Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2015-06-26T12:34:11Z No. of bitstreams: 1 LuisMachado.pdf: 875680 bytes, checksum: e6382dece8db46bfbf9e76baa4beff73 (MD5) / Made available in DSpace on 2015-06-26T12:34:11Z (GMT). No. of bitstreams: 1 LuisMachado.pdf: 875680 bytes, checksum: e6382dece8db46bfbf9e76baa4beff73 (MD5) Previous issue date: 2010 / Nenhuma / O objetivo do presente estudo é analisar as formas institucionais de regulação da produção de petróleo dos Estados Unidos, Brasil, Argentina, Venezuela e Arábia Saudita, especialmente do ponto de vista da forma contratual com que esses países concedem a exploração em seus territórios e do nível de intervenção direta do Estado nessa exploração. Nesse sentido, a regulamentação dos setores de serviços de utilidade pública tende a acomodar os interesses de grupos de pressão, especialmente as indústrias reguladas e os consumidores, como forma de apoio político ao governo. Portanto, é imperioso que se adicione à teoria econômica da regulação o aspecto institucional dos países como elemento-chave na valorização de um ou outro grupo de interesse através da intervenção na economia. O petróleo é um produto estratégico no cenário da economia internacional, e por isso os Estados tendem a apoiar ou criar empresas para a produção de petróleo, procurando participar dos ganhos pela sua extração. Assim, buscou-se investigar quatro realidades distintas quanto à produção, consumo, capacidade de refino e comércio do petróleo nos países estudados. Além disso, foram analisadas as formas de arranjos institucionais distintos pelos quais esses países regulam a exploração do petróleo em seus territórios, seja através de contratos de concessão e institucionalização de agências reguladoras do setor, como os Estados Unidos, ou apenas com contratos de concessão sem agências reguladoras, como no caso argentino. No caso venezuelano, existe apenas contrato de joint-venture entre a iniciativa privada e a empresa estatal de exploração, sem agência reguladora. A Arábia Saudita não possui agência reguladora e explora a extração de petróleo diretamente pela empresa do súdito, da qual detém 100% de participação, e assim contrata apenas serviços específicos, sem concessão. Finalmente o caso brasileiro, onde se verifica a existência de contratos de concessão, bem como agência reguladora do setor e uma empresa de capital estatal atuando no mercado. / The objective of the present study is to analyze the institutional forms of oil production regulation in United States, Brazil, Argentina and Venezuela, especially the contractual point of view these countries grant the exploration in its territories and about the level of direct intervention of the State in this exploration. Regulation about public utility services tends to especially accommodate the interests of lobbies, the regulated industries and the consumers, as form to support government politician. Therefore, is necessary adds in the economic regulation theory the institutional aspect about the countries like element-key in the valuation if one or another interest group through the intervention in the economy for regulation. Oil is a strategical product in the scene of the international economy, and therefore the States tend to support or to create companies for oil production, being looked for to participate in profits for its extraction. Four different realities how much production, consumption, capacity of refining and commerce of the oil was studied. Moreover, the forms of distinct institutional arrangements had been analyzed for which these countries regulate the exploration of the oil in its territories, either through contracts of concession and institutionalization sector regulating agencies, as the United States, or only with concession contracts without regulating agencies, as in the Argentine case. In the Venezuelan case, only contract of joint-venture between the private initiative and the state-owned company of exploration exists, without regulating agency. Saudi Arabia does not possess regulating agency and explores the extration of oil directly for the company of the subject, of which it withholds 100% of participation, and thus contracts only specific services, without concession. Finally the Brazilian case, where there is concession contract, as well as regulating agency of the sector and a company of state capital acting in the market.

132f. / Submitted by Suelen Reis (suziy.ellen@gmail.com) on 2013-04-10T19:22:39Z No. of bitstreams: 1 AntonioAraujo.pdf: 2047390 bytes, checksum: bda07bb1207880fe36653d55fd259804 (MD5) / Approved for entry into archive by Rodrigo Meirelles(rodrigomei@ufba.br) on 2013-05-08T11:59:57Z (GMT) No. of bitstreams: 1 AntonioAraujo.pdf: 2047390 bytes, checksum: bda07bb1207880fe36653d55fd259804 (MD5) / Made available in DSpace on 2013-05-08T11:59:57Z (GMT). No. of bitstreams: 1 AntonioAraujo.pdf: 2047390 bytes, checksum: bda07bb1207880fe36653d55fd259804 (MD5) Previous issue date: 2011 / Diversos sistemas existentes, seja na natureza, sejam nas sociedades humanas apresentam acaracterística peculiar de serem formados por subsistemas ou indivíduos que cooperam entre si para a realização de suas tarefas, e consequentemente alcançarem um objetivo global. Exemplos desses sistemas vão de uma sociedade de abelhas a uma organização financeira, de uma comunidade animal a um sistema de manufatura, entre tantos outros. Ao se observar um campo de produção de petróleo, por exemplo, é possível perceber a natureza distribuída e colaborativa desse tipo de sistema, onde diversos poços (de produção ou injeção), estações de coleta e tratamento de óleo, entre outros, distribuídos geograficamente, têm uma distribuição que naturalmente os induziria a cooperar constantemente para que os objetivos de produção sejam atingidos. Apesar disso, as atuais arquiteturas para operação de um campo de produção utilizam uma estrutura centralizada para o seu gerenciamento. As oportunidades advindas da automação, do aumento da capacidade de processamento dos sistemas computacionais e da diminuição do custo das plataformas de hardware têm, por sua vez, ensejado oportunidades para aperfeiçoar o modo de produção dos campos de petróleo. Os sistemas de manufatura por sua vez, têm sido cada vez mais influenciados pelos sistemas multiagentes, um novo paradigma para o desenvolvimento de software que tem no agente o seu elemento principal. Alguns dos benefícios deste paradigma são: maior poder de atuação, por conta comportamento autônomo e inteligente, maior capacidade de lidar com as falhas nos sistemas de computação, por conta da descentralização do controle, e maior flexibilidade por conta da estrutura dinâmica dos agentes. Este trabalho apresenta um modelo baseado em sistema multiagente para o gerenciamento e controle de um campo de petróleo, cuja proposta visa a dotar cada unidade do sistema produtivo com a capacidade de análise local de suas condições baseada no conhecimento especialista embarcado, de comunicação e de negociação com outras unidades a fim de estabelecer decisões com base em uma visão global do sistema que permita consequentemente, melhorar a sua operação. / Salvador

The production of oil in Brazil is increasing to each year. As result a increasing volume of production water is generated as part of the process itself of formation of the deposits, many times in bigger ratios that the oil itself. The produced water owns a great diversity of contaminants, what ally to its high volume, brings serious problems at the moment of giving a final disposal to it. Some solutions had been attempted; on account of the more restrictive laws each day of the ambient agencies, but still in palliative way. The present work considers evaluating eletrodics materials and experimental conditions adjusted to be used in reactors, propitiating the electrochemical treatment of this residue. As anodes Ânodos Dimensionalmente Estáveis (ADE) commercial had been used, with nominal composition of Ti/Ru0,34Ti0,66O2. As cathodes the ADE, 316 steel and graphite had been evaluated. Amongst the cathodic materials studied the steel it presented performance better, as much with regard to efficiency of the process how much in the reduction of the cathodic incrustation. In the studies in batelada, was verified that the dilution of the solution and the current density influence significantly in the efficiency of the oxidation process. The increase of the dilution increases greatly the efficiency of the process. Already the speed of reduction of the DQO increases with the current density, however, the increase is not linear indicating that the process is controlled for diffusion. The efficiency of the electrochemical process in the reduction of the DQO of the sample was associated the three factors:a) the direct oxidation of the organic residue on the electrode; b) removal of the powdered material for the eletrogenerated gases (eletroflotation); c) indirect oxidation through reactive intermediate species. In the case of the reactor, it could be noticed that, the control of the outflow and the current density, as well as of the height of the flotant column influence in the efficiency of the treatment of the effluent, one beyond minimizing the incrustation on the cathodes. In the optimized conditions: outflow 800mLh-1, current density of 50mAcm-2 and flotant column of 10cm got it considerable reduction of the DQO and the TOG, proving the possibility of the application of the electrochemical technology in the treatment of the produced water of oil. / Conselho Nacional de Desenvolvimento Científico e Tecnológico / A produção de petróleo no Brasil vem aumentando a cada ano. Em conseqüência, um volume crescente de água de produção é gerado como parte do próprio processo de formação das jazidas, muitas vezes em proporções maiores que o próprio óleo. A água produzida possui uma grande diversidade de contaminantes, o que aliado ao seu alto volume, traz sérios problemas no momento de lhe dar uma disposição final. Várias soluções foram tentadas, por conta das leis cada dia mais restritivo dos órgãos ambientais, mas ainda de maneira paliativa. O presente trabalho propõe avaliar materiais eletródicos e condições experimentais adequadas para serem utilizadas em reatores, propiciando o tratamento eletroquímico desse resíduo. Como ânodo foram utilizados ânodos dimensionalmente estáveis (ADE) comerciais, com composição Ti/Ru0, 34Ti0,66O 2. Como cátodos foram avaliados os ADE, aço 316 e grafite. Dentre os materiais catódicos estudados o aço apresentou melhor desempenho, tanto com relação à eficiência do processo, quanto na redução da incrustação catódica. Nos estudos em batelada, verificou-se que a diluição da solução e a densidade de corrente influenciam significativamente na eficiência do processo de oxidação. O aumento da diluição aumenta acentuadamente a eficiência do processo. Já a velocidade de redução da DQO aumenta com a densidade de corrente, no entanto o aumento não é linear indicando que o processo é controlado por difusão. A eficiência do processo eletroquímico na redução da DQO da amostra foi associada a três fatores: a) ((oxidação direta do resíduo orgânico sobre o eletrodo; b) remoção do material particulado pelos gases eletrogerados (eletroflotação); c) oxidação indireta através de espécies intermediárias reativas. No caso do reator, pôde ser notado que, o controle da vazão e da densidade de corrente, bem como da altura da coluna flotante influenciam na eficiência do tratamento do efluente, além de minimizar a incrustação sobre os cátodos. Nas condições otimizadas: vazão de 800 cm3. h-1, densidade de 50mAcm-2 e coluna flotante de 10 cm obteve-se a redução considerável da DQO e do TOG, comprovando a possibilidade da aplicação da tecnologia eletroquímica no tratamento da água produzida de petróleo.

Nigeria and the United Kingdom are leading oil producers within their region. Both countries are linked by their exploration and production maturity within their regions and the fact that they produce similar oil grades. Their institutional similarities and their economic status as developed and developing economies provide the platform upon which this study basis its comparative investigation. On account of the oil price phenomenon and oil supply concerns by way of reserves, this study investigates the effect of the Hotelling theory, the Hubbert theory and Engineering decline curve theory on actual production rates within Nigeria and the United Kingdom. It develops individual models for both countries, applying each theory to each country to analyse the individual effect and the effect in comparison to one another. The level of adherence of each country to these production theories is measured, following which a combination of all three theories is applied to both country cases to quantify the level of significance and relationship to actual production behaviour. The results leave us with the understanding that the Hubbert theory does indeed capture the behaviour of production in both countries even where it was not expected in the case of Nigeria. The reserve constraint remains a key factor in future production plans for both countries. The results of the empirical analysis provide evidence of strong support in the United Kingdom for the Hotelling theory and weak support in the case of Nigeria. Oil price also remains a key factor in production modelling, though less so in the case of Nigeria. The engineering modelling approach on the other hand failed to explain Nigeria’s production profile, while it captured that of the UK; indicating that production in Nigeria is yet to decline. This result is corroborated by the projected peak production date seen in the Hubbert forecast model for Nigeria, and the weaker support Nigeria shows for the Hubbert theory. The study concludes by developing a combined model using all three theories to quantitatively analyse which of them best explains the country oil production profile. The results lead us to conclude that despite the fact that there is weak support for a production theory; an empirical analysis of the data does show that the relationship may not be insignificant. The augmentation along with the combination of the production models sheds more light on actual behaviour as it provides a more in-depth understanding on actual oil production behaviour.

Face aux difficultés d’identification des arbres de la famille des Lauracées et dans un but d’établissement d’une production durable et certifiée d’huiles essentielles, l’entreprise KLR a voulu mettre en place un nouvel outil d’identification multi-critères qui consiste à croiser plusieurs types de caractères afin d’obtenir une détermination taxonomique fiable.Trois méthodes d’identification ont été utilisées : la chimie (analyse de l’odeur de l’écorce), le code barre génétique et l’anatomie du bois. Une analyse par positionnement multidimensionnel non-métrique a permis de faire un tri des critères anatomiques du bois pertinents dans l’identification des Lauracées étudiées, au final 14 critères ont été retenus. D’après les données anatomiques dont nous disposons, nous proposons une clé de détermination au genre. En génétique, la technique de code barres de la région chloroplastique intergénique trnH-psbA a été appliquée aux individus récoltés. En comparant les séquences de tous les individus, 40% des individus peuvent être identifiés au genre car ils présentent des haplotypes uniques. Des haplotypes au sein des genres ont été déterminés et permettent d’identifier l’espèce pour 40 à 100 % des individus selon le genre. L’analyse SPME de l’odeur des écorces a permis de définir pour chaque espèce une carte d’identité chimique à partir des molécules majoritaires. Des chémotypes ont été définis au sein des espèces Licaria cannella, Ocotea indirectinervia et Sextonia rubra. Ces trois domaines d’observation ont permis de constituer une base de type Xper2 pour l’identification.Un criblage des essences intéressantes de par leur rendement en huile essentielle a été effectué. En ce qui concerne la partie extraction de l’huile essentielle, trois espèces montrent un rendement en huile essentielle de l’écorce supérieur à 0,3% et 5 espèces montrent un rendement en huile essentielle du bois supérieur à 0,3%. Les évaluations olfactives menées par un expert ont permis de séparer les huiles distillées en 5 différentes notes. Nous avons jugé intéressant de savoir si l’outil d’identification multicritère pouvait prédire les compositions et rendements en huile essentielle de l’espèce identifiée avant l’abattage de l’arbre. Des études complémentaires ont donc été menées. Pour plus de la moitié des arbres, la signature de l’écorce, la composition de l’huile essentielle de l’écorce et la composition de l’huile essentielle du bois sont similaires. La tentative d’établir un lien entre les caractéristiques des cellules à huile et le rendement s’est avérée infructueuse. / In order to develop a sustainable production of certified essential oils extracts, the compangy KLR has struggled with the identification of the trees from the family of Lauraceae. This is the main reason that led to the project of creating an identification tool for this family.This tool would be based on three approaches: chemistry (chemical signature of the dark), wood anatomy and genetic barcoding. By means of a non-metric multidimensional scaling analysis, we selected a list of 14 relevant wood anatomical features that can identify the trees to the genera. For the genetics, the barcoding of trnH-psbA an intergenic region of the DNA has been used. When comparing all the sequences from all the individuals, 40% of the individuals can be identified to the genera because of their specific haplotype. Haplotypes within the genera have also been established. Chemical identity cards have been set with the SPME analysis technique and chemotypes have been defined inside the species Licaria cannella, Ocotea indirectinervia and Sextonia rubra. These three identification approaches allowed us to construct an identification tool on Xper².During the study, we aimed to screen what species would be interesting for its essential oil yield. Three species showed an essential oil bark yield above 0.3%. five species showed an essential oil wood yield above 0.3%. An expert held an olfactory evaluation on the oils and divided the essential oils in five olfactory groups. We found it relevant to improve the identification tool by trying to predict the compositions of the oils before the three was cut down. In the case of the half of trees studied by SPME analyses, the composition of the wood and the bark essential oils were similar. On the other hand, no link was found between the oil cells dimensions and the essential oil yield.

[pt] Ambientes industriais, especialmente no setor de petróleo e gás, apresentam desafios únicos para técnicas de identificação de sistemas. Apesar dos avanços, ainda existe uma lacuna em nossa compreensão da integração de algoritmos black box, abordagens grey box e aprendizado de máquina para simulação de dados reais. Com o objetivo de otimizar a compreensão e previsão em ambientes industriais complexos, foram explorados a simulação de dados do mundo real na produção de petróleo e análise da velocidade de perfuração. Este estudo propõe uma análise da integração de algoritmos black box, abordagens grey box e aprendizado de máquina na simulação de dados reais, com ênfase na produção de petróleo e o estudo da interação broca rocha no processo de perfuração de poços de petróleo. Neste trabalho foram empregados técnicas de aprendizado de máquina, como redes neurais e métodos clássicos de identificação de sistemas, como modelos lineares como ARX (AutoRegressive with eXogenous inputs) e não lineares como o NARX (Nonlinear AutoRegressive with eXogenous inputs), para capturar os comportamentos dinâmicos dos processos em estudo utilizando dados reais da produção de petróleo e de perfuração, levando em consideração as características específicas e desafios operacionais desses ambientes. Com base nos resultados obtidos, as técnicas utilizadas demonstraram viabilidade de aplicação, na qual a utilização desses modelos híbridos, que combinam conhecimentos físicos com abordagens de múltiplos modelos formados por algoritmos de identificação de sistemas e aprendizado de máquina, demonstrou potencial para aprimorar as simulações. Esses resultados ressaltam a eficácia desses métodos, indicando que pesquisas futuras podem se dedicar à implementação dessa técnica na identificação de sistemas complexos. / [en] Industrial environments, especially in the oil and gas sector, presentunique challenges for system identification techniques. Despite advancements,there still exists a gap in our understanding of integrating black box algorithms,grey box approaches, and machine learning for simulating real-world data.With the aim of optimizing understanding and prediction in complex industrialenvironments, real-world data simulation in oil production and drilling speedanalysis was explored. This study proposes an analysis of the integration ofblack box algorithms, grey box approaches, and machine learning in simulatingreal-world data, with an emphasis on oil production and the study of the drill-rock interaction in the oil well drilling process. In this work, machine learningtechniques such as neural networks and classical system identification methods,such as linear models like ARX (AutoRegressive with eXogenous inputs) andnonlinear ones like NARX (Nonlinear AutoRegressive with eXogenous inputs),were employed to capture the dynamic behaviors of the processes understudy. Additionally, real data from oil production and drilling were utilized,considering the specific characteristics and operational challenges of theseenvironments. Based on the results obtained, the techniques used demonstratedapplicability and yielded satisfactory outcomes. Specifically, the use of hybridmodels, combining physical knowledge with multiple model approaches formedby system identification algorithms and machine learning, showed potentialfor enhancing simulation. These findings underscore the effectiveness of thesemethods, suggesting that future research could focus on implementing thistechnique in identifying complex systems.

Submitted by Sandra Pereira (srpereira@ufrrj.br) on 2017-04-06T12:02:08Z No. of bitstreams: 1 2012 - Rodrigo Rossi Rodrigues.pdf: 2827044 bytes, checksum: 1e0bfbc21aa38412c1823ea16ac2e2ed (MD5) / Made available in DSpace on 2017-04-06T12:02:08Z (GMT). No. of bitstreams: 1 2012 - Rodrigo Rossi Rodrigues.pdf: 2827044 bytes, checksum: 1e0bfbc21aa38412c1823ea16ac2e2ed (MD5) Previous issue date: 2012-03-27 / The goal of this study is to show how the resource?s rationalization can improve the operational results of one maritime transport company. The tanker maket follows a world tendency on the hiring price of charter. It makes the process of costs reducing through the optimization of fuel and diesel or/and through the performance?s control of time, velocity and operational disponibility a mandatory factor. The control of the consume of fuel and diesel is a way to reduce the discounts on the ships earns. The losses can be caused by technical problems on the MCP's (Main Combustion Engine), MCA's (Auxiliary Combustion Engine) and auxiliary machinery of the ships, or by the necessity of revaluation on the contractual price of loading and unloading products on shipping terminals and also by the climate changes in the Brazilian coastline, which causes some different conditions of velocity and time trips. / Atrav?s deste trabalho pretende-se apresentar como a racionaliza??o de recursos pode contribuir para melhorar o resultado operacional de uma empresa de servi?o de transporte mar?timo. O mercado de afretamento de navios petroleiros segue uma tend?ncia mundial na contrata??o do pre?o do frete, tornando importante o processo de redu??o de custos atrav?s da otimiza??o de ?leo combust?vel e ?leo diesel, e controle das performances operacionais: tempo, velocidade e disponibilidade operacional. Uma das formas de reduzir os descontos na receita dos navios ? o controle do consumo de ?leo combust?vel e ?leo diesel. Sejam eles provocados por problemas t?cnicos nos MCP?s (Motor de Combust?o Principal), MCA?s (Motor Auxiliar de Combust?o) e M?quinas Auxiliares dos navios, ou at? mesmo pela necessidade de reavalia??o de taxas contratuais de bombeio de carga e descarga de produtos nos terminais, e velocidade/tempo de viagem em alguns trechos do litoral brasileiro, devido ?s mudan?as clim?ticas e geogr?ficas.

The formation of modern society has been dominated by coal and oil, and together these two fossil fuels account for nearly two thirds of all primary energy used by mankind.  This makes future production a key question for future social development and this thesis attempts to answer whether it is possible to rely on an assumption of ever increasing production of coal and oil. Both coal and oil are finite resources, created over long time scales by geological processes. It is thus impossible to extract more fossil fuels than geologically available. In other words, there are limits to growth imposed by nature. The concept of depletion and exhaustion of recoverable resources is a fundamental question for the future extraction of coal and oil. Historical experience shows that peaking is a well established phenomenon in production of various natural resources. Coal and oil are no exceptions, and historical data shows that easily exploitable resources are exhausted while more challenging deposits are left for the future. For oil, depletion can also be tied directly to the physical laws governing fluid flows in reservoirs. Understanding and predicting behaviour of individual fields, in particularly giant fields, are essential for understanding future production. Based on comprehensive databases with reserve and production data for hundreds of oilfields, typical patterns were found. Alternatively, depletion can manifest itself indirectly through various mechanisms. This has been studied for coal. Over 60% of the global crude oil production is derived from only around 330 giant oilfields, where many of them are becoming increasingly mature. The annual decline in existing oil production has been determined to be around 6% and it is unrealistic that this will be offset by new field developments, additional discoveries or unconventional oil. This implies that the peak of the oil age is here. For coal a similar picture emerges, where 90% of the global coal production originates from only 6 countries. Some of them, such as the USA show signs of increasing maturity and exhaustion of the recoverable amounts. However, there is a greater uncertainty about the recoverable reserves and coal production may yield a global maximum somewhere between 2030 and 2060. This analysis shows that the global production peaks of both oil and coal can be expected comparatively soon. This has significant consequences for the global energy supply and society, economy and environment. The results of this thesis indicate that these challenges should not be taken lightly.

Oil is the black blood that runs through the veins of the modern global energy system. While being the dominant source of energy, oil has also brought wealth and power to the western world. Future supply for oil is unsure or even expected to decrease due to limitations imposed by peak oil. Energy is fundamental to all parts of society. The enormous growth and development of society in the last two-hundred years has been driven by rapid increase in the extraction of fossil fuels. In the foresee-able future, the majority of energy will still come from fossil fuels. Consequently, reliable methods for forecasting their production, especially crude oil, are crucial. Forecasting crude oil production can be done in many different ways, but in order to provide realistic outlooks, one must be mindful of the physical laws that affect extraction of hydrocarbons from a reser-voir. Decline curve analysis is a long established tool for developing future outlooks for oil production from an individual well or an entire oilfield. Depletion has a fundamental role in the extraction of finite resources and is one of the driving mechanisms for oil flows within a reservoir. Depletion rate also can be connected to decline curves. Consequently, depletion analysis is a useful tool for analysis and forecasting crude oil production. Based on comprehensive databases with reserve and production data for hundreds of oil fields, it has been possible to identify typical behaviours and properties. Using a combination of depletion and decline rate analysis gives a better tool for describing future oil production on a field-by-field level. Reliable and reasonable forecasts are essential for planning and nec-essary in order to understand likely future world oil production.

Made available in DSpace on 2014-12-17T14:08:36Z (GMT). No. of bitstreams: 1 AmandaGO.pdf: 1269225 bytes, checksum: bb3d37e917eb9a6a37d5196f82512218 (MD5) Previous issue date: 2010-01-27 / The objective of reservoir engineering is to manage fields of oil production in order to maximize the production of hydrocarbons according to economic and physical restrictions. The deciding of a production strategy is a complex activity involving several variables in the process. Thus, a smart system, which assists in the optimization of the options for developing of the field, is very useful in day-to-day of reservoir engineers. This paper proposes the development of an intelligent system to aid decision making, regarding the optimization of strategies of production in oil fields. The intelligence of this system will be implemented through the use of the technique of reinforcement learning, which is presented as a powerful tool in problems of multi-stage decision. The proposed system will allow the specialist to obtain, in time, a great alternative (or near-optimal) for the development of an oil field known / O objetivo da engenharia de reservat?rios ? o de gerenciar campos de produ??o de petr?leo de forma a maximizar a produ??o dos hidrocarbonetos obedecendo a restri??es f?sicas e econ?micas existentes. A defini??o de estrat?gias de produ??o ? uma atividade complexa por envolver diversas vari?veis do processo. Deste modo, um sistema inteligente, que auxilie na otimiza??o das alternativas de desenvolvimento do campo, torna-se bastante ?til no dia-a-dia dos engenheiros de reservat?rios. Este trabalho prop?e o estudo preliminar de um sistema inteligente de aux?lio ? tomadas de decis?es, no que diz respeito ? otimiza??o de estrat?gias de produ??o em campos de petr?leo. A intelig?ncia desse sistema ser? implementada por meio do uso da t?cnica de aprendizado por refor?o, a qual se apresenta como uma poderosa ferramenta em problemas de decis?o multi-est?gios. O sistema estudado visa permitir que o especialista obtenha, em tempo h?bil, a alternativa ?tima (ou quase-?tima) para o desenvolvimento de um campo de petr?leo conhecido