Originally posted by: EdKlotz
By "solve the resulting tree" I presume you mean you take the branches (i.e. bound changes) that led to the single leaf that prompted you to abort the run, modify either your original model with those bound changes, and solve. If so, any bound changes will be viewed as invalidating the tree, so a subsequent optimization will generate cuts again and start at the root node.
If this is not what you are doing, then explain exactly how you take information from the leaf node of interest and create the "resulting tree".
The C API has a CPXgetcallbacknodelp function that would allow you to get the node LP and, with a bit of work, identify the cuts CPLEX has added. So you could use the C API to provide cuts from the solve that gave you the leaf node to your "resulting tree" problem. But it will still start at the root node.
#CPLEXOptimizers#DecisionOptimization