Modelagem de dados de falhas de equipamentos de sub-superfície em poços de petróleo da Bacia Potiguar

This master´s thesis presents a reliability study conducted among onshore oil fields in the Potiguar Basin (RN/CE) of Petrobras company, Brazil. The main study objective was to build a regression model to predict the risk of failures that impede production wells to function properly using the information of explanatory variables related to wells such as the elevation method, the amount of water produced in the well (BSW), the ratio gas-oil (RGO), the depth of the production bomb, the operational unit of the oil field, among others. The study was based on a retrospective sample of 603 oil columns from all that were functioning between 2000 and 2006. Statistical hypothesis tests under a Weibull regression model fitted to the failure data allowed the selection of some significant predictors in the set considered to explain the first failure time in the wells

Contribuições da pesquisa operacional para gestão da produção e operações : uma análise exploratória da literatura

This research aims to investigate the evolution presented during three decades (1980, 1990 and 2000) of using the tools of Operations Research (OR) as a suport to decision making in Production Operation Management (POM). Hypothesis tests were made to verify the proportional growth of a given area over the decades to the detriment of the areas of facility layout, capacity planning, production scheduling and inventory management. Six journals were selected and from them more than 800 articles were used for classification and analysis in the grounds of review. It also discussed possible ways for future research and comparisons are made with other papers of literature review. As a result, it was found that areas of heuristics and simulation showed a greater quantity of contributions in all POM areas of this study

Modelo matemático para otimização dos custos de um sistema híbrido eólico-solar : aplicação no bombeamento de petróleo

This work purposes the application of a methodology to optimize the implantation cost of an wind-solar hybrid system for oil pumping. The developed model is estimated the implantation cost of system through Multiple Linear Regression technique, on the basis of the previous knowledge of variables: necessary capacity of storage, total daily energy demand, wind power, module power and module number. These variables are gotten by means of sizing. The considered model not only can be applied to the oil pumping, but also for any other purposes of electric energy generation for conversion of solar, wind or solar-wind energy, that demand short powers. Parametric statistical T-student tests had been used to detect the significant difference in the average of total cost to being considered the diameter of the wind, F by Snedecor in the variance analysis to test if the coefficients of the considered model are significantly different of zero and test not-parametric statistical by Friedman, toverify if there is difference in the system cost, by being considered the photovoltaic module powers. In decision of hypothesis tests was considered a 5%-significant level. The configurations module powers showed significant differences in total cost of investment by considering an electrical motor of 3 HP. The configurations module powers showed significant differences in total cost of investment by considering an electrical motor of 5 HP only to wind speed of 4m/s and 6 m/s in wind of 3 m, 4m and 5 m of diameter. There was not significant difference in costs to diameters of winds of 3 m and 4m. The mathematical model and the computational program may be used to others applications which require electrical between 2.250 W and 3.750 W. A computational program was developed to assist the study of several configurations that optimizes the implantation cost of an wind-solar system through considered mathematical model

Testes de hipóteses em modelos de sobrevivência com Fração de cura

Survival models deals with the modeling of time to event data. However in some situations part of the population may be no longer subject to the event. Models that take this fact into account are called cure rate models. There are few studies about hypothesis tests in cure rate models. Recently a new test statistic, the gradient statistic, has been proposed. It shares the same asymptotic properties with the classic large sample tests, the likelihood ratio, score and Wald tests. Some simulation studies have been carried out to explore the behavior of the gradient statistic in fi nite samples and compare it with the classic statistics in diff erent models. The main objective of this work is to study and compare the performance of gradient test and likelihood ratio test in cure rate models. We first describe the models and present the main asymptotic properties of the tests. We perform a simulation study based on the promotion time model with Weibull distribution to assess the performance of the tests in finite samples. An application is presented to illustrate the studied concepts