Pietro Belotti is one
of the developers of the FICO-Xpress optimization suite, one of the
leading solvers of optimization problems. His research interests lie
in Mixed Integer Nonlinear Programming, Robust Optimization, and
Network Optimization. Prior to joining FICO in 2013, he worked at
Carnegie Mellon University, Lehigh University, and Clemson
University. He collaborates with researchers in several universities
on topics including bound reduction in optimization problems,
solution methods for multiobjective optimization, optimal pedigree
selection in forest management, and others.
TITLE OF TALK: Modeling and optimization of large-scale
industrial problems
ABSTRACT: Solving practical optimization problems
in industry requires sophisticated software that can scale to
possibly millions of variables and constraints. Leading commercial
optimization software packages provide robust, fast solvers, but
also tools for simulation, visualization, and modeling, for a more
streamlined path from problem to solution. I will briefly present
the features of one such software package, the Xpress optimization
suite, and discuss in depth some of its main optimization
algorithms, as well as other parts of the optimization stack.