Re: [Help-glpk] reincarnation of tspsol

[Andrew Makhorin]

> Why would you forbid all primal heuristics, e.g. the feasibility pump, to be 
> used in conjunction with row generation?
> The result of the feasibility pump could be valid in which case the callback 
> would not add any lazy constraint.
> 

As I understand you suggest to apply primal heuristics only when all
rows have been generated, right? There is a technical point that only
rows violated within a tolerance for *current* LP relaxation can be
added, so applying a heuristic may result in a wrong solution. Another,
probably better way could be passing a solution found by a heuristic to
the callback to accept or reject it.