What Is OPTI?

OPTimization Interface (OPTI) Toolbox is a free MATLAB toolbox for constructing and solving linear, nonlinear, continuous and discrete optimization problems. A range of open source and academic solvers are supplied for the Windows user - no compilation required!

The toolbox is supported with demos, detailed documentation and implementation experience. The toolbox is released under the BSD 3-Clause License.

What can I do with OPTI?

A brief outline of OPTI functionality is listed below. For more details, see the Examples section.

  • Construct optimization problems in MATLAB and solve them using a range of supplied solvers
  • Automatically detects the problem type being solved and uses the best solver available for the problem
  • Use powerful open source solvers such as IPOPT, SCIP, NOMAD and others
  • Perform solver benchmarking, plot optimization contours and validate the solution
  • Read and write standard optimization files such as MPS and LP
  • Read and solve AMPL problems

A Typical User

A typical OPTI user will be:

  • Comfortable using MATLAB (student level is fine)
  • Understand the basics of optimization (how to define an objective and apply constraints)
  • Working in Windows (sorry, no Linux or Mac support)
  • Looking for a free optimization package

You may also be familiar with some of the following statements / questions which OPTI aims to solve:

  • I'd like to use (insert solver here), but I'm not a strong C/C++ programmer
  • I'm an OK programmer, but I just can't work out how to compile (insert solver here)
  • Will a different solver give a better solution to my problem?
  • Can I try (insert solver here) without changing my problem?
  • I'd like to solve Mixed Integer problems in MATLAB
  • My problem is taking a long time to solve. Is there a faster solver which is still robust?

Solvers

The main contribution of OPTI is providing the solvers compiled, with MEX interfaces. This allows a MATLAB user to call them just like any other MATLAB function.

All supplied solvers have been compiled from open source code obtained via the links listed in the Solvers page. Where possible, all authors are aware of the inclusion in OPTI, and have OK'd this work. However if you use a solver in your research, please make sure you cite the original solver reference (and OPTI if you are feeling generous).

If you are using a solver in any commercial work please check the individual license restrictions of the solver.

Further Reading

My thesis also contains a huge amount of information on the development of OPTI , its motivation and some interesting case studies.