Filipe Brandão graduated from the Faculdade de Ciências of Universidade do Porto with a master’s degree in computer science in 2012. His master’s thesis was entitled “Bin Packing and Related Problems: Pattern-Based Approaches” and one of its outcomes was the initial version of project VPSolver. In 2012, he started a PhD in computer science, which he is currently finishing, at the Faculdade de Ciências of Universidade do Porto under the supervision of João Pedro Pedroso. His research focus mainly on the application of very strong integer programming models based on arc-flow formulations to the solution of a large variety of cutting and packing problems. Those models are based on the general arc-flow formulation with graph compression, proposed in Brandão and Pedroso (2013), which is produced by VPSolver.
VPSolver – a vector packing solver based on an arc-flow formulation with graph compression.
VPSolver generates very strong models (equivalent to Gilmore and Gomory 1961’s) that can be solved using general- purpose mixed-integer programming solvers such as Gurobi and GLPK. For modelling more complex problems easily, VPSolver also includes a Python API (pyvpsolver), a mod- elling toolbox (PyMPL) and a Web App.