CODEWITHSUNDEEP
gams-math
user9990604
user9990604

Reputation:

How can I make this set?

let s be a index of scenario and j index of Network Nodes.(s={1*200} , j={1*100}) d(s,j)= total impacts of contamination scenario s, if the contaminant is first detected by scenario at j. W(s)=the subset of Network Nodes which d(s,j)<>0. W'(s)= the subset of W(s) such that d(s,j)<>d(s,j') , for all j,j' belong to W'(s).

How can I make the subset W'(s)?

I got error with below code! My try:

Set
     i /1*100/
     s/1*200/

;

Alias (i,j);
set
     Node(j)
     Wp(Node)
     same(Node)
     U(Node)
     ;

Parameter d(s,j);
scalar t,posMin;


  $call GDXXRW.exe data.xlsx  par=d rng=sheet2!A1:X200
  $GDXIN data.gdx
  $LOAD d
  $GDXIN


Loop(s,
     Node(j)=(j.val);
     while(Card(Node)<>0,
             t=smin(Node$(d(s,Node)),d(s,Node));
             same(Node)$(d(s,Node)=t)=yes;
             posMin=smin(Node,same(Node));

             display t,same, posMin;
             Wp(Node)=Wp(Node)+( Node$(Node.val=posMin));
             display Wp;
             same(Node)$(Node.val=posMin)=no;
             u(Node)=same(Node);
             Node=Node-U;

     );


 );

Upvotes: 0

Views: 75

Answers (1)

Berenger
Berenger

Reputation: 376

If I've understood your problem and type of data correctly, would something like this work?:

Given some data:

set s / 1*5 /,
    j / 1*3 /;

alias(j,i);

table d(s,j)

    1   2   3
1   0   20  20
2   0   0   20
3   10  20  30
4   0   0   0
5   0   20  30

;

Set W is a subset of s containing only non-zero rows:

set W(s);
W(s) = yes$sum(j, d(s,j));    
display W;

Set Wp is a subset of W(s) containing the elements of W minus the non-zero repeated elements across j:

set Wp(s);
Wp(s) = yes$W(s);

loop((W(s),i,j),
    if(((d(s,i) eq d(s,j)) and (d(s,i) ne 0) and (ord(i) ne ord(j))),
    Wp(W) = no;
        ));

display Wp;

For me, this makes: s={1,2,3,4,5} W={1,2,3,5} Wp={2,3,5}

** EDIT (from comments) **

Create a set Q. If your data is over two dimensions and you want your set to be defined over two dimensions, this will suffice:

set Q(s,j);
Q(s,j) = yes$d(s,j);
display Q;

Create parameter storing min values across j for each s (in Q):

parameter xp(s);
xp(s) = smin(j$Q(s,j), d(s,j));
display xp;

Create a copy of original data that only retains the min values across j:

parameter d_copy(s,j);
d_copy(s,j) = d(s,j)$(d(s,j) eq xp(s));

Remove any duplicates as j increases:

loop((s,i,j),
    if((d_copy(s,j) eq d(s,i))$(ord(j) gt ord(i)),
        d_copy(s,j) = 0;
    );
);

display d_copy;

Create a subset of Q that corresponds to the min values:

set Qp(s,j);
Qp(Q) = yes$d_copy(Q);
display Qp;

Upvotes: 2

Related Questions