Projects

Data mining and processes to support decision and automation

ExACTa is an initiative by PUC-Rio for research, agile experimentation, and co-creation for digital transformation. The initiative was created by four professors from the Informatics Department, research in the areas of Data Science, Software Engineering, Optimization, and Human-Computer Interaction. The initiative’s vision is to strengthen the links between research and development, allowing the application of research results to trigger innovation and operational excellence in partner companies. Among the differentials of the initiative are the strong integration with cutting-edge research being developed in the Department of Informatics at PUC-Rio (CAPES maximum concept in research in the field of computing, 7), the own and innovative process for the co-creation of research solutions and development with agility, a highly qualified team and an environment specially designed for this purpose. The ExACTa initiative allows to apply in practice research results generated both in the context of the initiative and in thematic laboratories of the Department of Informatics, such as DASLAB, GALGOS, IDEIAS and LES. With these differentials, ExACTa sets up the ideal initiative for the digital transformation of partner companies.

Stochastic Optimization Models for Short-Term Oil Production

The main challenge of this project is to provide engineers with a decision support system to maximize oil production while minimizing flaring and respecting the constraints imposed by the wells processing plant and burn limit. The objective of this project is to propose, develop and test the use of the mathematical optimization models, using more advanced uncertainty optimization techniques to incorporate into the decision process the uncertainty inherent in the production of offshore platforms using the continuous gas lift. This new model is expected to be a more reliable solution considering the trade-off between the expected increase in oil production. This is a joint work with Petrobras, PUC-Rio (LAMPS and Tecgraf and UAI.

Incorporating the effect of climate variability and contingencies in the optimal contracting strategy of transmission-usage amounts

The main goal is to develop a computational tool capable of accounting for the uncertainty due to contingencies and climate variability in the optimal transmission-usage contracting strategy for a distribution company with high integration of renewable-distributed generation.Energisa

Research in credit card fraud and churn detection

The purpose of this project was to build a tool using state-of-the-art methods for fraud and churn detection on credit card machines. In this project we have to deal with a huge amount of transaction data, it was a challenge to process all data and to construct a machine learning that can deal with this.Stone