How to create an list of routes that come from a solution x[i,j]?

Question

I'm trying to create an automatic method that, through a solution x[i,j] obtained in my model, manages to generate an list[list] of all routes, which collects each solution x, in each iteration, but without duplication.

Example:

routes = []
    for i in range(N):
        for j in range(N):
                if x[i,j].x:
                    print(x[i,j])
                    routes.append([x[i,j]])

The solution obtained is the following:

<gurobi.Var x[0,1] (value 1.0)>
<gurobi.Var x[1,3] (value 1.0)>
<gurobi.Var x[3,0] (value 1.0)>

This solution is admissible, but I thought that by doing the routes.append([x[i,j]]) the array appeared in format routes=[[(0, 1), (1, 3), (3,0)]], that was the format that I would like to work, to be more simple.

Instead I get the result [[<gurobi.Var x[0,1] (value 1.0)>], [<gurobi.Var x[1,3] (value 1.0)>], [<gurobi.Var x[3,0] (value 1.0)>]], that doesn't help me.

Should I do it in another format? By the way, can you help me to create a way, for each iteration and solution, to check the routes first and avoid duplicate insertion of the same route?

Answer 1

You're looking for the indexes where x[i,j] is 1. The code should look something like this:

routes = []
    for i in range(N):
        for j in range(N):
                if x[i,j].x:
                    print(x[i,j])
                    routes.append((i,j))

Alternately, use Python list comprehension to create the list of routes:

routes = [(i,j) for i in range(N) for j in range(N) if x[i,j].X > 0.5]

And if x is a dictionary (or a Gurobi tupledict), you can simplify this to:

routes = [(i,j) for i,j in x.keys() if x[i,j].X > 0.5]