Solving Mixed Integer quadratic and conic optimization problems Event
- Time:
- 16:00 - 17:00
- Date:
- 20 October 2016
- Venue:
- Room 3041 Building 2, Southampton Business School
For more information regarding this event, please email Dr Yuan Huang at yuan.huang@soton.ac.uk .
Event details
Abstract: Modern optimization software packages are capable of solving large optimization problems, as long as the functions involved are linear, even if some of the variables are integer---the class known as mixed integer programming, or MIP. However, size is still an issue in problems with a quadratic objective function or quadratic constraints, even if convex. Yet this class of problems, named mixed integer quadratically constrained programming, or MIQCP, admits numerous practical applications: from finance to chemical engineering, from logistics to facility location. We show some of the techniques that have proven successful for solving efficiently the problems in this class, with a special focus on the subclass in which the quadratic constraints are second-order cones, or Lorentz cones.
Speaker information
Dr Pietro Belotti ,FICO, Birmingham,Pietro Belotti received a PhD from the Technical University of Milan in 2003. He is a member of the team that develops the FICO Xpress-Optimizer, one of the leading solvers for linear and nonlinear optimization problems. He previously worked at Carnegie Mellon University, Lehigh University, and Clemson University. His main research interests lie in global optimization, quadratically constrained optimization, and optimization under uncertainty.