Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!news.sprintlink.net!simtel!noc.netcom.net!netcom.com!departed
From: departed@netcom.com (just passing through)
Subject: Free market planning optimization
Message-ID: <departedD8nHwq.GLL@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
Date: Tue, 16 May 1995 03:05:14 GMT
Lines: 27
Sender: departed@netcom17.netcom.com

Is there any work being done out there on economic models for decision-
making in other realms (besides the market)?

In a wargame, for example, you have a number of objectives and a limited
number of pieces to accomplish those objectives.

You could have a number of agents (one for each goal), which start out
with differing amounts of virtual cash, and these agents could trade
pieces amongst themselves.

These agents might value more highly pieces that were close to their 
goals, and be willing to 'sell' pieces that were in surplus of what they
judged they needed to accomplish their goals.

The reason I'm thinking about this is that one HUGE failure of common
strategic games (perfect general, warcraft, warlords ][) is that they
do not allocate force to an objective properly -- units trickle in one
by one towards whatever objective the computer thinks is important --
and they get crushed.  

Anyhow, you kind of get the idea -- has anyone done interesting work
on using economic models to make decisions in other theaters, like
strategic planning?

-- Richard W.
departed@netcom.com

