From: "Clay Olbon II" <Clay.Olbon@dynetics.com>
Date: Sat, 27 Jul 1996 00:55:19 +0800
To: "jim bell" <cypherpunks@toad.com>
Subject: Re: Twenty Bank Robbers -- Game theory:)
Message-ID: <AE1E48FA-4FC47D@193.239.225.200>
MIME-Version: 1.0
Content-Type: text/plain


jim bell <jimbell@pacifier.com> wrote:
>My previous answer was incomplete, of course.  I continue to believe that 
>the problem is unsolveable as stated, if for no other reason than the 
>"weight" of the negative represented by dying is not stated.  It's a VERY 
>complex problem, unless there's some trick I'm not seeing.

Jim is right.  The problem with any optimization problem is when
unquantifiable negatives are included.  The "classic" example of this is an
inventory problem.  The optimal solution is minimal (or no) inventory,
however there are unquantifiable negatives that arise when a customer
cannot get his product when he wants it.  I don't think the problem is that
complex if the negative is that you get no money rather than "you die".  As
an aside, many game theory problems (possible including the simplified
version of this one) are solvable using linear programming (and no, that is
not writing C one line at a time ;-).  It has been far too long since my
last game theory course to consider trying to set this problem up however
(I've found that I don't use either game theory or LP a whole lot in the
"real world" of engineering).  

	Clay


***************************************************************************
Clay Olbon II       *      Clay.Olbon@dynetics.com
Systems Engineer    *    PGP262 public key on web page
Dynetics, Inc.      * http://www.msen.com/~olbon/olbon.html
***************************************************************** TANSTAAFL