Originally posted by: NSopt

Hello.
I'm working on branching study by using C callable API.
I saw below answer in previous discussed thread.

" Not necessarily fatal if you're willing to do a bit of extra work. Suppose you want to split node A into three children (B, C and D). At node A you call the branch callback twice. The first time you add a combination of constraints and bounds that defines B. The second time you add some single constraint or bound, however vacuous, that applies to both C and D, creating child A'. Then at A' you split into C and D. It requires a bit of extra bookkeeping to keep track of which nodes need which new pieces, but it's not horrible. CPLEX allows you attach a user data object to each node, so you can use a custom object to tag each node with information about whether (and how) it needs further splitting." by Paul Rubin

Could Paul or anyone please explain more about this? (i.e., branch on more than two children)
Especially, I'd like to know how we can attach a user data object to each node and how we can use attached information for this implementation.
Thank you.

Best,
NS
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

I wrote a blog post about it, including sample Java code. The C API is mechanically different, but you should be able to make the transition.

Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: NSopt

Thank you very much for your really good answer!
It was a great help for me.
Sincerely
NS
#CPLEXOptimizers
#DecisionOptimization