No abstract

* Corresponding author / autor para quem as correspondências devem ser encaminhadas Recebido em 07/2002, aceito em 12/2002 após 1 revisão ResumoO principal objetivo deste trabalho é avaliar a complementação do uso de técnicas de previsão de vendas com Redes Neurais Artificiais. Como aplicação, foi analisada a demanda industrial no setor de cerveja da Austrália. A maioria dos modelos de previsão atuam de forma isolada, ou seja, tratando problemas por enfoques que se excluem. A sugestão deste trabalho é utilizar estes métodos de forma cooperativa, buscando alcançar melhores resultados que aqueles obtidos individualmente. O trabalho inicia-se com considerações sobre previsões de séries temporais, caracterização breve sobre redes neurais e outros métodos de previsão. A seguir, o estudo de caso: uma série histórica de vendas de cerveja na Austrália. Encerra-se com uma conclusão e recomendações, e lista de referências bibliográficas.Palavras-chave: previsão de demanda, redes neurais artificiais, cooperação. AbstractThe main goal of this article is to evaluate the complementation between forecasting techniques, in particular Neural Networks. The environment of the application is a foreign industrial sector, sales of beer in Australia. Most of the forecasting models act in a separate way, that is, treating problems through different views that exclude one another. The suggestion of this work is to use the methods in a cooperative way, looking forward to achieve better results. The article begins with some considerations on Time Series Studies, followed by a brief characterization of Neural Networks and the other methods used on this work. Continuing, the case is presented and discussed and finishing, conclusions and a list of references.

In this paper we propose a DEA (Data Envelopment Analysis) based ranking for Brazilian soccer teams. Ranking is a major issue for the soccer clubs, either when looking for sponsorship or expanding their supportive group. We first include value judgements, applying a method to consolidate the results of the national and international matches. Then we use both aggregate results as outputs for applying the DEA model. Rankings published by the Brazilian Soccer Confederation (CBF) and 'Placar', a traditional sports magazine, do not give any credit to international titles and national cups, focusing only on the results in the National League. We compare the DEA ranking with the Federation's and the magazine's, raising some controversial issues and drawing unexpected conclusions.Keywords: soccer in Brazil; data envelopment analysis; decision aid. ResumoNeste artigo é proposto um ranking baseado em Análise Envoltória de Dados (DEA) para os times de futebol brasileiros. Ranking é uma questão importante para os clubes, quer seja quando procuram patrocínio ou expansão da torcida. Primeiramente incluídos julgamentos de valor, utilizando um método para consolidar resultados de torneios nacionais e internacionais. Em seguida usamos ambos agregados como outputs para a aplicação do modelo DEA. Os rankings da Confederação Brasileira de Futebol (CBF) e da tradicional revista Placar, por exemplo, não dão nenhum crédito a títulos internacionais e nacionais, focalizando apenas no resultado da Liga Nacional. O ranking obtido via DEA é comparado com os da CBF e da revista, o que ocasiona certa controvérsia e conclusões inesperadas.Palavras-chave: futebol no Brasil; análise envoltória de dados; apoio à decisão.

In Brazil, for the last four decades, Petrobras, a state-owned company, had retained a monopolistic concession for the exploration of Oil & Gas prospects, when, in 1998, the law was changed and a broad range of opportunities were created. Since then, Petrobras has been offering a series of Join Venture opportunities for foreign oil companies to farm in. Presently, a number of international oil companies installed offices locally and started to participate in the upstream sector. ANP, the National Petroleum Agency, is offering for tender various blocks both offshore and onshore. The first bid round was a major success, in mid 1999, the second bid round being scheduled for the third quarter of year 2,000. Blocks leased in the first bid round are now being explored by means of 2D and 3D seismic, and the first exploratory wells will be soon drilled. This article presents a framework to perform an economic evaluation of oil & gas prospects, taking into account the risk factor. The Net Present Value of each block is first computed based upon an average value. Following that, Monte Carlo simulation is used to compute the downside risk. The inputs are determined by means of probability distributions. Reservoir parameters, capital and operating costs and oil prices are fed into the program, revealing the NPV sensitivity to each factor. The problem is further structured by means of decision trees. The Expected Monetary Value, based upon perceived probabilities of success, is estimated, utility functions being used to account for risk aversion. The optimum level of participation in each prospect is determined, using the so-called Certainty Equivalent concept. Finally, the formation of an optimum portfolio is discussed, so that any oil company can establish its strategy in joint bids and farm ins. Four hypothetical prospects are evaluated; with different levels of investment and risk profiles, namely, from higher probability of success to new frontier, lower probability of success. The correspondence with real life may be, respectively, the Campos Basin, where 70% of Brazilian oil production comes from, and basins like Pelotas, where there are no producing wells, with little seismic available. The framework suggested in this work could be useful for any company that is currently involved with exploration activities. Additionally, it may be used for analyzing any investment with a certain degree of uncertainty or volatility.

The paper uses a case study approach to present the challenges to develop a large and thick oil carbonate reservoir, full of opportunities but also of uncertainties. Additionally, Libra block development is under a Production Share Contract that was award to a Consortium where Petrobras is the operator in partnership with Shell, Total, CNOOC and CNPC. The paper will briefly present the de-risking plan and detail the EWT Program implementation and the way it is helping the full field development, which consists of four mega ultra deepwaters projects.