Decision Optimization

Decision Optimization

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


#Analytics
#DecisionOptimization
#DecisionOptimization
 View Only
  • 1.  combinatorial benders cut (conflict refiner cplex)

    Posted 11/27/13 01:47 AM

    Originally posted by: blessandguess


    I am  implementing combinatorial benders cut (conflict refiner cplex) for  my optimization problem and coding in c/c++.I am facing two problems.first  when you get solution from the master problem how to do fix the solution to slave problem to check for feasibility. also after you generate the cut for  the master problem how do you add the constraint to the master problem and resolve(maybe callback function).I can upload my code if you want to see it


    #CPLEXOptimizers
    #DecisionOptimization


  • 2.  Re: combinatorial benders cut (conflict refiner cplex)

    Posted 11/27/13 10:09 AM

    Originally posted by: EhsanN


    1) Depending on when you're adding benders cuts, you should either use the user cut callback (i.e., adding cuts to fractional nodes) or lazy constraint callback (i.e., adding cuts to integer nodes). You could also use both callbacks.

    2) To get the solution of node and fix it for the SP, you should define a IloNumArray object within the callback and set its value equal to the desired part of the node solution. To do this, you should use the getValues method.

    3) All the necessary operations (e.g., retrieving the solution, deriving the cut, and adding the benders cuts) is done within the callback.

    To better understand the process, you might take a look at the ilobendersatsp.cpp example in your CPLEX installation folder. You could find some useful info using the links below:

    http://pic.dhe.ibm.com/infocenter/cosinfoc/v12r5/index.jsp?topic=%2Filog.odms.studio.help%2FCPLEX%2FReleaseNotes%2Ftopics%2Freleasenotes123%2Freleasenotescplex123_22.html

    http://orinanobworld.blogspot.com/2012/07/benders-decomposition-in-cplex.html

     

     


    #CPLEXOptimizers
    #DecisionOptimization