The GNU Linear Programming Kit (GLPK) is an open-source project for solving optimization problems (including LP and mixed-integer programs). This installation accepts optimization models expressed in the GNU MathProg modeling language (a subset of the AMPL language) and the glpsol solver for defining and solving small instances of such problems for classroom use.

Model input. On the GLPK page of this OR Toolkit site,

  1. Enter a title to be displayed in the report heading
  2. Enter the model in the GNU MathProg language in the text area
  3. Select those processing reports to be displayed:
    • Input model: an echo print of the model as entered above
    • Generated model: the model created from the MathProg description, with all constraints expanded, in Cplex format
    • Solver statistics: messages from the glpsol solver as the model is processed and solved
    • Final solution: report on optimal values of solution variables
    • Sensitivity: sensitivity and ranging report
  4. Press Solve model button to initiate execution and creation of reports

Resources available for learning GMU MathProg include:

  • GNU MathProg Language Reference (HTML, PDF)
  • Example problems, which can be copied and pasted into the GLPK input form (select file INDEX for short descriptions)

Debugging your model

  • If nothing follows the “GLPK output:” heading, check the Solver statistics display checkbox to view the glpk errors
  • Execution time limit for solver is 10 cpu seconds
[Home] [LP] [BLP] [Mr. T] [Mr. Pivot] [GLPK] [Networks] [Decision Analysis] [Stochastic] [Forecasting]