Decision Optimization

Decision Optimization

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


#Analytics
#DecisionOptimization
#DecisionOptimization
 View Only
  • 1.  Extension of Basic Parallel Machine Scheduling CP Model with multiple period.

    Posted 12/05/13 09:04 AM

    Originally posted by: Cagri


    Hello,

    I am trying to write a CP to deal with standart parallel machine scheduling problem with sequence dependent setup times. And we have a single machine operator following this parallel process (e.g. Incubation Operation). It will be a hybrid methodology combined with MILP dealing with planning issues. I built the code already working properly without d index of day explaining each planning day. To use in this hybrid method, I would like to extend the basic model to make schedules for each day of planning horizon, simultaneously.When I add d index of Day to sequence variable, I get some errors.

    Could you please help me to overcome this challenge?

    Question Marks

    • I am also not clear whether it is possibe to add d index of Day to sequence dvar in this concept or not?
    • Do I need to call the same CP model for each planning period in a recursive manner using a loop?
    • Which alternative can provide computational efficiency?


    Note: Errors!, Scheduling model and data file is attached in this massage. (Errors! & OPL CP Mod, Dat, Ops Files)

    Cheers!

    Cagri SEL
    cagrisel@gmail.com
     


    #DecisionOptimization
    #OPLusingCPOptimizer


  • 2.  Re: Extension of Basic Parallel Machine Scheduling CP Model with multiple period.

    Posted 12/09/13 06:42 AM

    Originally posted by: ol


    Hello,

    replace your declaration of sequence by:


    dvar sequence schedule[d in Days, m in Machines] in all(d in Days, j in Jobs) opttask[d][j][m] types all(d in Days, i in Jobs) T[i];

     

    and this will work.

    Regards,

    Olivier


    #DecisionOptimization
    #OPLusingCPOptimizer


  • 3.  Re: Extension of Basic Parallel Machine Scheduling CP Model with multiple period.

    Posted 12/09/13 07:26 AM

    Originally posted by: Cagri


    Thanks lot for your kindly interest. It works.

    But, I could not clearly understand the definition and usage of T variable. What is the fundamental mission of this variable?

    int T[Jobs]=...; 

    types all(d in Days, i in Jobs) T[i];


    #DecisionOptimization
    #OPLusingCPOptimizer


  • 4.  Re: Extension of Basic Parallel Machine Scheduling CP Model with multiple period.

    Posted 12/11/13 03:45 AM

    Originally posted by: ol


    Hello,
    The type is used to specify setup times between two operations executed on the same resource. If an operation 1 of type A is followed by an operation 2 of type B, a matrix (Setup in your model) expresses the setup times between type A and type B.
    In your model this is done by the constraint:
    forall(d in Days, m in Machines)
        noOverlap(schedule[d][m], Setup);
    You can find a complete explanation in the user man of CPO, see "Modeling sequence-dependent setup times".
    Regards,
    Olivier


    #DecisionOptimization
    #OPLusingCPOptimizer


  • 5.  Re: Extension of Basic Parallel Machine Scheduling CP Model with multiple period.

    Posted 12/11/13 03:47 AM

    Originally posted by: Cagri


    Thank you very much. 


    #DecisionOptimization
    #OPLusingCPOptimizer