Maximize production with new NAG Library Mixed Integer Nonlinear Programming algorithms

New in the NAG Library - Mixed Integer Nonlinear Programming improves business planning, scheduling, production, design and portfolio optimization

September 9, 2015 - The Numerical Algorithms Group (NAG), numerical software, engineering and HPC service provider, announce new algorithms in the NAG Library for Mixed Integer Nonlinear Programming. The underlying algorithm is a modified Sequential Quadratic Programming (SQP) stabilised by using trust regions and can deal with both convex and nonconvex problems and problems with possibly expensive function evaluations. In addition, it is not assumed that the mixed integer problem has to be relaxable; the function evaluations are requested only at integral points. This is a distinctive feature of the solver since the usual approaches rely on the relaxation of the discrete variables.

Examples of Mixed Integer Nonlinear Programming can be found in many areas including:

  • Portfolio optimization
  • Design of distribution networks
  • Flight path / route / scheduling optimization
  • Optimal response to catastrophic oil spills
  • Protein folding

The NAG Library is tried and trusted by users for its quality, depth of coverage, support and documentation. Available for multiple programming languages, environments and platforms that NAG Library is embedded in thousands of applications all over the world. 30 day trials are available on request.