Re: Fuzzy Arithmetic Request

WSiler (wsiler@aol.com)
Wed, 22 Jul 1998 20:16:05 +0200 (MET DST)

>Through Internet we bought a CD from Academic Press, which contained 4 Books
>more some softwares, but none of those talk about the basic Fuzzy arithmetic
>intervals (+,-,/,*,ln,etc.) I Beg you help us with the theory involved in this
operations

The basic theory behind fuzzy arithmetic is Zadeh's extension principle. That
is, to evaluate C = A o B, where A, B and C are fuzzy numbers and o is some
operator, we go about it this way. Let x be a member of A, and y be a member of
B. We want to calculate the membership of z = x o y in C = A + B. If we do
this for all possible z values, we will have the membership function for C.

To do this, we select a value for z. Now for all x and y with x o y = z, we
calculate the MAXIMUM of all the MINIMUMs of the memberships of x in A and y in
B. This "max of all the mins" is Zadeh's famous extension principle.

For example, we will take a fuzzy 2 and a fuzzy 5, and add them.

1.0 - a b
.8 - a a b b
.6 - a a b b
.4 - a a b b
.2 - a a b b
0.0 ---------a--------------a---------------------b------------------------
0 1 2 3 4 5 6 7
8 9

Now we take every pair of numbers x and y which add to some value z = x + y. We
calculate the membership of x in A, a(x); of y in B, b(y); and take the MINIMUM
of these. This is a "trial" value for the membership of z = x + y in C . We do
this for every x-y pair for which x + y = z, and take the MAXIMUM of all these
trial values as the membership of x + y in C. = Say (to start) z = 7. For a sum
of 7, say x is 3 and y is 4. The membership of x in A is zero, and the
membership of y in B is .5. We take the minimum of these, which is zero, and
note it. Now we try x = 2..5 and y = 4.5.The membership of x in A is 1, and the
membership of y in B is also 1. The minimum of these is 1. This is bigger than
our previous value of .5, so we discard the .5 and keep it. If we keep on going
we will get values which are smaller than 1, so the membership of 7 in C is 1.

Of course, we will do this analytically if we can. For triangular membership
functions this is easy to do and we will end up with this for C = A + B;

1.0 - a b c
.8 - a a b b c c
.6 - a a b b c
c
.4 - a a b c b
c
.2 - a a b c b
c
0.0 ---------a--------------a-------c-------------b-----------------------c-
0 1 2 3 4 5 6 7
8 9

Hope this helps some - William Siler

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html