Parameter sets

Explains how to use parameter sets with multiobjective optimization.

During the solution process, you may prefer that optimization problems with different priorities are solved with different parameter settings. CPLEX makes this possible by using parameter sets. If parameter sets are passed to the CPXmultiobjopt() function (or the corresponding functions in the other APIs), then those sets are used one-by-one in the optimizations.

Using a parameter set means that, with each subproblem, first all parameters are set to their default values and then the parameters listed in the set are changed to the values specified in that set.

There are a few exceptions. The parameters in the environment that set a global limit (time limit, deterministic time limit, etc.) act as a limit on the cumulative resource use of the sequence of optimizations.

The following two parameters also do not reset to their default values before applying the new parameter set with each subproblem:

  • CPXPARAM_Parallel: since parallel mode cannot be changed during the algorithm, any parallel mode settings in the parameter sets will be disregarded. See parallel mode switch.
  • CPXPARAM_ScreenOutput: the value in the master environment will be used in all subproblems, except in those where the parameter set overrides it. See messages to screen switch
Note: After optimization the parameters in the environment are reset to whatever values they had before the optimization.

For more information about using parameter sets, see Parameter sets in the CPLEX Callable Library (C API) Reference Manual.

For a list of examples using parameter sets with multiobjective optimization, see Examples for multiobjective optimization.