Stefano,
If you assign a feasible solution to the variables before invoking CPLEX,
then CPLEX will use the objective value corresponding to that solution to
compute its initial upper bound (when minimizing), and if it is a good bound
then the search time may be reduced. In addition CPLEX has a heuristic that
can take an initial solution (feasible or infeasible, I think, though I'm
not entirely sure) as a start in making a quick search for a better
solution, which would provide an even better bound.
In general warm starting does not work all that well for mixed-integer
programming, though. Even if you send CPLEX a solution that is optimal, in
fact, it may need a long time to establish that there really are no better
solutions.
Bob Fourer
4er@...
> -----Original Message-----
> From:
ampl@... [mailto:
ampl@...]
> On Behalf Of Stefano Coniglio [
stefano.coniglio@...]
> Sent: Wednesday, June 17, 2009 6:49 AM
> To: AMPL Modeling Language
> Subject: [AMPL 2601] Warm start with mixed-integer problems
>
>
> Greetings,
>
> to what level can a warm start be done via AMPL+CPLEX for mixed-
> integer linear problems?
>
> It looks like the satuses sent via
> option send_statuses 1
> are devised for linear problems only (i.e. they only describe the
> basic status of the variables in the solution).
>
> Can I obtain some warm start by setting the initial solution to some
> value?
>
> Thanks a lot,
> -Stefano
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "AMPL Modeling Language" group.
To post to this group, send email to
ampl@...
To unsubscribe from this group, send email to
ampl+unsubscribe@...
For more options, visit this group at
http://groups.google.com/group/ampl?hl=en-~----------~----~----~----~------~----~------~--~---
