Couenne

The NEOS Server offers Couenne for the solution of mixed-integer nonlinearly-constrained (MINLP) and global optimization problems. Problems can be submitted to Couenne on the NEOS server in AMPL or GAMS format.

Couenne (Convex Over and Under ENvelopes for Nonlinear Estimation) is a branch-and-bound algorithm to solve mixed integer nonlinear programs (MINLPs) and global optimization problems. Couenne aims to find global optima of nonconvex MINLPs. It implements linearization, bound reduction, and branching methods within a branch-and-bound framework. Its main components are:

• an expression library;
• separation of linearization cuts;
• branching rules;
• bound tightening methods.

Couenne was developed and is currently maintained by Pietro Belotti and others. More details on the solver are available on the Couenne page at COIN-OR.

Using the NEOS Server for Couenne/GAMS

The user must submit a model in GAMS format to solve an optimization problem. For security purposes, the model submitted must adhere to the following conventions:

• It must be self contained, i.e., no $include or $batinclude statements.
• It may not execute external programs, i.e., no $call or execute statements.
• No file creation, i.e. no put files or $echo except for a file named 'results.txt'. Other files created in the GAMS model file will be deleted.

If you are unfamiliar with GAMS, the GAMS Documentation includes a GAMS Tutorial and User's Guide. Examples of models in GAMS format can be found in the GAMS model library.

By default, the NEOS Server limits the amount of output generated in the listing file by turning off the symbol and unique element list, symbol cross references, and restricting the rows and columns listed to zero. This behavior can be changed by specifying the appropriate options in the model file. See the documentation on GAMS output for further information.

You may optionally submit an options file if you wish to override the default parameter settings for the solver. Currently, the NEOS Server can only use optfile=1 with GAMS input. Therefore, any model that specifies a different options file will not work as intended.