Hi everybody,
I would like to ask a ruther simple question, at least for the Gurus of GLPK.
I have a very large and difficult optimization problem to solve. GLPK works for days and still it does not find any solution. On the other hand, I have manage to get a good solution (which I do not know if it is optimal or just near optimal, instead) using Computational Inteligence heuristics. The question is : is it possible to "feed" GLPK with this solution and force GLPK to use this solution as a starting point for further searching process? If I can do so, I hope that GLPK would be helped to find at least a better solution compared with that found using heuristics, even if it is not optimal.
Please, note that I use Mathprog language and the terminal to run GLPK (command prompt). Nevertheless, being beginner with the capabilities of GLPK, I would appriciate a detailed answer with analytical description of steps I have to follow in order to achieve the above mentioned goal.
Thanks for your time and for any possible answer.