W dniu 2010-04-13 13:47, Philipp Marcus pisze: > >> Ok. So what is your exact task to do: >> 1) to prove NP-completeness of the problem >> > That is more a "bonus" task. Do you have any ideas how to prove it? I > think the best way is to map it to a problem that is known to be > NP-complete, but I don't know which one is suited best for that.. > Your problem is a variation of k-median problem that is NP-complete. Maybe you should check this path... >> 2) write a solver for a problem using ECLiPSe ? >> > Thats what I have to do. Currently I am working on a version that uses > ic_symbolic and includes all your hints. I could post it to the list > when I'm done. > You do not need to use ic_symbolic. You could have a mapping: 1-A, 2-B, 3-C and use ic or fd library. Best regards -- [ Wit Jakuczun http://pl.linkedin.com/in/jakuczunwit ]Received on Tue Apr 13 2010 - 14:47:17 CEST
This archive was generated by hypermail 2.2.0 : Thu Feb 02 2012 - 02:31:58 CET