Decision Optimization

Originally posted by: SystemAdmin

Hi:

I've been trying to implement an inventory optimization model in CPLEX. Unfortunately I have problems with the index ranges... The model contains two sets of indexes (stages and periods) and I have to sum up over both of them.
Also, I sometimes want to include variables in the index ranges. So for example I might want to sum up over all periods starting with the current period and ending with the period that equals the transportation lead time of the stage. I don't really know how to do this in OPL as I don't know the right syntax. I've been trying to find out how this works for a while but I don't really find anything in the documentation or elsewhere.

I hope you can help me. I would be so grateful!

Thanks,
Carolin
#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: SystemAdmin

data file
#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi,

first of all, in the OPL language the equality constraint is not = but ==.

I changed a little your code in order to run it.
This can be a start for you in order to change it to what you need.

Regards

using CP;   

int Nbstages =...; range stages = 1..Nbstages; 

int Nbperiods =...; range periods = 1..Nbperiods+1; 

int holdingcost [stages]=...; 

int backordercost=...; 

int WIPtransportleadtime [stages]=...; 

int Demand [periods]=...; dvar int+ qtyordered [stages] [periods]; dvar int+ qtybackordered [stages] [periods]; dvar int+ WIPintransit [stages] [periods]; dvar int+ WIPwaiting [stages] [periods]; dvar int+ initialinventory [stages] [periods]; dvar int+ finalinventory [stages] [periods]; dvar int+ basestock [stages]; minimize sum (j in stages:j!=1) ((holdingcost[j]* sum (t in 1..Nbperiods+1)finalinventory [j][t]))+ backordercost* sum(t in 1..Nbperiods+1)(qtybackordered[1][t]); subject to 
{ forall (j in stages:j!=1) forall (t in periods:t==1) initialinventory [j] [t]>=0; forall (j in stages:j!=1) forall (t in 2..WIPtransportleadtime[j]: t in periods) initialinventory [j][t] == finalinventory[j][t-1]; forall (j in stages:j!=1) forall (t in 2..Nbperiods+1: (t in periods) && (t>=WIPtransportleadtime[j]+1)) ((t>=WIPintransit[j][t]+1) => (initialinventory [j][t] == finalinventory [j][t-1]+WIPintransit[j][t-WIPtransportleadtime[j]])); forall (j in stages:j==1) forall (t in 1..Nbperiods+1) initialinventory [j][t]==Demand [t]; forall (j in stages) forall (t in 1..Nbperiods+1) finalinventory [j][t] == initialinventory [j][t]-qtyordered[j][t]; forall (j in stages) forall (t in 1..Nbperiods+1) qtyordered [j][t]<=initialinventory [j][t]; forall (j in stages) forall (t in periods:t==1) qtyordered [j][t]<=Demand [t]; forall (j in stages) forall(t in periods:t==1) qtybackordered [j][t]==Demand[t]-qtyordered[j][t];   forall (j in stages) forall(t in 2..Nbperiods+1) qtyordered [j][t]<=Demand[t]+qtybackordered[j][t-1];   forall (j in stages) forall(t in 2..Nbperiods+1) qtybackordered[j][t]==Demand[t]+qtybackordered[j][t-1]-qtyordered[j][t]; forall (j in stages:j!=Nbstages) forall (t in periods) WIPintransit [j][t]==qtyordered[j][t];   
} ;

#DecisionOptimization
#OPLusingCPLEXOptimizer