Originally posted by: Arnautarrago

Hi everyone,

I'm having an error when trying to run a cplex code. Basically the error says that the index is out of bound for the array x. 

Please find attach the code as well as the data. 

Thanks in advance.

 

#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi,

you forgot to attach the .mod

regards

https://www.linkedin.com/pulse/making-decision-optimization-simple-alex-fleischer/

#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: Arnautarrago

Hi Alex,

I don't know but i can't upload the model itself, i only can upload the project file.

If you can't see the model i copy it here:

#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi,

you have many out of bounds.

If you write

 

    

     using CP;
 
// NETWORK+PARAMETERS
 
int trucks=...; // set of trucks
range truck= 1..trucks;
 
int times=...; // set of time windows
range time=1..times;
 
int capacity [truck]=...; // capacity of a truck
 
int WBs=...; // working bench (WB1,WB2)
range WB= 1..WBs;
 
int OCs=...; // ore crusher (OC1,OC2)
range OC= 1..OCs;
 
int nodes=...; // (WB1,WB2,OC1,OC2)
range node= 1..nodes;
 
int nodecalling [node][time]=...; // 1 if node n calls for truck t at time k
 
// VARIABLES
 
dvar boolean x [truck,node,time]; // 1 if truck t is at node n at time k, 0 otherwise
 
// OBJECTIVE FUNCTION
 
dexpr float totalproduction =
sum (t in truck, n in OC, k in time) x [t,n,k] * capacity [t];
 
maximize totalproduction;
 
//CONSTRAINTS
 
subject to {
 
forall(t in truck,n in node,k in time)
 
  ct1: x[1,1,1]==1 || x[1,2,1]==1;
 
  ct2: x[2,1,1]==1 || x[2,2,1]==1;
 
forall (k in time:(k+1) in time) ct3:(nodecalling [1,k]==1) =>
 
x [1,3,k+1] ==1 || x [2,3,k+1] ==1 || x [1,4,k+1] ==1 || x [2,4,k+1] ==1 || x [1,1,k] ==0 || x [2,1,k] ==0;
    
forall (k in time:(k+1) in time) ct4:(nodecalling [2,k]==1) =>
 
x [1,3,k+1] ==1 || x [2,3,k+1] ==1 || x [1,4,k+1] ==1 || x [2,4,k+1] ==1 || x [1,2,k] ==0 || x [2,2,k] ==0;
 
forall (k in time:(k+1) in time) ct5:(nodecalling [3,k]==1) =>
 
x [1,1,k+1] ==1 || x [2,1,k+1] ==1 || x [1,2,k+1] ==1 || x [2,2,k+1] ==1 || x [1,3,k] ==0 || x [2,3,k] ==0;
    
forall (k in time:(k+1) in time) ct6:(nodecalling [4,k]==1) =>
 
x [1,1,k+1] ==1 || x [2,1,k+1] ==1 || x [1,2,k+1] ==1 || x [2,2,k+1] ==1 || x [1,4,k] ==0 || x [2,4,k] ==0;
 
};

 

it will be much better

regards

#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: Arnautarrago

Hi Alex,

 

Thank you very much for the help.

 

By the way, I'm writing a new code and i have a problem trying to write these expressions (red colour), which aim to compare the values of two different sums:

 

Here you can see the code:

 

using CP;
 
 // NETWORK+PARAMETERS
 
float shifttime= 8;
 
int shovels=...; // set of shovels
range shovel=1..shovels;

int dumpsites=...; // set of dump sites
range dumpsite= 1..dumpsites;

float O [shovel]=...;

float W [shovel]=...;

float demand [dumpsite]=...;

float distance [dumpsite][shovel]=...;

float loadingspeed= 1.848;

float dumpingspeed= 3.078;

float capacity= 154;

// VARIABLES

dvar int+ x [dumpsite][shovel];

dvar int+ y [dumpsite][shovel];

// OBJECTIVE FUNCTION 

dexpr float totalcost= 
sum (i in dumpsite, j in shovel) (1.283 * x[i,j] * distance [i,j] + 1 * y[i,j] * distance [i,j]);

minimize totalcost;

//CONSTRAINTS

subject to {

forall (i in dumpsite) ct1: demand [i] <= sum (j in shovel) capacity * x[i,j]; // demand constraint (1)

forall (j in shovel) ct6: sum (i in dumpsite) x[i,j] * (capacity/loadingspeed) <= shifttime; // maximum number of trucks a shovel can serve in a shift (6)

forall (i in dumpsite) ct7: sum (j in shovel) x[i,j] * (capacity/dumpingspeed) <= shifttime; // maximum number of trucks a dumping site can serve in a shift (7)

forall (i in dumpsite) ct8: sum (j in shovel) x[i,j] = sum (j in shovel) y [i,j]; // number of trucks at each dumpsite remain unchanged at the end of the shift (8)

forall (j in shovel) ct9: sum (i in dumpsite) x[i,j] = sum (i in dumpsite) y [i,j]; // number of trucks at each loading area remain unchanged at the end of the shift (9)

};

 

Thank you in advance.

Regards

 

 

 

#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi

try == instead of =

regards

#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: Arnautarrago

Hi Alex,

Thank you for the help it worked perfectly! Even though, when I simulated the code, after 1h 30 min running the simulation I aborted it because I didn't find that normal. Do you know why this happens?

Once again, thank you in advance.

Regards

#DecisionOptimization
#OPLusingCPLEXOptimizer