Tuples OPL in VRP

Question

I am trying to build a constraint of the form x[i j] +x[j i] <y[i][j] where x is a decision variable if arc ij is used.

The arcs are defined in terms of tuples. However, not all arcs are available in that [2 3] is presented while [3 2] is not.

y[i][j] is a decision variable too but it is using nodes (not tuple values) representing if the two nodes are covered by the same trip or not.

How can I write this constraint?

Note: When I run it using this constraint, it uses only same arc tuples

   forall(i,j in Setlocations :  i < j && i!=j )

   forall( g in Graph : g.o==j && g.d==i ) 

   forall(m in Graph : m.o==j && m.d==i) 
    x[g] + x[m]  <= z[i][j] ; 

Answer 1

in How to with OPL I shared a few TSP starting points.

TSP with remove circuits is an easy one.

.mod

 // Cities
int     n       = ...;
range   Cities  = 1..n;

// Edges -- sparse set
tuple       edge        {int i; int j;}
setof(edge) Edges       = {<i,j> | ordered i,j in Cities};
int         dist[Edges] = ...;

// Decision variables
dvar boolean x[Edges];

 {int} nodes={i.i | i in Edges} union {i.j | i in Edges};

range r=1..-2+ftoi(pow(2,card(nodes)));


{int} nodes2 [k in r] = {i | i in nodes: ((k div (ftoi(pow(2,(ord(nodes,i))))) mod 2) == 1)};

 

 

/*****************************************************************************
 *
 * MODEL
 *
 *****************************************************************************/

// Objective
minimize sum (<i,j> in Edges) dist[<i,j>]*x[<i,j>];
subject to {
   
   // Each city is linked with two other cities
   forall (j in Cities)
        sum (<i,j> in Edges) x[<i,j>] + sum (<j,k> in Edges) x[<j,k>] == 2;
        
   // Subtour elimination constraints.
 
   
forall(k in r)  // all subsets but empty and all
    sum(e in Edges:(e.i in nodes2[k]) && (e.j in nodes2[k])) x[e]<=card(nodes2[k])-1;  

 }  
 
 
 setof(edge) solEdges       ={e | e in Edges : x[e]==1};
 execute
 {
   writeln(solEdges);
 }