I have two IP models that should have the same optimal objective value judging from the way they are created, and in fact another MILP solver gives me an objective of -176.0 for both. Cbc (used through NEOS and also through google's or-tools' JAVA interface) gives me -175.0 and -176.0 respectively.
Is there any suggestion on what I am missing here? I attach mps and lp files for both models.
Attachment(s): https://github.com/s-c-e/cbc-trac-migration-attachments/blob/master/trac-ticket-169.zip
Hi,
I have two IP models that should have the same optimal objective value judging from the way they are created, and in fact another MILP solver gives me an objective of -176.0 for both. Cbc (used through NEOS and also through google's or-tools' JAVA interface) gives me -175.0 and -176.0 respectively.
Is there any suggestion on what I am missing here? I attach mps and lp files for both models.
Thanks, Sven