[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-glpk] [Fwd: A query and a request for help
|
From: |
Michael Hennebry |
|
Subject: |
Re: [Help-glpk] [Fwd: A query and a request for help |
|
Date: |
Thu, 9 Apr 2015 19:54:05 -0500 (CDT) |
|
User-agent: |
Alpine 1.00 (DEB 882 2007-12-20) |
From: sanaullah swati <address@hidden>
sir i have this objective function for creating districts by clustering
of census data (gis data)
Inline image 1
Not seeing it.
The above objective function of the optimal redistricting model is to
minimize the totalpopulation weighted distance to each district center
summed across all the districts.
What you seem to want would be difficult to impossible with a finite MILP.
Replacing distance with L1 distance or some other suitable
approximation to distance would render it just diffucult.
What you seem to be trying to do is called clustering.
http://en.wikipedia.org/wiki/Cluster_analysis#Centroid-based_clustering
One large MILP is not usually how they are solved.
--
Michael address@hidden
"SCSI is NOT magic. There are *fundamental technical
reasons* why it is necessary to sacrifice a young
goat to your SCSI chain now and then." -- John Woods