Cette thèse est consacrée à la description, la prédiction et la prévention des défaillances de systèmes. Elle se compose de quatre parties relatives à la modélisation stochastique de dégradation, au pronostic de défaillance du système, à l'estimation du niveau de défaillance et à l'optimisation de maintenance.Le processus d'Ornstein-Uhlenbeck (OU) dépendant du temps est introduit dans un objectif de modélisation des dégradations. Sur la base de ce processus, le premier instant de passage d’un niveau de défaillance prédéfini est considéré comme l’instant de défaillance du système considéré. Différentes méthodes sont ensuite proposées pour réaliser le pronostic de défaillance. Dans la suite, le niveau de défaillance associé au processus de dégradation est estimé à partir de la distribution de durée de vie en résolvant un problème inverse de premier passage. Cette approche permet d’associer les enregistrements de défaillance et le suivi de dégradation pour améliorer la qualité du pronostic posé comme un problème de premier passage. Le pronostic de défaillances du système permet d'optimiser sa maintenance. Le cas d'un système contrôlé en permanence est considéré. La caractérisation de l’instant de premier passage permet une rationalisation de la prise de décision de maintenance préventive. L’aide à la décision se fait par la recherche d'un niveau virtuel de défaillance dont le calcul est optimisé en fonction de critères proposés / This thesis is dedicated to describe, predict and prevent system failures. It consists of four issues: i) stochastic degradation modeling, ii) prognosis of system failures, iii) failure level estimation and iv) maintenance optimization. The time-dependent Ornstein-Uhlenbeck (OU) process is introduced for degradation modeling. The time-dependent OU process is interesting from its statistical properties on controllable mean, variance and correlation. Based on such a process, the first passage time is considered as the system failure time to a pre-set failure level. Different methods are then proposed for the prognosis of system failures, which can be classified into three categories: analytical approximations, numerical algorithms and Monte-Carlo simulation methods. Moreover, the failure level is estimated from the lifetime distribution by solving inverse first passage problems. This is to make up the potential gap between failure and degradation records to reinforce the prognosis process via first passage problems. From the prognosis of system failures, the maintenance optimization for a continuously monitored system is performed. By introducing first passage problems, the arrangement of preventive maintenance is simplified. The maintenance decision rule is based on a virtual failure level, which is solution of an optimization problem for proposed objective functions

Consider the following questions in the early stage of new product development. What should be the target market for proposed design concepts? Who will be the competitors and how fast are they moving forward in terms of performance improvements? Ultimately, is the current design concept and targeted launch date feasible and competitive? To answer these questions, there is a need to integrate the product benchmarking with the assessment of performance improvement so that analysts can have a risk measure for their R&D target setting practices. Consequently, this study presents how time series benchmarking analysis can be used to assist scheduling new product releases. Specifically, the proposed model attempts to estimate the "auspicious" time by which proposed design concepts will be available as competitive products by taking into account the rate of performance improvement expected in a target segment. The empirical illustration of commercial airplane development has shown that this new method provides valuable information such as dominating designs, distinct segments, and the potential rate of performance improvement, which can be utilized in the early stage of new product development. In particular, six dominant airplanes are identified with corresponding local RoCs and, inter alia, technological advancement toward long-range and wide-body airplanes represents very competitive segments of the market with rapid changes. The resulting individualized RoCs are able to estimate the arrivals of four different design concepts, which is consistent with what has happened since 2007 in commercial airplane industry. In addition, the case study of the Exascale supercomputer development is presented to demonstrate the predictive use of the new method. The results indicate that the current development target of 2020 might entail technical risks considering the rate of change emphasizing power efficiency observed in the past. It is forecasted that either a Cray-built hybrid system using Intel processors or an IBM-built Blue Gene architecture system using PowerPC processors will likely achieve the goal between early 2021 and late 2022. This indicates that the challenge to improve the power efficiency by a factor of 23 would require the maximum delay of 4 years to reach the Exascale supercomputer compared to the existing performance curve.

Made available in DSpace on 2011-05-04T12:36:19Z (GMT). No. of bitstreams: 0 Previous issue date: 2009 / O Instituto Oswaldo Cruz, unidade de pesquisa biomédica da Fundação Oswaldo Cruz, quer chegar em 2015 como um Instituto de Pesquisa e Tecnologia internacional de excelência, formador de cientistas e técnicos, reconhecido pela qualidade de sua ação de referência de diagnóstico, assistência e vigilância epidemiológica e capaz de responder às demandas na área de saúde com rapidez e confiabilidade. Este trabalho apresenta uma proposta de estudo prospectivo como ferramenta indispensável de planejamento estratégico para subsidiar a tomada de decisões e a formulação das políticas institucionais que façam o Instituto Oswaldo Cruz alcançar suavisão de futuro. A metodologia utilizada envolveu a realização de revisão da literatura para a construção do referencial teórico, a análise comparativa de três estudos prospectivos realizados no país e um realizado por instituição estrangeira e a identificação dos instrumentos e ferramentas utilizados nesses estudos. O resultado foi a apresentação de uma proposta de estudo prospectivo dividido em três fases: Inicial, de coleta, tratamento e sistematização da informação; Principal, de análise e interpretação da informação e produção do conhecimento; e de Comprometimento, de disseminação dos resultados, que propiciará ao IOC alcançar a excelência descrita em sua visão de futuro. / The Oswaldo Cruz Institute, the Oswaldo Cruz Foundation's biomedical research unit, is aiming to reach 2015 as "a international Institute of Research and Technology of excelence, former of scientists and technicians, acknowledged by the quality of its actions of reference to diagnosis, assistance and epidemiologic surveillance, and able to respond to the demands in the health area with agility and reliability." This paper represents a proposal of prospective study as an indispensable strategic planning tool to subsidize the decision making and the formulation of institutional policies to make the Oswaldo Cruz Institute reach its vision of the future. The methodology used involved the realisation of a literature review in order to build the theoretical referential, the comparative analysis of three prospective studies made in the country and one made by a foreign institution and the identification of the instruments and tools used in these studies. The result was the presentation of a prospective study proposal divided in three phases: the Inicial one, of collection, treatment and sistematization of the information; the Principal one, of analysis and interpretation of this information and the knowledgement production; and the Commitment one, of dissemination of the results, wich will propiciate the Institute Oswaldo Cruz to reach the excelence described in its vision of the future.