Originally posted by: SystemAdmin

[ivanmo said:]

Hello everybody,

I'm trying to model this constraint. This is what I did, but didn't work, it shows me this message "Operator not available for range -int" .  I appreciate any help.
I attach the file .doc to be more clearer

summation from t-1 to t+1 of x(j)(t).

forall(j in J, f in F,t in T)
ct2:
//d[f][t]==sum(t+1 in T, t-1 in T)x[j][t];
d[f][t]==sum(t in T-1..T+1)x[j][t];

Thanks

Ivan
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

[jfk said:]

hello,
I assume you are talking about OPL...
the following code works though I have no idea what are the indexes of the array x:

int F = 4;
int J = 4;
int T = 4;

dvar int d[f in 1..F, t in 1..T];
dvar int x[j in 1..J, t in 1..T]; //you write in the doc: "t for all f f is element in F" what is that??

subject to {
  forall(f in 1..F, t in 2..T-1)
      d[f,t] == sum(j in 1..J, k in t-1..t+1) x[j,k];
/*     
  forall(f in 1..F) {
      d[f,T] == ??;
      d[f,1] ==??;
  }
*/ //you have to take care of these elements separately     
}

in your shoes I would take an example and start from there...

I hope it helps

cheers
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

[ivanmo said:]

Hello

thanks for the help...
Yes I'm talking about OPL.
In the doc there's a mistake .

To be more clearer I sent the hole code, anyways I'm gonna try with the info you gave me and again THANKS A LOT

int PlanHor=...;
range T=1..PlanHor;

{string} J=...; 
{string} R=...;
{string} F=...; 
{string} JP=...;
{string} JS=...;

float p[J]=...;
float cv[J]=...;
float ch[J]=...;
float cf[J]=...;
float co[R]=...;
float D[J][T]=...;
float C[R][T]=...;
float c[R][J]=...;
float B=...;
float M=...;
float O[R][T]=...;
float u[J][JS]=...;
float Pc[J]=...;
float Rb[J][T]=...;
float penalup[J][T]=...;
float penaldwn[J][T]=...;

float Inv[J]=...;

dvar int+ d[J][T];
dvar int+ x[J][T];
dvar int+ xs[JS][T];
dvar int+ i[J][0..PlanHor];
dvar int+ o[R][T];

dvar boolean y[J][T];
dvar boolean Zup[J][T];
dvar boolean Zdwn[J][T];

maximize
sum(j in J, t in T) p[j] * d[j][t]-
sum(j in J, t in T) ch[j] * i[j][t]-
sum(j in J, t in T) cv[j] * x[j][t]-
sum(j in J, t in T) cf[j] * y[j][t]-
sum(j in J, t in T) Rb[j][t] * x[j][t]-
sum(j in J, t in T) Pc[j] * x[j][t]-
sum(r in R, t in T) co[r] * o[r][t]-
sum(j in J, t in T) penalup[j][t]*Zup[j][t]-sum(j in J, t in T)penaldwn[j][t]*Zdwn[j][t];

subject to {

forall(j in J, t in T)
ct1:
i[j][t-1]+x[j][t]==d[j][t]+sum(js in JS)u[j][js]*xs[js][t] + i[j][t];

forall(j in J, f in F,t in T)
ct2:
//d[f][t]==sum(t+1 in T, t-1 in T)x[j][t];
d[f][t]==sum(t in T-1..T+1)x[j][t];

forall(j in J, t in T)
ct3:
d[j][t]-D[j][t]<= Zup[j][t]*M;<br />
forall(j in J, t in T)
ct4:
D[j][t]-d[j][t]<= Zdwn[j][t]*M;<br />
forall(j in J, t in T, r in R)
ct5: sum(j in J)c[r][j]*x[j][t]<=C[r][t]+o[r][t];<br />
forall(r in R, t in T)
o[r][t]<=O[r][t];<br />
forall(j in J, t in T)
ct6: x[j][t]<=B*y[j][t];<br />
forall(j in J, t in T)
i[j][1]==Inv[j];

forall(j in J, t in T)
i[j][0]==Inv[j];
}
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

[ivanmo said:]

Hello, can any one tell me where can I find examples, close to this problem, I've tried to find something similar and I haven't found nothing like this.

Thanks
#DecisionOptimization
#MathematicalProgramming-General