Contribute to the library of problems and models

To contribute to the library of MINLP and GDP problems (linear and nonlinear) through a new problem, new models to an existing problem, or new instances to a model of an existing problem, the guidelines for submitting the corresponding files are as follows (see an MINLP example problem here, see a GDP example problem here). Below we present example files for an MINLP model. Files for GDP models are similar. These can be submitted using the Contribute Problems tab. If you need help, or would rather send us the files so that we upload them for you, please send to minlp@andrew.cmu.edu

1. Problem statement (pdf). Describe precisely the application problem. (see an example file here)
2. Supply a session description file (pdf). Provide in one page a qualitative summary of the application problem, major features of the mathematical model and/or instances, and list solvers used. Include the title of the problem, authors, affiliations, email addresses. (see an example file here)
3. Model(s) (pdf).One or several alternative formulations of optimization model. Present the derivation of each model by introducing the corresponding notation. For uniformity, please use minimization of your objective function. Display the final model at the end of this section. Also, include relevant references. (see an example file here)
4. Input modeling files. (GAMS, AMPL, AIMMS input text files). Include title of problem, authors, affiliations, email addresses. Provide comments on the model, and use notation compatible with the one used in notation of section b) above. To make the equations more readable it would be helpful if the variables are in upper case and constants are in lower case. (see an example file here)
5. Output files (text file). Provide one output file corresponding to one of the solutions presented in section f. (see an example file here)
6. Results and discussion (pdf). Present two sections: computational results and interpretation of solutions. For the former provide version of modeling system used, version of optimization solvers used, choice of values of special parameters (eg optimality tolerance). Present results in tables in terms of objective function value, problem size, CPU time and relevant indicators (eg number of nodes, relaxation, number iterations). Present comparisons among various solvers if appropriate. For interpretation of solution present diagrams, plots or qualitative descriptions of solution obtained for application problem. Note: Place your data in an Appendix of the Results file. Values of parameters of the instance (s) solved for the model(s). If this file is not provided, describe in the model file. Also, try to supply data to create models of varying size. (see an example file here)