Newsgroups: comp.constraints
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!gatech!newsxfer.itd.umich.edu!zip.eecs.umich.edu!umn.edu!news
From: "James Albert Larson" <larso171@maroon.tc.umn.edu>
Subject: Re: Optimal assignment of students to classes?
To: tmh@ictv.com
Message-ID: <90594.larso171@maroon.tc.umn.edu>
X-Minuet-Version: Minuet1.0_Beta_16
Sender: news@news.cis.umn.edu (Usenet News Administration)
Nntp-Posting-Host: dialup-4-27.gw.umn.edu
X-Popmail-Charset: English
Organization: University of Minnesota, Twin Cities
Date: Sun, 14 May 1995 04:33:14 GMT
Lines: 23

On 10 May 1995 12:48:05 -0700, 
Todd Hoff  <tmh@ictv.com> wrote:

>but for a real life
>business problem. A friend asked me how to optimally assign
>people to classes where:
>1. there are 10s of thousands of people who need training (no joke)
>2. each student has a series of classes that they must take and
>   the classes must happen in a certain order
>3. there are N geographically dispersed sites offering similar classes

... and several other conditions / objectives / constraints

This looks like a very large problem that can't be optimally solved.  
Genetic algorithms are a good way to get a very good, though usually 
suboptimal solution to a problem like this.  I know that GA has tackled 
large classroom assignment problems successfully, but I don't have much 
personal knowledge of it.  You can try the comp.ai.genetic newsgroup,
which deals with genetic algorithms.  I almost guarantee that someone that 
is expert in classroom assignment problems in GA will answer.

Jim Larson

