Decision Optimization

Decision Optimization

Delivers prescriptive analytics capabilities and decision intelligence to improve decision-making.

 View Only
Back to discussions
Expand all | Collapse all

Finding good feasible solution in MIP in range of certain value

  • 1.  Finding good feasible solution in MIP in range of certain value

    Posted Wed November 23, 2011 11:25 AM

    Originally posted by: nghe


    Hello,
    I am implementing the code solving Min mip and I have a (const) certain value z0 > Optimal value of min problem. Now I just want to find a good feasible solution of mip so that its objective value is less than or equal to 95%z0 (let assume the percentage like this) and quit solving in cplex. If I cannot find a feasible solution like it, I need to solve min problem upto optimality.

    I know that we can use Callback to do that but I donot want to lose the dynamic search and parallel computation.

    Do you know any parameter in cplex so that I can set the range of objective value of a feasible solution with a certain upperbound?

    Thank you,

    nghe
    #CPLEXOptimizers
    #DecisionOptimization


  • 2.  Re: Finding good feasible solution in MIP in range of certain value

    Posted Wed November 23, 2011 01:09 PM

    Originally posted by: SystemAdmin


    Just use an info callback to check your abort criterion. This is compatible with dynamic search.

    Please take a look at the "mipex4.c" example (or ilomipex4.cpp or similar examples for other APIs), which explains how to use the info callback.
    Tobias
    #CPLEXOptimizers
    #DecisionOptimization


  • 3.  Re: Finding good feasible solution in MIP in range of certain value

    Posted Wed November 23, 2011 02:33 PM

    Originally posted by: nghe


    Thank you so much. I will see .

    Nghe
    #CPLEXOptimizers
    #DecisionOptimization


Related Content