Vinaora Nivo Slider 3.xVinaora Nivo Slider 3.xVinaora Nivo Slider 3.x
Welcome to ORCS Lab
Welcome to the Laboratory of Operational Research and Complex Systems. Feel free to explore the website and get to know our projects and the people behind the algorithms.
ORCS Lab - Optimisation, Data Science and Complex Systems
Research in the ORCS Lab focuses mainly on the development and application of optimisation methods to complex systems; on data science and statistical modeling; and on the applications of computational intelligence to large-scale power and energy systems.

Latest News

Paper accepted: ITOR

We're happy to announce that our paper A Recombination-based Matheuristic for Mixed Integer Programming Problems with Binary Variables, co-authored by André Maravilha, Eduardo Carrano and Felipe Campelo, has just been accepted at the International Transations on Operations Research (IF: 1.745). It was under review for almost two years! Congratulations to all authors for their great work (and resilience)!

Click below for the abstract.

Read more: Paper accepted: ITOR

Paper published: Expert Systems with Applications

Now that's some good news! André Batista has just published his very first journal paper, at Elsevier's Expert Systems with Applications (IF: 3.928), vol. 99, pp. 180-192, 2018. The paper, titled Demand Side Management Using a Multi-Criteria epsilon-Constraint based Exact Approach, was co-authored by André and his father uncle advisor, Prof. Lucas S. Batista.

The paper is already available online, and is under an early open-access window until early March. Go take a look, or click below for the abstract.

Read more: Paper published: Expert Systems with Applications

ExpDE 1.4 now on CRAN

The second release version of R package ExpDE, a standardised modular implementation of Differential Evolution for the experimental investigation of operators, was released today on CRAN: This version includes some minor bug fixes and features two new differential mutation variants.

Package CAISEr v. 0.2.1 released!

The first release version of R package CAISEr, which implements a methodology for sample size calculations in experimental comparisons of algorithms, was released today. The package is available from CRAN, and can be installed from the R prompt using install.packages("CAISEr").

This package is the result of work by ORCS members Prof. Felipe Campelo and Fernanda Takahashi. The full description of the methodology and results generated with this package were submitted as a research paper to the Journal of Heuristics.

Best paper award - ENIAC 2017

We're very pleased to report that our paper "A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition", by Daniel de Souza, Fillipe Goulart, Lucas Batista and Felipe Campelo was selected as one of the top 3 papers at ENIAC2017 - Encontro Nacional de Inteligência Artificial e Computacional (National Meeting on Artificial and Computational Intelligence), currently happening in Uberlândia/MG, Brazil. Daniel presented the paper, which was written as part of his M.Sc. work, this last afternoon. Congratulations!

You can read the abstract by clicking "Read More" below.

Read more: Best paper award - ENIAC 2017

Fernanda's Qualification Exam

We're happy to invite everyone to Fernanda Takahashi's Ph.D. qualification exam. Her work Sample Size Estimation For Power And Accuracy In The Experimental Comparison of Metaheuristics will be presented on May 31st 2017 2 p.m., at the Seminar Room T005. Besides Fernanda's advisor, ORCS Lab's Prof. Felipe Campelo, the qualification committee will be composed by:

  • Prof. Thiago Noronha, DCC-UFMG
  • Prof. Luiz Duczmal, Dep. Statistics - UFMG
  • Prof. Helio Barbosa, LNCC

Click below for the abstract.

Update: click here for pictures!

Read more: Fernanda's Qualification Exam

April Defense Madness

In the next two weeks we'll be having four (FOUR!) defenses by ORCS Lab's students! All students are invited as usual, and the academic community in general is always very welcome as well!

Check below for details.

Read more: April Defense Madness

MOEADr package 0.2.1 released

The first release version of R package MOEADr, a standardised modular implementation of Multiobjective Evolutionary Algorithms based on Decomposition (MOEA/D), was released today. Version 0.2.1 contains several variants of decomposition strategies, aggregation functions, variation operators, neighborhood assignment rules, update strategies and more. It also integrates seamlessly with package smoof, which provides standard implementations of several benchmark sets for multiobjective optimization.

This package is the result of a joint project between ORCS Lab's Felipe Campelo and Lucas Batista and University of Tsukuba's Claus Aranha, and is part of an upcoming paper on component-wise modeling of MOEA/Ds (stay tuned!).

You can install the package from the R console by typing