You are on page 1of 10

Evaluating Land Suitability using Fuzzy Logic

Method
Sri Hartati
Faculty of Mathematics and Natural Science, Gadjah Mada University
shartati@ugm.ac.id
Agus Harjoo
Faculty of Mathematics and Natural Science, Gadjah Mada University
aharjoo@ugm.ac.id
!mas Suaesih Sitanggang
Faculty of Mathematics and Natural Science, "ogor Agricultural University
imas@fmi#a.i#$.ac.id
A$stract
A system for evaluating land suita$ility using fu%%y logic method has $een
develo#ed. &he system determining a degree of delimitation, land ratings, suita$ility
classes, and their actual land suita$ility su$classes of given land characteristics. &he
main #arts of the evaluation #rocess, 'hich included in the fu%%y inference engine of
the no'ledge $ased system, are fu%%ification and defu%%ification. &hey are
discussed #artly in this #a#er to sho' ho' the system 'ors.
&he results of sho's that the system can evaluate the land suita$ility for different
ty#es of cro#s of given land characteristics or vise versa. &he evaluated rating
indicates the degree of suita$ility for cro#s to $e cultivated on a given land area 'ith
the e(isting conditions.
Keywords ) Knowledge-Based System, Land Evaluation, Fuzzy
* !ntroduction
+and evaluation is a #rocess for #redicting land suita$ility class of a given area,
and it is #otential use for agriculture, s#ecially to determine 'hich a##ro#riate cro#s can $e
cultivated in the given area, or given cro#s, 'hat ty#es of land is a##ro#riate to #lant. &his
ty#e of 'ors can $e automated using a com#uter system, a fu%%y no'ledge $ased
system, 'hich emulate the decision,maing a$ility of human agriculture e(#ert.
&here are several no'ledge,$ased system have $een develo#ed for land
suita$ility evaluation. For e(am#le, -.G!S is a system integrating e(#ert shell
system designed for mani#ulating no'ledge on land use suita$ility for agricultural
#ur#oses 'ith G!S, a commercial com#uter #acage from A/01!NF2 34ialouris et
al., *5567. !t is an rule,$ased e(#ert system used for land suita$ility evaluation and
climate condition in southern #art of Greece for five cro#s) mai%e, olive, tomato,
'heat, and gra#e. Another e(am#le is A+-S 3/ossiter 8 Armand, *5567. !t is a
com#uter #rogram for land evaluation, $oth #hysically or economically, follo'ing
the method of Food and Agriculture 2rgani%ation 3FA27. &his research focuses on
develo#ing a fu%%y no'ledge,$ased system for #hysical land suita$ility food
cro#s. &he system utili%e fu%%y logic method for evaluating land suita$ility.
9 Architecture of the no'ledge $ased system
&he structure of the no'ledge $ased system for land suita$ility evaluation
consists of t'o major #arts) the develo#ment environment and the consul tation
environment such as sho'n in Figure *.
User Interface
Recommended
Land Suitability
Class
Input Data:
Land
Characteristics
Knowledge Base:
Crop reuirements and rules for
limitation le!el determination and
land suitability classes
"u##y Inference
$ngine
Knowledge
%cuisition
C&'SUL(%(I&' $')IR&'*$'( D$)$L&+*$'( $')IR&'*$'(
User
,uman $-perts in
Land Suitability
$!aluation
Knowledge
$ngineer
&he no'ledge of this system is a$out the re:uirements for the different ty#es of cro#s
3food, horticultural, industrial or estate, and s#icy cro#s7 and it is taen from Soil and Agro
climate /esearch 0enter, "ogor. &his no'ledge has also $een collected and 'ell
documented $y researchers of ;uslit$angna 30enter of /esearch <evelo#ment for Animal
;roduction7 and ;uslit$anghort 30enter of /esearch for Horticultural <evelo#ment7
3<jaenudin et al, 9===7.
&he land suita$ility classes are evaluated using the limitation method regarding
num$er and intensity of limitations. &he evaluation is carried out using fu%%y inference
method. Under the limitation method, the land characteristics 3or :ualities7 are com#ared
'ith the re:uirements data for a s#ecific cro#. &hen, the land classes are defined according
to the num$er and the intensity of limitations. +imiting method suggests first evaluation of
climatic limitations. +and suita$ility class determination on climatic limitation is $ased on
the most severe one 3Sys et al , *55*7.
According to the FA2 3Food and Agriculture 2rgani%ation7 frame'or there are
t'o suita$ility orders) Suitable and Not Suitable and three suita$ility classes) very suitable
3S*7, moderately suitable 3S97, and marginally suitable 3S>7. &a$le * sho's the ty#es of
the limitation and their corres#onding land characteristics.
&he no'ledge re#resentation techni:ue used for this system is #roduction rule. &he
no'ledge is re#resented in the form of !F,&H-N. &here are t'o grou#s of !F,&H-N
rules, i.e. for determining the limitation level and for determining land suita$ility class.
&he #remise #art of the rule can $e a fu%%y #re#osition or non,fu%%y #re#osition 'hich is
related to land characteristic grou#, 'hile the conclusion #art is related to limitation level.
> !nference -ngine
&his com#onent is essentially the $rain of the no'ledge $ased system that #rovides
methodology for reasoning a$out information in the no'ledge $ase and on the 'oring
9
Figure *. &he structure of the no'ledge $ased system for land suita$ility evaluation
memory for formulating conclusion. &he reasoning mechanism taes #lace in the
inference engine according to then facts received from the user interface.
+and suita$ility class is determined through a t'o,#hase inference $ased on in#ut
data e(#ressed as cris# value and fu%%y set. First, inference #rocess is done to set the
limitation level, and secondly it is done to determine the suita$ility class of the land. &hese
t'o #rocesses are de#icted in Figure 9.
&a$le *. +imitations and land characteristics $eing used.
+imitation +and 0haracteristics
climate average tem#erature, rainfall, air humidity
+andsca#e and soil
&o#ogra#hy slo#e
?etness flooding , drainage
;hysical soil te(ture, coarse fragments, soil de#th
Soil fertility soil 0-0 3+and @&@7, #H, $ases saturation, 0,2rganic
Salinity and alalinity Salinity, alalinity
+and #re#aration surface roc, folded dro'n of roc
s l o p e
c o a r s e
f r a g m e n t s
s o i l d e p t h
l a n d K ( K
p ,
C . & r g a n i c
b a s e s s a t u r a t i o n
s a l i n i t y
a l / a l i n i t y
D e t e r m i n a t i o n
o f l a n d s c a p e
0 s o i l
l i m i t a t i o n
p h y s i c a l s o i l
l i m i t a t i o n
s o i l
f e r t i l i t y
l i m i t a t i o n
s a l i n i t y a n d
a l / a l i n i t y
l i m i t a t i o n
l a n d
s u i t a b i l i t y
c l a s s
D e t e r m i n a t i o n
o f w e t n e s s
l i m i t a t i o n l e ! e l
f l o o d i n g
d r a i n a g e
w e t n e s s l i m i t a t i o n
c l i m a t i c l i m i t a t i o n
a ! e r a g e
t e m p e r a t u r e
r a i n f a l l
a i r h u m i d i t y
D e t e r m i n a t i o n
o f l i m i t a t i o n
l e ! e l r e l a t e d t o
t e - t u r e
t e - t u r e
t e - t u r e l i m i t a t i o n
( o t a l L a n d
$ ! a l u a t i o n
s u r f a c e r o c /
f o l d e d d r o w n o f
r o c /
l a n d
p r e p a r a t i o n
l i m i t a t i o n
t o p o g r a p h i c
l i m i t a t i o n
i n f e r e n c e I i n f e r e n c e I I
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
f u # # i f i c a t i o n i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
3.1 Phase I Inference
!n #hase ! inference, limitation level is concluded from in#utted values of land
characteristics. For land characteristics e(#ressed as linguistic varia$les, i.e.
>
Figure 9. !nference #rocess
characteristics related to climate, to#ogra#hy, #hysical soil 3e(ce#t soil te(ture7,
soil fertility, salinity and alalinity, and land #re#aration, limitation level is
o$tained from fu%%y inference. For illustration, a #ortion of the inference #rocess
to set these limitation levels $ased on in#ut of cris# values, associated 'ith
climate, is illustrated in Figure >.
"ased on fu%%y inference, there are four #hases in limitation level setting
from cris# value in#ut, namely fu%%ification, inference, com#osition, and
defu%%ification. !nference method ado#ted in the #resent research is min, 'hile
max is used for the com#osition. 0om$ination of the t'o is no'n as max-min
inference. &his method is 'idely used in inference engine using fu%%y system due
to its easiness in com#utation 3?ang, *5567. <efu%%ification method used in the
study is center average defu%%ifier. !t is easier to im#lement com#ared to the other
t'o, center gravity and ma(imum defu%%ifier.
! e r y l o w
l o w
m o d e r a t e l y l o w
m o d e r a t e
m o d e r a t e l y h i g h
h i g h
! e r y h i g h
l o w
m o d e r a t e l y l o w
m o d e r a t e l y h i g h
h i g h
a ! e r a g e
t e m p e r a t u r
r a i n f a l l
a i r h u m i d i t y
! e r y s e ! e r e
s e ! e r e
m o d e r a t e
s l i g h t
c l i m a t i c
l i m i t a t i o n
f u # # i f i c a t i o n
i n f e r e n c e c o m p o s i t i o n d e f u # # i f i c a t i o n
R u l e 1 :
I " a ! e r a g e t e m p e r a t u r e i s c o l d
% ' D r a i n f a l l i s ! e r y l o w
% ' D a i r h u m i d i t y i s l o w
( , $ ' c l i m a t i c l i m i t a t i o n i s ! e r y s e ! e r e
R u l e 2 :
I " a ! e r a g e t e m p e r a t u r e i s s o m e w h a t c o l d
% ' D r a i n f a l l i s l o w
% ' D a i r h u m i d i t y i s m o d e r a t e l y l o w
( , $ ' c l i m a t i c l i m i t a t i o n i s s e ! e r e
R u l e 3 :
I " a ! e r a g e t e m p e r a t u r e i s m e d i u m
% ' D r a i n f a l l i s m o d e r a t e
% ' D a i r h u m i d i t y i s h i g h
( , $ ' c l i m a t i c l i m i t a t i o n i s s l i g h t
R u l e 4 :
I " a ! e r a g e t e m p e r a t u r e i s c o o l
% ' D r a i n f a l l i s m o d e r a t e l y l o w
% ' D a i r h u m i d i t y i s m o d e r a t e l y h i g h
( , $ ' c l i m a t i c l i m i t a t i o n i s m o d e r a t e
o t h e r r u l e s
c o l d
s o m e w h a t c o l d
c o o l
m e d i u m
w a r m
s o m e w h a t h o t
h o t
3.2 Fuzzification
Grou# mem$ershi# function of each land characteristics taes tra#e%ium sha#e,
determined $ased on overla##ing interval of #lant re:uirement data. &he general form of
mem$ershi# function is de#icted in Figure A, its e:uation is not sho'n here.
c1 d1
555
land characteristic 6-7
1
group 1 group 2 group 4 group i group n.1 group n
ai bi ci di an bn
2 - o!erlaping degree
2 - o!erlaping degree
2 - o!erlaping degree
2 - o!erlaping degree
&he mem$ershi# function of grou#s of average tem#erature for as#aragus is de#icted in
Fig. B, and defined com#letely CSitanggang, 9==9D. &his sho's that average tem#erature
value is 'ithin medium grou# 'ith mem$ershi# level of * if the value lies 'ithin 3*E.B,
A
Figure >. <etermination of limitation associated 'ith climate.
Figure A. General form of land characteristic mem$ershi# function.
9A.BD interval. &his interval is o$tained $y narro'ing the re:uirement interval of average
tem#erature for the medium grou#, i.e. C*E, 9BD.
a!erage
temperature 6at7
1
cold
1859
:59
somewhat
cold
cool medium warm
somewhat
hot
hot
2359 1;59
2959 1959
1359
4:59
2<59 <59
1:59
4359
4959
2 - o!erlaping degree
3:
2 - o!erlaping degree
&he mem$ershi# functions for the other land characteristics is similarly determined as it
is done for average tem#erature, and they can $e seen in CSitanggang, 9==9D. &he
mem$ershi# function for limitation level determined from limitation rating r 'ith overla#
B is sho'n in Figure F. All e:uations defining the mem$ershi# function for the land
characteristics are not sho'n here, $ut are 'ritten in CSitanggang, 9==9D.
r a t i n g 6 r 7
1
! e r y
s e ! e r e
4 9 3 9
: 5 9
= 9 9 9
s e ! e r e
8 : < :
m o d e r a t e
1 : :
s l i g h t
n o
As an illustration of fu%%ification #rocess, an e(am#le of data in#ut of land
characteristics related to the climate is given)
- Average temperature : 29.75 C
- Rainfall : 1996 mm
- Air humidity : 40
From grou# mem$ershi# function of average tem#erature, rainfall, and air humidity, the
mem$ershi# value of in#ut data in each grou# can $e com#uted. &he mem$ershi# value
of average tem#erature data in#ut is in warm and somewhat hot areas. Similarly,
mem$ershi# value for rainfall data in#ut is in the areas of medium and moderately high.
&he mem$ershi# value for air humidity data in#ut is high area. &hese mem$ershi# values
are as follo's)
Average tem#erature 3at7)
'arm
395.6B7 G =.6B,
some'hatHhot
395.6B7 G =.9B.
/ainfall 3r7)
moderate
3*55F7 G =.F,
moderatelyHhigh
3*55F7 G =.A.
Air humidity 3ah7)
moderately!high
3A=7 G *.
Using rules for setting climatic limitation level CSitanggang, 9==9D, 'e come to the
follo'ing rules)
Rule 1: !" at i# $arm A%& rf i# m'derate A%& ah i# m'derately high
()*% +limati+ limitati'n i# m'derate
Rule 2: !" at i# warm A%& rf i# m'derately high
A%& ah i# m'derately high ()*% +limati+ limitati'n i# m'derate
Rule ,: !" at i# #'me$hat h't A%& rf i# m'derate
A%& ah i# m'derately high ()*% +limati+ limitati'n i# #evere
Rule 4: !" at i# #'me$hat h't A%& rf i# m'derately high
A%& ah i# m'derately high ()*% +limati+ limitati'n i# #evere
B
Figure B. Mem$ershi# function of grou#s of average tem#erature for as#aragus.
Figure F. Mem$ershi# function for grou# of limitation level.
"y a##lying min o#erator to the a$ove rules, the truth values of the #remises are o$tained
and de#icted in Fig 6.
warm moderate
moderately
high
moderately
high
moderately
high
moderate
somewhat
hot
at > 2<5;9
:5;9
:5=
:5;9
:5=
:529
:529
:53
:53
1
1
1
1
input 1
rf > 1<<=
input 2
ah > 3:
input 4
at is warm rf is moderate ah is moderately high
rf is moderate
rf is moderately high
at is somewhat hot
%'D %'D
%'D %'D
%'D %'D
%'D %'D
+remis of
rule 1:
:5=
truth !alue of
the premise of
rule 1
:53
:529
:529
"u##ify inputs
%pply %'D operator 6*I'7
warm
at is warm ah is moderately high
moderately
high
ah is moderately high
moderately
high
ah is moderately high rf is moderately high
moderately
high somewhat
hot
at is somewhat hot
truth !alue of
the premise of
rule 2
truth !alue of
the premise of
rule 4
truth !alue of
the premise of
rule 3
+remis of
rule 2:
+remis of
rule 4:
+remis of
rule 3:
3.3 Min Inference
&he mem$ershi# function of the limitation level is derived from the generated
rules, truncated at the height corres#onding to the com#uted truth #remise value
using the res#ective rule such as sho'n Fig. E.
3. Max !o"#osition
&he climatic limitation rating cris# data.is o$tained using max com#osition
method, 'hich is com#osited from ma(imum #oints of all fu%%y sets generated
from the results of min inference such as sho'n in Fig. E. &he mem$ershi#
function #roduced from com#osition #rocess is as follo's)
F
Figure 6. Fu%%ification #rocess of cris# in#ut data.

<

<
<

<
<

=
other'ise , =
5= EA ,
*=
5=
EA F* , F . =
F* B . B6 ,
*=
BB
B . B6 B . >6 , 9B . =
B . >6 >B ,
*=
>B
7 3
2ut#ut
r
r
r
r
r
r
r
r
r
w a r m
m o d e r a t e l y
h o t
m o d e r a t e
m o d e r a t e l y
h i g h
m o d e r a t e l y
h i g h
m o d e r a t e l y
h i g h
m o d e r a t e l y
h i g h
m o d e r a t e l y
h i g h
m o d e r a t e
m o d e r a t e l y
h i g h
m o d e r a t e l y
h o t
w a r m
a t > 2 < 5 ; 9
: 5 ; 9
: 5 =
: 5 ; 9
: 5 =
: 5 2 9
: 5 2 9
: 5 3
: 5 3
1
1
1
1
i n p u t 1
r f > 1 < < =
i n p u t 2
a h > 3 :
i n p u t 4
a t i s w a r m r f i s m o d e r a t e a h i s m o d e r a t e l y h i g h
a h i s m o d e r a t e l y h i g h
a h i s m o d e r a t e l y h i g h
a h i s m o d e r a t e l y h i g h
r f i s m o d e r a t e
r f i s m o d e r a t e l y h i g h
r f i s m o d e r a t e l y h i g h
a t i s w a r m
a t i s m o d e r a t e l y h o t
a t i s m o d e r a t e l y h o t
% ' D % ' D
% ' D % ' D
% ' D % ' D
% ' D % ' D
R u l e 1 :
R u l e 2 :
R u l e 4 :
R u l e 3 :
" U ? ? I " @ I ' + U ( S % p p l y % ' D o p e r a t o r 6 * I ' 7
m o d e r a t e
m o d e r a t e
s e ! e r e
s e ! e r e
( , $ '
( , $ '
( , $ '
( , $ '
l i m i t a t i o n i s m o d e r a t e
l i m i t a t i o n i s m o d e r a t e
l i m i t a t i o n i s s e ! e r e
l i m i t a t i o n i s s e ! e r e
& u t p u t o f
t h e r u l e 1
I "
I "
I "
I "
* I ' I ' " $ R $ ' C I ' A
* % B C & * + & S I ( I & '
C $ ' ( $ R
% ) $ R % A $
D $ " U ? ? I " I $ R
r a t i n g 6 r C 7 > = 9
& u t p u t
& u t p u t o f
t h e r u l e 2
& u t p u t o f
t h e r u l e 4
& u t p u t o f
t h e r u l e 3
3.$ %efuzzification
Fu%%y sets #roduced from com#osition #rocess are converted into cris# form
through center average defu%%ifier method. 0enter of fu%%y sets of rule *, 9, >, and
rule A out#uts 3Figure E7 res#ectively are
*
r
G 69.B,
9
r
G 69.B,
>
r
G B=,
A
r
G B=,
'ith height '
*
G =.F, '
9
G =.A, '
>
G =.9B, '
A
G =.9B. 0enter average defu%%ifier
method com#uted limitation rating related to the climate 3r
I
7 as follo's)
6
Figure E. !nference #rocess for determining climatic limitation
rating cris# data.
rI G
9B . = 9B . = A . = F . =
9B . = B= 9B . = B= A . = B . 69 F . = B . 69
+ + +
+ + +
G FB.
From limitation level mem$ershi# function 'e o$tain r
I
G FB, $elonging to medium grou#
'ith mem$ershi# level
medium
3r
I
7 G *.
/ating of limitation related to to#ogra#hy, #hysical soil 3e(ce#t soil te(ture7, soil fertility,
salinity and alalinity, and land #re#aration can $e o$tained in the same 'ay as it is
sho'n for the climate
A /esults
&he fu%%y system out#uts rating and level of limitation, land suita$ility class and its
actual su$,class according to the land characteristics given $y the user. &he in#ut data can
$e e(#ressed either in cris# values or in fu%%y sets. As an e(am#le, given a set of cris#
in#ut data such as sho'n in Fig.5, the system results in the total land evaluation is
Figure 5 -(am#le of cris# in#ut data Figure *= -(am#le of out#ut
Figure ** 0ro# re:uirements Figure *9 -(am#le of fu%%y in#ut data
&a$le A. +and suita$ility class associated 'ith climate, landsca#e, soil for cris# in#ut data
0limatic
0lass
0ro#s +andsca#e
and soil 0lass
0ro#s
S* ,, S* ,,
E
S9 As#aragus, Aster, Star Fruit, Gladiolus,
0orn, 0itrus, ;eanut, Soy$ean, 0ananga,
Juinine, @lengeng Fruit, Mango, /ose,
;inea##le, "itter Melon, ;a#aya,
&u$erosa, Stra'$erry, &o$acco, S'eet
;otato
S9 ,,
S> S#inach, "ean, <urian Fruit, Mung
"ean, +ong "ean, 0innamon, /adish,
?ater Melon, 0ucum$er, "anana,
0antalou#e, -gg ;lant, &omato, 0arrot
S> "roccoli, 0orn, 0otton, 0innamon,
Soy$ean, 0a$$age, ;e##er, "netum
gnemon, ?ater Melon, 0ucum$er,
Nutmeg, "itter Melon, 0antalou#e, -gg
;lant, &omato, S'eet ;otato
N A##le, "roccoli, 0otton, 0a$$age,
;e##er, "netum gnemon, Nutmeg,
;etsai, +eaf mustard, +ettuce, &ea.
N A##le, As#aragus, Aster, S#inach, Star
Fruit, "ean, <urian Fruit, Gladiolus,
0itrus, Mung "ean, +ong "ean, ;eanut,
0ananga, Juinine, @lengeng Fruit,
/adish, Mango, /ose, ;inea##le, ;a#aya,
;etsai, "anana, +eaf mustard, &u$erosa,
+ettuce, Stra'$erry, &ea, &o$acco, 0arrot
unsuitable 3N7 for cro# of as#aragus such as de#icted in Fig.*=. !t is marginally suitable
3S>7 for corn, cinnamon, soy$ean, 'ater melon, cucum$er, $itter melon, cantalou#e, egg
#lant, tomato, and s'eet #otato such as sho'n in the &a$le A.
&a$le A indicates the land suita$ility class for each cro# related to climate as 'ell as
landsca#e and soil. &he system is also ca#a$le of sho'ing re:uirements of a given cro#.
Figure ** sho's the as#aragus cro#Ks re:uirements are given $y the system. Another
e(am#le is sho'n in Figure *9, the set of fu%%y in#ut data results in the total land
evaluation is moderately suitable 3S97 for cro# of as#aragus.
B /emars
&he develo#ed system has $een evaluated $y the e(#erts from Faculty of
Agriculture $oth Gadjah Mada University and Soil and Agro climate /esearch 0enter
"ogor, the result sho's that it is ca#a$le of determining the land suita$ility class
given the characteristics of the land, therefore, it can determine the a##ro#riate
cro#s to $e cultivated. !n addition, the system is also ca#a$le of determining the
a##ro#riate land characteristic for a given ty#es of cro#s.
Fu%%y inference method can re#resent and mani#ulate agriculture no'ledge that
is incom#lete or vague and can $e used to determine land limitation rating. &he rating
value is used to determine limitation level of the land. At the similar limitation level for
different ty#es of cro#s, the rating value is used to determine 'hat the most suita$le cro#s
to cultivate for the e(isting condition of the land. &he greater the rating value the more
suita$le cro# for land of interest.
&eferences
C*D <jaenudin, <., Mar'an, H., Su$agyo, H., Mulyani, A., 8 Suharta, N., 9===, Kriteria
Kesesuaian Lahan untu# Komoditas $ertanian, ;usat ;enelitian &anah dan
Agrolimat "adan ;enelitian dan ;engem$angan ;ertanian, "ogor, 9===
C9D Havinga H.N.L, van der Meer ;., "rou'er, L. 0ser., Fuzzy Logi%, &echnical /e#ort of
Faculty of 0ivil -ngineering and Geosciences, <elft University of &echnology,
Netherlands, *555
C>D @lir, G. L 8 4. "o., Fuzzy Set and Fuzzy Logi%& 'heory and ())li%ations. ;rentice,
Hall !nternational, !nc, Ne' Lersey, *55B
5
CAD Negoita, 0. M., Ex)ert Systems and Fuzzy Systems, &he "enjamin1cummings
;u$lishing 0om#any, !nc, 0alifornia, *5EB
CBD /ossiter G. <. 8 Armand /. M. ?., (utomated Land Evaluation System, A+-S
Mersion A.FB UserKs Manual, 0ornell University, *556
CFD Sitanggang, !.S., Sistem Berbasis $engetahuan untu# EvaluasiKesesuaian Lahan
untu# 'anaman Budidaya dengan *esin +nerensi Fuzzy, &hesis ;ascasarjana FM!;A
Universitas Gadjah Mada, 9==9
CFD Sys, 0., /anst M., 8 <e$aveye L., Land Evaluation $art ++ & *ethods in Land
Evaluation, Agriculture ;u$lications, "elgium, *55*
C6D &ur$an, -, ,e%ision Su))ort and Ex)ert Systems & *anagement Su))ort System,
;rentice,Hall !nternational, !nc., Ne' Lersey, 9===
CED ?ang, +. ( -ourse in Fuzzy Systems and -ontrol. ;rentice,Hall !nternational, !nc.,
Ne' Lersey, *556
C5D 4ialouris, 0. ;., Massilii @., Nios A. +., <ionisios @., 8 Ale(ander ". S., (n
+ntegrated Ex)ert "eogra)hi%al +normation System or Soil Suitability and Soil
Evaluation, Lournal of Geogra#hic !nformation and <ecision Analysis, Mol. *, no.9,
##.5=,*==, *556.
*=

You might also like