coupling or decoupling the trade-off metrics and presenting them visually. We conducted a user study to understand how visualizations affected comparison decisions and quality of decision explanations. The findings from this study highlight that techniques that visually separate the encoding of priorities and ideal solutions lead to higher mental load and lower self-reported trust but may support more varied decision strategies than integrated visualizations. But they were always preferred over a baseline visualization.We conclude the thesis with a list of discussions and perspectives for future directions stemming from the results of this work
Abichou, Bouthaïna. "Contribution à la formalition de bilans/états de santé multi-niveaux d'un système pour aider à la prise de décision en maintenance : agrégation d'indicateurs par l'intégrale de Choquet." Thesis, Université de Lorraine, 2013. http://www.theses.fr/2013LORR0020/document.
Повний текст джерелаThis work is addressing the health assessment of a multi-component system by means of multi-levels health check-up. Thus scientific Ph. D. objective aims to establish items of a generic health check-up concept. It focuses specifically on the functions of anomaly detection, normalization and aggregation of different indicators to develop a synthetic index representing the overall health status for each element within the system. In that way, it is proposed a new approach for detecting conditional anomalies. This approach has the advantage of quantifying the deviation for each indicator compared to its nominal behavior while taking into account the context in which the system operates. An extension of the Choquet integral used as an operator aggregating indicators is also proposed. This extension regards on the one hand, a process of an unsupervised learning of the capacity coefficients for the lowest level of abstraction, namely components level, and on the other hand, an approach to inference them from one level to another. These contributions are implemented on a ship diesel engine which is the most critical system for the BMCI project of the MER-PACA pole to which this thesis is attached
Fayard, Nicolas. "Capability approach inspired tools for aiding policy design." Electronic Thesis or Diss., Université Paris sciences et lettres, 2024. http://www.theses.fr/2024UPSLD043.
Повний текст джерелаThis research explores the application of the Capability Approach (CA) within decision-aiding frameworks, focusing on public policy design.The CA is presented as an alternative to traditional welfare measures, offering a multidimensional framework that accounts for diversity and subjectivity.We propose an improved approach by incorporating systemic factors through mixed-integer linear programming, obtaining capability sets as Pareto frontiers.A proof of concept was developed by applying the CA to assess mental health capabilities of older adults in the urban context of Paris.Rather than comparing and aggregating, possibly multidimensional, solutions, new methodologies are required to compare and aggregate sets of Pareto-efficient solutions.The second part of this manuscript is dedicated to adapting classical aggregation methods to the CA, specifically focusing on aggregating capability sets in the presence of a utility function and dealing with uncertainty, both with and without subjective probabilities
Hoang, Giang Tran Thi. "D’un « diagnostic partagé » à « l’action partagée » : comment le mode projet Living Lab peut alimenter le processus d’aide à la décision du « faire la ville » en intégrant les citoyens pour une ville intelligente." Electronic Thesis or Diss., Université de Lorraine, 2019. http://www.theses.fr/2019LORR0168.
Повний текст джерелаIn the current era, Smart City projects have to deal with big social, ecological and technological challenges such as digitalization, pollution, democratic aspirations, need to be safe, etc. The higher involvement of multi-stakeholders (such as citizens, users, engineers, researchers, elected representatives, etc.) in the different phases of the projects, is one strategy enabling a variety of perspectives to be considered and thus develop a shared vision of the city. Paradoxically, the dynamic and multiple nature of stakeholders appears to be a source of complications and uncertainty in the decision-making process. Getting a consensus, which means legitimizing a group decision-making process, is often difficult to reach because the different expectations, needs and behaviours. The main hypothesis of the present research is that the consensus process could be facilitated by human and automatic tools through identifying conflicting points and following their evolution after several rounds of discussion. This study will propose a general methodology for two kinds of decision-making process supporting the negotiation between stakeholders within a Living Lab environment, an automatic control system that substitutes the facilitator’s actions and a process supervised by human facilitator. First, a methodology based on a multi-agent system is developed for automatic reaching consensus in large scale group of making decisions without facilitators and direct interaction. Then, a second methodology combining a human facilitator and ICT–based supporting tool. A proposition of multi-criteria and multi-stakeholders methodology supporting consensus reaching processes is used as decision support system tool in Smart City projects under Living Lab environment. Both Living Lab decision-making methodologies, the automatic and the supervised process, were developed, executed and validated in concrete case studies. Throughout using both methods, we can compare how different between the theoretical model and a process of experimenting decision-making tools in the real-life. The results of data analysis from experimentation provide suggestions to improve decision-making process of stakeholders with diverse objective and background in Smart City projects
Zowid, Fauzi Mohammed. "Development and performance evaluation of multi-criteria inventory classification methods." Thesis, Bordeaux, 2020. http://www.theses.fr/2020BORD0331.
Повний текст джерелаThis thesis deals with the issue of inventory classification within supply chains. More specifically, it aims to provide new alternative classification methods to address the multi-criteria inventory classification (MCIC) problem. It is well known that the ABC inventory classification technique is widely used to streamline inventory systems composed of thousands of stock-keeping-units (SKUs). Single-criterion inventory classification (SCIC) methods are often used in practice and recently MCIC techniques have also attracted researchers and practitioners. With regard to the MCIC techniques, large number of studies have been developed that belong to three main approaches, namely: (1) the machine learning (ML), (2) the mathematical programming (MP), and (3) the multi-criteria decision making (MCDM). On the ML approach, many research methods belonging to the supervised ML type have been proposed as well as a number of hybrid methods. However, to the best of our knowledge, very few research studies have considered the unsupervised ML type. On the MP approach, a number of methods have been developed using linear and non-linear programming, such as the Ng and the ZF methods. Yet, most of these developed methods still can be granted more attentions for more improvements and shortcomings reduction. On the MCDM approach, several methods have been proposed to provide ABC classifications, including the TOPSIS (technique for order preference by similarity to ideal solution) method, which is well known for its wide attractiveness and utilization, as well as some hybrid TOPSIS methods.It is worth noting that most of the published studies have only focused on providing classification methods to rank the SKUs in an inventory system without any interest in the original and most important goal of this exercise, which is achieving a combined service-cost inventory performance, i.e. the maximization of service levels and the minimization of inventory costs. Moreover, most of the existing studies have not considered large and real-life datasets to recommend the run of MCIC technique for real life implementations. Thus, this thesis proposes first to evaluate the inventory performance (cost and service) of existing MCIC methods and to provide various alternative classification methods that lead to higher service and cost performance. More specifically, three unsupervised machine learning methods are proposed and analyzed: the Agglomerative hierarchical clustering, the Gaussian mixture model and K-means. In addition, other hybrid methods within the MP and MCDM approaches are also developed. These proposed methods represent a hybridization of the TOPSIS and Ng methods with the triangular distribution, the Simple additive weighting (SAW) and the Multi-objective optimization method by ratio analysis (MOORA).To conduct our research, the thesis empirically analyzes the performance of the proposed methods by means of two datasets containing more than nine thousand SKUs. The first dataset is a benchmark dataset originating from a Hospital Respiratory Theory Unit, often used in the literature dealing with the MCIC methods, composed of 47 SKUs. The second dataset consists of 9,086 SKUs and coming from a retailer in the Netherlands. The performances of the proposed methods are compared to that of existing MCIC classification methods in the literature. The empirical results reveal that the proposed methods can carry promising performances by leading to a higher combined service-cost efficiency
Turcanu, Catrinel. "Multi-criteria decision aiding model for the evaluation of agricultural countermeasures after an accidental release of radionuclides to the environment." Doctoral thesis, Universite Libre de Bruxelles, 2007. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210642.
Повний текст джерелаThis work explores the application of multi-criteria decision aid methods for optimising food chain countermeasure strategies after a radioactive release to the environment.
The core of the thesis is dedicated to formulating general lines for the development of a multi-criteria decision aid model. This includes the definition of potential actions, construction of evaluation criteria and preference modelling and is essentially based on the results of a stakeholders’ process. The work is centred on the management of contaminated milk in order to provide a concrete focus and because of its importance as an ingestion pathway in short term after an accident.
Among other issues, the public acceptance of milk countermeasures as a key evaluation criterion is analysed in detail. A comparison of acceptance based on stochastic dominance is proposed and, based on that, a countermeasures’ acceptance ranking is deduced.
In order to assess “global preferences” taking into account all the evaluation criteria, an ordinal method is chosen. This method allows expressing the relative importance of criteria in a qualitative way instead of using, for instance, numerical weights. Some algorithms that can be used for robustness analysis are also proposed. This type of analysis is an alternative to sensitivity analysis in what concerns data uncertainty and imprecision and seeks to determine how and if a model result or conclusion obtained for a specific instance of a model’s parameters holds over the entire domain of acceptable values for these parameters.
The integrated multi-criteria decision aid approach proposed makes use of outranking and interactive methodologies and is implemented and tested through a number of case studies and prototype tools.
Doctorat en Sciences de l'ingénieur
info:eu-repo/semantics/nonPublished
Ciorna, Vasile. "VIANA : visualisation analytique au service de la conception des pneumatiques." Electronic Thesis or Diss., Bordeaux, 2024. http://www.theses.fr/2024BORD0500.
Повний текст джерелаWith the acceleration of technical breakthroughs in the past decades, more and more technical means are available to support companies in their product developments. While not new, lately, there is a clear expansion in the utilization of machine learning models within the internal processes of companies. However, the way these models are utilized is influencing the risk-reward balance as well as their adoption significantly. This project reports reflections on the design of visualizations to support effectively the utilization of machine learning methods in the frame of product design. We identify model explainability and rapid what-if scenario creation as central pillars for utilization and adoption. These learnings were integrated into two new visualization applications. Firstly, Interact is a novel What-If tool for regression models and an explainable artificial intelligence method. The tool provides single-model support to test and understand a model, validate its mechanisms through the utilization of statistical interactions, and anticipate on actions. The evaluation of this tool with experts from the tire industry revealed that the design of the user interface widgets was key for its effective utilization. More precisely, feedforward information and navigation cues were appreciated by our evaluators. Interact includes simple widgets that convey such information. However, our users requested, in addition, new capabilities for multi-model support. This new direction called for a more in-depth analysis of prior work on how to design and use navigation cues and embedded visualizations to convey feedforward information. Accordingly, we surveyed the visualization literature for work including “scented” or sophisticated widgets. An online browser allowing exploration of relevant work comprising “scented widgets” is made available. This literature survey allowed to identify a research opportunity related to suitable visualizations for multi-models or tradeoff contexts. Respectively, we conducted a crowdsourcing experiment to compare line charts and colorfields as effective cues for multi-inputs, multi-outputs contexts. Line charts showed better performance compared to colorfields for novice users. Secondly, VisDM, builds on previous experiences such as the development of Interact and focuses on multi-inputs, multi-outputs with a specific emphasis on decision-making in product design. The tool allows using simultaneously up to 7 models that work together to assess tradeoffs. Scented widgets allow anticipating on actions effectively, and Multi Criteria Decision Making methods allow deciding on competing alternatives
De, Smet Yves. "A multicreteria perspective on reverse auctions." Doctoral thesis, Universite Libre de Bruxelles, 2005. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210974.
Повний текст джерелаDoctorat en sciences appliquées
info:eu-repo/semantics/nonPublished
Willot, Hénoïk. "Certified explanations of robust models." Electronic Thesis or Diss., Compiègne, 2024. http://www.theses.fr/2024COMP2812.
Повний текст джерелаWith the advent of automated or semi-automated decision systems in artificial intelligence comes the need of making them more reliable and transparent for an end-user. While the role of explainable methods is in general to increase transparency, reliability can be achieved by providing certified explanations, in the sense that those are guaranteed to be true, and by considering robust models that can abstain when having insufficient information, rather than enforcing precision for the mere sake of avoiding indecision. This last aspect is commonly referred to as skeptical inference. This work participates to this effort, by considering two cases: - The first one considers classical decision rules used to enforce fairness, which are the Ordered Weighted Averaging (OWA) with decreasing weights. Our main contribution is to fully characterise from an axiomatic perspective convex sets of such rules, and to provide together with this sound and complete explanation schemes that can be efficiently obtained through heuristics. Doing so, we also provide a unifying framework between the restricted and generalized Lorenz dominance, two qualitative criteria, and precise decreasing OWA. - The second one considers that our decision rule is a classification model resulting from a learning procedure, where the resulting model is a set of probabilities. We study and discuss the problem of providing prime implicant as explanations in such a case, where in addition to explaining clear preferences of one class over the other, we also have to treat the problem of declaring two classes as being incomparable. We describe the corresponding problems in general ways, before studying in more details the robust counter-part of the Naive Bayes Classifier
Slavich, John. "Processus de quantification des attributs logiciels, PQAL, basé sur une méthode de prise de décisions multicritères." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2000. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape3/PQDD_0008/MQ60922.pdf.
Повний текст джерелаPérinel, Emmanuel. "Segmentation en analyse de données symboliques : le cas de données probabilistes." Paris 9, 1996. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1996PA090079.
Повний текст джерелаFarah, Mohamed. "Approche multicritère des problèmes de recherche documentaire." Paris 9, 2006. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=2006PA090028.
Повний текст джерелаResearch in Information Retrieval shows performance improvement when many sources of evidence are combined to produce a ranking of documents. When documents are evaluated according to some attributes or criteria, most current approaches assess document relevance by computing a single score which aggregates performance values w. R. T. These attributes. When only ranked lists of relevant documents are provided by multiple ‘blackbox’ methods, most current approaches combine these lists into one single output result set. In both cases, we propose ranking methods within a multiple criteria framework using aggregation mechanisms based on decision rules identifying positive and negative reasons for judging whether a document should get a better ranking than another. The proposed methods deal well with the Information Retrieval distinctive features. Experimental results are reported in both contexts
Plantié, Michel. "Extraction automatique de connaissances pour la décision multicritère." Phd thesis, Ecole Nationale Supérieure des Mines de Saint-Etienne, 2006. http://tel.archives-ouvertes.fr/tel-00353770.
Повний текст джерелаLe modèle qui supporte notre système interactif d'aide à la décision de groupe (SIADG) s'appuie largement sur des traitements automatiques de la connaissance. Datamining, multicritère et optimisation sont autant de techniques qui viennent se compléter pour élaborer un artefact de décision qui s'apparente à une interprétation cybernétique du modèle décisionnel de l'économiste Simon. L'incertitude épistémique inhérente à une décision est mesurée par le risque décisionnel qui analyse les facteurs discriminants entre les alternatives. Plusieurs attitudes dans le contrôle du risque décisionnel peuvent être envisagées : le SIADG peut être utilisé pour valider, vérifier ou infirmer un point de vue. Dans tous les cas, le contrôle exercé sur l'incertitude épistémique n'est pas neutre quant à la dynamique du processus de décision. L'instrumentation de la phase d'apprentissage du processus décisionnel conduit ainsi à élaborer l'actionneur d'une boucle de rétroaction visant à asservir la dynamique de décision. Notre modèle apporte un éclairage formel des liens entre incertitude épistémique, risque décisionnel et stabilité de la décision.
Les concepts fondamentaux de connaissance actionnable (CA) et d'indexation automatique sur lesquels reposent nos modèles et outils de TALN sont analysés. La notion de connaissance actionnable trouve dans cette vision cybernétique de la décision une interprétation nouvelle : c'est la connaissance manipulée par l'actionneur du SIADG pour contrôler la dynamique décisionnelle. Une synthèse rapide des techniques d'apprentissage les plus éprouvées pour l'extraction automatique de connaissances en TALN est proposée. Toutes ces notions et techniques sont déclinées sur la problématique spécifique d'extraction automatique de CAs dans un processus d'évaluation multicritère. Enfin, l'exemple d'application d'un gérant de vidéoclub cherchant à optimiser ses investissements en fonction des préférences de sa clientèle reprend et illustre le processus informatisé dans sa globalité.
Othmani, Imed. "Optimisation multicritère : fondements et concepts." Phd thesis, Université Joseph Fourier (Grenoble), 1998. http://tel.archives-ouvertes.fr/tel-00004900.
Повний текст джерелаSerrano, Franck. "Aide multicritère à la décision en matière d'économie d'énergie." Aix-Marseille 2, 1991. http://www.theses.fr/1992AIX24008.
Повний текст джерелаThe aim of this work is to develop a multicriteria decision aid method to solve town's energy efficiency problems. The second goal is to realize a sensibility analysis to prepare the programming stage
Gzara, Mariem. "Méthode coopérative d'aide multicritère à l'ordonnancement flou." Lille 1, 2001. http://www.theses.fr/2001LIL10117.
Повний текст джерелаRommel, Yves. "Apport de l'intelligence artificielle à l'aide à la décision multicritère." Paris 9, 1989. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1989PA090004.
Повний текст джерелаThe aim of this thesis is to show the contribution of artificial intelligence technics (AI) to multicriteria decision making. The AI technics which have been used concern knowledge representation technics in a declarative manner, deductive technics of problem resolution and also "inductive" technics of machine learning. Research has been developped into two opposite directions relatively to the modelisation approach as follows: -the first direction consists of defining how much AI contributes to the traditional multicriteria analysis frame i. E. We consider here the multicriteria problems using a scores table and according to the three operational approaches to the aggregation of preferences (synthesis criterion, outranking and interactive procedure); - the second direction consists of considering multicriteria methodology within the expert-system architecture and modelisation framework. This rests on the fact that expert-systems naturally deal with multicriteria problems