Introduction to Grey System Theory

Deng Julong
Deynrt~iientof A~rton~crtion,H~mzhongU ~ ~ i ~ ~ eofr 'Science
s i t ~ ~ rrncl Technologj~,W~rl7cr11,
People's Rey~rblicof Cl7incr

Received October 1988

Outline

Grey System theory was initiated in 1982 [7]. As far as information is con-
cerned, the systems which lack information, such as structure message, opera-
tion mechanism and behaviour document, are referred to as Grey Systems.
For example, the human body, agriculture, economy, etc., are Grey Systems.
Usually, on the grounds of existing grey relations, grey elements, grey
numbers (denoted by 8 ) one can identify which Grey System is, where
"grey" means poor, incomplete, uncertain, etc. The goal of Grey Systeni and
its applications is to bridge the gap existing between social science and natu-
ral science. Thus, one can say that the Grey System theory is inter-
disciplinary, cutting across a variety of specialized fields, and it is evident that
Grey System theory stands the test of time since 1982.
As the case stands, the developn~entof the Grey System-as well as theo-
retical topic-is coupled with clear applications of the theory in assorted
fields.
The conccept of the Grey System, in its theory and successful application,
is now well known in China. The application fields of the Grey System in-
volve agriculture [23, 77-81, 911, ecology [59], economy [61, 102, 103, 1041,
meteorology [58, 74,911, medicine [55, 891, history [63, 641, geography [I], in-
dustry [61, earthquake [73, 87, 881, geology [76, 1191, hydrology [98, 1121, irri-
gation strategy [261, military affairs, sports [116], traffic [67], management [30,
97, 1051, material science [82, 831, environment [ 1081, biological protection
[69,70], judicial system [loo], etc.
Projects which have been successfully completed with the Grey System
theory and its applications are as follows:
1. Regional econonlic planning for several provinces in China;
2. To forecast yields of grain for some provinces in China:
The Jo~rr~lal
of Grey Sj~sretii0957-3720 / 89/ S 03.50 @ Sci-Tech Information
. Services. England and
China Petroleum Industry Press, P.R. China. Printed in P.R.China.
 3. To analyse agricultural economy in China;
4. To make satisfactory planning of irrigation (watershed of the People's
Victory Channel in Henan Province);
5. To build models available for biological protection;
6. To control the water level for boilors by grey prediction control;
7. To estimate the economic effect;
8. To build a diagnosis model available for medicine;
9. To forecast weather.
As far as properties of implement fields are concerned, it can be seen that
the greatest part of them can be referred to as abstract systems, which are
short of physical prototypes, according to the terminology used throughout
the Grey System theory. They can be called latent Grey Systems.
The aims of Grey System theory are to provide theory, techniques, notions
and ideas for resolving (analysing) latent and intricate systems, for example:
1. To establish a non-function model instead of regressive analysing;
2. To define and constitute Grey Process replacing the stochatic process
and to find the real time techniques instead of statistical model to deal
with the grey process, in order to obtain an approach to modelling with
few data, avoiding searching for data in quantities;
3. To turn the disorderly raw data into a more regular series by grey gen-
erating techniques for the benefit of modelling instead of modelling with
original data;
4. To build a differential model-so called grey model (GM)--by using the
least 4 data to replace difference modelling in vast quantities of data;
5. To develop a novel family of grey forecasting methods instead of time
series and regressive methods-all of these may be referred to as an ap-
proach to deal with grey processes;
6. To develop innovational techniques and concepts for decision making-
so called grey decision making;
7. To develop novel control techniques. e. g., the grey forecasting control
replacing classical control which is referred to as afterward control, also
relational control, generating control and programming control;
8. To study the mechanism theory, including grey sequence theory and
grey structure theory;
9. To study feeling and emotion functions and fields with whitening func-
tions;
10. Based on grey relations, grey elements and grey numbers to be used to
study grey mathematics instead of classical mathematics study.
The essential contents and topics of Grey System theory encompass the fol-
lowing areas:

grey relational space

grey generating space
grey forecasting
 Introduction to Grey System Tlteory

grey decision making

grey control
grey mathematics
grey theory

Why have so many researchers in the fields of software science utilized

Grey System theory, and why has there been dialogue between them and the
Grey System theory? At least two reasons immediately spring to mind. First,
the software researchers tend to be methodologically accepting, when they
are grey mathematically efficient and sure of being grey methodologically
innovative; second, some of the element techniques and concepts of Grey
System theory have been implemented and investigated by a variety of areas
and even incorporated in some theoretical schema, such as grey market and
economic programming.

Grey Relational Space

The grey relational space (GRS) [I 1, 181 is one that describes the posture rela-
tionships between one main factor and all the other factors in a given system.
A GRS is a binary set denoted by (X, T)where X is a collection composed of
sequences x i to be compared and reference sequence x,, r is a map set cal-
led grey relational map set, YETis an appointed relational map in GRS. As-
sume that

is an image at point k from the series to real number with map y and
y(xo, x i ) is an image at all points with k = 1, 2, 3, n , where
a * . ,

Let y(xo, x i ) satisfy that

then y(xo(k), xi(k)) is said to be a grey relational coefficient at the point k

and y(xo, x i ) to be a grey relational grade, iff r satisfies the following axi-
oms [18]:
 Deng Jirlong

(Al) (Norm Interval)

y(x,(lc), si(k))€(O, 11, V k

y(x, ( k ) , x i (1~))= 1, iff s o( k )= x i (I<), VIZ
y(s, ( k ) , x i (12)) = 0, iif x OE(P, X , E (P

where cp is an empty set.

(A2) (Duality Symmetric)

~ ( x(k),
, x i (k)) = y(xi (Ir), x o(k)), iff
X = {x,, x i )

(A3) (Wholeness)

?(so( k ) , x i (I<))# y(x, (I<), x,, ( k ) ) ,

almost always, iff

x = {xj[i= 0, 1, ..., n , I? > 2)

(A4) (Aproachability)

y(x, ( k ) , x i ( k ) ) decreases along with A(k) increasing, where

Now, we have a following expression for y(x0(k), x i ( k ) ) which satisfies all

of the mentioned axioms, that which

where [E(O, 1) is of the distinguished coefficient [53].

Remark. The norm interval (0,1] which means 'dy(x,(k), x,(k))~(O,l]'dk

was first developed by Deng (1985)[ 181 , while the norin interval [ ( I , 01,
y(x, (I<), s,( k ) ) ~ [ c ib], where a, b are arbitrary real numbers was developed
by Guo (1985)[531. The concept of grey relational grade of interval was pro-
posed by Wang (1985)[941.
 Introduction to Grey System Tlzeoty

A variety of successful applications for GRS have been exploited by many

scientists in different areas, such as material science [83], biological analysis
[841, seismological reseach 11871, judical system [loo], economic analysis [117],
and geology [76].

Grey Generating Space

In the viewpoint of Grey System theory, the concept and generating tech-
niques are important ideas by which the disorderly raw data can be turned to
a regular series for the benefit of grey modelling, transferred to a
dimensionless series in order to obtain an appropriate fundamental for grey
analysing and changed into a unidirectional series for decision making.
Let x, y, $ be the series and x ~ X , yEY, $E@, where X, Y, @ are collections
of dimension n. If

then @ is a basis space, $, a basis of @, x an image of J) and X an image

space spanned by the basis $, and the coordinates Y.
Hence

is a generating space [40].

Suppose that ,

and that

x (0) EY, x (1) EX, $:.')E@

column
 Deng Julong

Then

By AGO the above generating operation is denoted, i.e.

thus AGO x"' is said to be an Accumulated Generating Operation which is

vital to grey modelling. Similarly, the jth order AGO is as follows:

and that

Moreover, some theorems concerning the jth order AGO of x"' have been
established 1401.
Assuming that

(X(') (k))= x (k)

a ( I ) (x(') (k))= a 'O) (x (k))- a (O)(x(') (k - 1))

then we have
 Introduction to Grey System Tl~eoi-y

By IAGO the above-mentioned generating operation is denoted, such that

Thus IAGO x'l) is said to be an Inverse Accumulated Generating Opera-

tion.
Summing up the results of grey generating operations, we have

1. (AGO):

2. (LAGO): a(i)(xG)(k))= O1(i-l)(s(j)(k))- a(i-l)(sG)(k - 1))

3. (MEAN): z '" (k) + 0 . 5 (~I ) (k - 1)

(k) = 0 . 5 'I'~-
4. (PI: Y = (Y(l), ..-, x(n)), s = (x(l),...,x(n)),
1 "
x(k) = Y(k) / x(p), x(p) = ; Y(m)
,,,= I
5. (INIT): ,F= (:(I), --.,Y(iz)), x = ( ~ ( 1 )---, ~ x(n)),
x(k) = T(k) / Y(1)
By AGO, the following important results can be obtained: the non-nega-
tive, smooth, discrete function can be transferred into a series, extended ac-
cording to an approximate exponential law [36,41] which is called the grey
exponential law and by which a reform to establish a suitable foundation in
building a differential model is said to be completed.

Grey Modeling

On the subject of differential modeling, T.C. Hxia follows that, for biological
phenomena, economy and biological medicine, one had better build a
differential model, but this has been impossible in the past. In Grey System
theory a dynamic model with a group of differential equations is developed
[15,20,42,44], which is called grey differential model (GM). To do this Deng
(1985)[20]inferred:
1. A stochastic process whose amplitudes vary with time is referred to as a
grey process;
2. The grey modelling is based on the generating series rather than on the
raw one;
3. The grey derivative, parallel shooting and grey differential equation are
 defined and proposed in order to build a GM;
4. T o build a G M model, only a few data (as few as 4) are needed to distin-
guish it.
(0) (1) (0)
Let .x be a raw series, x = AGOx , then the following equation

is a grey differential model, called GM(1,l) as it includes only one variable

(0)
x , where

b, are the coefficients; in Grey Systenl theory terms, cr is said to be a devel-

ci,

oping coefficient and b the grey input, x'O)(k) is a grey derivative which
maximizes the information density for a given series to be modelled.
According to the least square method, we have

there

Here B is called a data matrix.

By regarding the following equation

(0)
as a shadow for s (12) + c ~ z (12) = b
(1)

We have

where C3 (x'l') is said to be a background grey number for

is the whitening value of grey number @ (x")) and

6(x"') 4 +
x '"(t At)/At
x'"(t - At) / At

means the parallel morphism or shooting from 6(x"') to the components of

dx'" / dt [20, 351, then the response equation for GM(1,l) are as follows

'1) (0)
where ~ ' " ( k ) , P'O'(1c) means calculating values of x and x at point k.
A grey differential equation having N variables is called GM(l,N), whose
expression can be written as follows:

where b is said to be an ith influence coefficient, which means that x i exer-

,
cises influence on x (the behaviour variable). Then we have

T
ii = [CL, b,, bNl

By means of least square method, we have

Model GM(1,I) plays an important role in grey forecasting, grey Pro-

gramming and grey control. Model GM(1,N) has laid an important founda-
tion for regional economic programming and grey multivariable control [42,
1021.
Grey Forecasting

The subjects of grey forecasting 121,431include:

series forecasting
calamities forecasting
season calamities forecasting
topological forecasting
systematic forecasting

As all of these are based on GM(l,l), we call'them grey forecasting, whose

objective is to unify the field and to bridge the gap between grey process theo-
ry and practice. Our intention is to make forecasting useful for decision and
policy makers who need future predictions.

Grey Series Forecasting

Grey series prediction is a comlnon technique available for forecasting-

which is referred to essential prediction-and direct use of GM(1,l) will ena-
ble us to know where and how tke events to be forecasted would appear.
(01 (0)- (1) -(O)
Let x be a raw series, AGO x - x , x (t) the prediction value at
point t , then the series prediction process can be written as

(0)
IAGO . GM . AGO: x +.v^'O'([)
(0) - (0) (0)
X -( ( 1 ) , 'A ( I ) ) [ > 17,

(01 (0) (1)

where IAGO and AGO are maps which transfer ,?'I' to x and x
A

to x
respectively, and G M implies the grey modelling for GM(1,l) by x ( l ) . For a
series prediction, a variety of successful instances have been completed in dif-
ferent areas, such as grain yields prediction [62, 79, 1141, economic prediction
[ 108, 92, 1021, material science [82], epide~niology[891, electrical power [75],
traffic [671, technological progress [851, seismology [881,judicial systems [loo],
engineering [I201 and agriculture [I 17, 1181.

Calamities Prediction

While the amplitutes of events to be forecasted refer to an appointed interval

called "grey number of pests", then the predictions to know when the event
will occur belong to calamities prediction [21,43].
(0)
Let s be a raw series
 Introduction to Grey Sjwtenz Tl~eory

C3 ,, 63 represent grey numbers of pests such that

where is said to be a boundary value of pests and 8 , is a grey number of

low pests which means that the pest value is less than C; 63 If is a grey number
of upper pests in which the pest value is greater than 5. If we have a subseries
(0) (0)
'

x , ofx such that:

-
(0)
x r - (x (I,),
(0)
..-, x (0) (12,))
x'0'(m,)~C3,nx(0), .1= 1, 2, n

then

,
Suppose that p , p , and p, are maps such that

(0) (0)
PC:x -+x,
pll: x'O)(k)+k
(0)
p,: x (ni,)+nzr
p,: k , + k

then the calamity prediction can be written as follows

p, . p,, ' pr: " (0)

+ ( p , ( l , ) , p'A (2,),
- ..., ~ ~ ( 1 2 ~ ) )
IAGO GM - AGO: ( p , ( l , > , ..., pg(?zr))+(te)

where t S is a time to be forecasted when pests appear and ( t S ) is a series of

1:.
A variety of instances of pest predictions become effective in different
areas, such as meteorology [I 15, 581 and agriculture [78].
Seasonal Calamity Prediction

The pests to be forecasted which happen in a certain span of time of a year

are referred to as a seasonal calamity. It is obvious since an assignment to
forecast the seasonal pests is similar to the above-stated prediction-namely,
the calamity prediction-but their procedures are different [21,43].
Let h(k) be a pest time (date) located in a special span of a year; its scalar
quantity refers to a relative value regarding the minirnuln value among the
samples of pest date, and h is the series of h(1)

h , is the straight line given by

where A, is a straight line to link h(1z-1) with Iz(lr), then the collection

is a line graph called the seasonal pest graph.

Suppose d , =const which means a fixed date of pest and assume that
di cuts across Y at points 1, 2, ... 17 and it is said to be a contour line in the
pest graph. Let y i(lc) be the kth span from point 1 to I<,then

is a series for modelling, iff p i is non-empty, the potency of p , is inore than

4 and

The process of seasonal pest prediction thus can be written as

 IAGO GM AGO *fiy(t)
5 > 12;

where p^y([) is a year to be forecasted in which the seasonal pest would oc-
cur at date d i
Instances would be to forecast the date that the first frost would occur in
Shanxi Province [91] and the date that the seasonal wind would appear in
Hunan and Guangxi Provinces.

Topological Prediction

According to the above-mentioned procedures and techniques for seasonal

pest prediction, forecasting the future form of a given curve is called
topological prediction [21,43].
Let us suppose that Y is a graph concerning a given curve such that

{ d , } represents a subset of points belonging to a set of points which may be

located at a straight line to the given or another curve. Thus:

IAGO . GM AGO (P,}-(P^~"'(S),{di}*bi}

is a topological prediction model referred to { d , } and @ y ( t ) ) is a future

cuvre to be forecasted. In order to analyse and tp deal with a grey process
whose curves are not smooth, the topological prediction techniques may be
useful.

Systematic Prediction

To forecast a variety of different variables as a whole, referred to a given sys-

tem, it is said to be a systematic prediction. If the assumption is made that the
variables to be forecasted together are subordinated to, and modelled with,
GM(1,l) and the results for every variable can be obtained recursively one by
one, then such a forecasting is said to be a grey systematic prediction.
Let x i (i= 1, 2, n) be the behaviour series referred to a given system,
then the whole x i can be forecasted by means of a grey systematic
prediction, if the following relationships between variables x i , i= 1, 2, ... n
hold,
 (1)
where 2 I (i= 1, 2, ... 11) are the background grey numbers concerned with
derivatives

Let us denote the procedure to model GM(1,l)

by GM,, and by g177,, the result or the model of it. As a function of gn?,,, con-
sider the n -lth grey differential equation, as follows:
(1)
ds,, - 1 (1) (1)

dt + ' 1 ~ ~ - ~ . l l - ~ 2 , , - ~ - ( l ~ l - ~ . , ~ ~ Y , l

(1)
-.
*-d s I l _ , (I)
dt +",,-l.rr-~ Xrr-I -NII-~.lpf (g'",,)

Thus we can briefly denote it with

where G M , I - l implies the precedure to model 17 -1th relationship and

( g177 , , g177 ,,) is the result from GM - , which involves models g117 and
g n ~ ~ ,. _Hence,
, the grey systelnatic prediction call be presented briefly with
the following representations:
 GM, = ( g r ? ~ ,g1712,
, .-. g11711)

It is evident that to model a variety of variables together by the grey sys-

tematic prediction which is obtained directly from a variety of models,
GM(l,l), the complex procedures which are necessary to solve differential
equation set by common means, are to be avoided.
The effect and convenience of the grey systematic prediction has been illus-
trated by many instances; for example, to forecast grain yield, applying ferti-
lizers, yield per mu and irrigated area together in Shanxi Province [79].

Grey Decision-Making

Grey decision-making is primarily concerned with the grey strategy of situa-

tion, grey group decision making and grey programming.

Grey Strategy of Situation

Grey strategy of situation deals with the strategy-making based 011 multi-ob-
jects which are contradictory in the ordinary way. The situation named by
Grey System theory implies that the events to be attended to are coped over
by games; then we have

s = (ci, b ) = (event, game)

where s is of an appointed situation, cr the event, b the game to be used to

cope over the event.
The following are the essentials to make a grey strategy of the situation [22,
461:
1. Exploiting point- set topology: the construction of situation set can be
determined when an event is uncertain and games are unique. Let A, B
be collections for events and games respectively, then we have

40 = {sls = x Y i when i f j , then Y i= A , Y i = L ~ , E O ~ ,

i= l

where 0 is a system of open sets, q is the sub- basis of .v, x is the

topology of situation.
2. To discover the effects of a given game dealing with a raised event, let us
consider the map o.
 where a is an effect map from situation s, to its sample u:' describing
the effect value concerning the pth object. By arranging the effect sam-
ples of all the objects, p = 1,2, .-.nz, the following effect series is obtained
(I) (2) 011)
u.. = {21C , 21.. , a-., 21C }
r/ 0

3. Consider a grey relational space (X,I?) where X is a collection of u U, r is

a map set in relational space.
Assuming the grey target is

where zl, is a reference sample series, which implies the kenel of target,
then s ,i. is a satisfactory situation and b,. the satisfctory game deal-
ing with event a i ,iff

It is also important to make a satisfactory strategy by means of effect

measure maps, which transfer the disconrormity samples resulting from
different objects into identical scala~s.

A grey variety of problems have been solved using the grey strategy of situ-
ation, such as the irrigation strategy in Henan Province [26], a forestation
strategy in Jinagsu Province and the agricultural economic division in Shanxi
Province [93, 1181.

Decision-Makingof Group

A comprehensive treatment of the strategy having three levels, by means of

grey relational space, grey statistics, grey clustering and grey prediction, is re-
ferred to as a decision-making of group [22,28,46].
Citing a case as a group strategy, let us consider an investment strategy
whose items to be decided may be an exploited products-a category of
investment, namely, the invested number and range which are to be marked
by "much", "middle" and "less". The decision levels, namely, the collections
of the decision and policy makers, are mass groups whose strategy is usually
determined solely by grey statistics; the expert group whose original strategy
is obtained by grey prediction as usual and whose attention is paid only to the
exploited products; the final group of the three levels, namely, the managers
group, whose master strategy comes from official documentation and the
quantitative strategy concerned is obtained by grey clustering in an ordinary
way, taking only the category of investment into consideration.
A comprehensive strategy which gives consideration to the three levels can
be obtained by unifying the decision series referred to the three groups,
respectively, based on grcy relational space.
A case in the above-mentioned point is one to harness a chaanel in Henan
Province [26].

Grey Programming

To confirin traditional progralnming as a foundation, to incorporate grey

prediction model in order to make a dynamic programming and to regard the
coefficients as grey numbers with provisioll for adapting to the environment,
are the essential ideas and attempts of grey programming [28,46].
Let us denote the model of linear progralnmillg by

where €3 is a grey target. While the vector c is referred to as a grey one whose
elements are grey numbers denoted with 63 ( c i ) , A is a grey matrix denoted
by

g(5) represents a boundary vector at time r which is obtained with GM(1,l)

modelling can be written as follows:

l A G 0 . GM . AGO: b , + 6, (5)
where b , is a time series associated with the 12th coefficient belonging to
boundary vector b, and 2, ( 5 ) means the value of b , to be forecasted at fu-
r,
ture time which constitutes g((r).
Thenf (x * ) means a satisfactory cost and ;. * is a satisfactory resolution
of it, iff.f (x * ) falls in a grey target which is given.
Grey linear prograinining has been brought to practical use, for example,
 Deng J u l o ~ ~ g

to .adjust the structure of village economy 1811 and to make a development

plan coordinating the economy, society and science in Hubei Province.

Grey Prediction Control

The control principle of convetitional control theory, whether classical or

modern, is to control the system behaviour according to the state sample
which has already occurred. It is then referred to as an afterward manner for
control. The demerits of the afterward manner are as follows:

1. Impossible to avert accidents in advance;

2. Impossible to control timely;
3. Weakly adaptable.

The essential idea of grey prediction control [13,29,49, 501 is to control the
system behaviour in advance with the control strategy obtained from the pre-
diction controller [ 6 ] ,based on GM(1,l). Figure 1 illustrates the outline for a
grey prediction control system. It maintains a desired state within reasonably

e
accurate tolerances even though the output J*is varied.

System

-
Control
3
Strategy . -
9Co)- 2
OM ( 1 . 1 )

+Y
Fig. 1. Grey prediction control system.

In Fig. 1 the output of loop 1 (control object) is denoted as jr, the predic-
tion value of the systems behaviour is denoted as 9"' which is obtained from
controller 2, the assigned quantity is denoted as y * and zr is a control strate-
gy coming from loop 3.
The control principle is as follows: a transducer feeds the sampled data to
prediction controller 2, which functions as a computer. The prediction value
9"' as a calculated result of controller 2 is delivered to loop 3, which consti-
tutes a control strategy as compared with y * . The control strategy zr stems
 Introd~rctionto Grey Sjutem Theory

from loop 3 as an input, activating loop 3, and then stored therein. Should
the behaviour y 'O' occur in the future, the stored strategy u will deal with it
and lead the systems behaviour to a desired state.
A practical grey prediction controller has been initiated by [6].

References

[I] Cao Yanying (1987). Modelling for grey forecasting of calamities in geography.
You~l~geogrnphers, 2, 6-1 1 (in Chinese).
[2] Chen Mianyun (1982). Grey dynamic of the systems of a boring machine. Journnl of
Huazl1otzg Unhresitjr of Science and Technologj~, 6, 7-1 1 (in Chinese).
[3] Chen Mianyun (1985). The stabilization and stability of grey systems. Fuzzy Mcrthe-
matics, (Specie11Issue of Grey Systenls) 2, 101-6 (in Chinese).
[4] Chen Mianyun (1988). Grey dynatnic modelling and prediction control of
inacroeconomic systems. In: Grey Sjrstetll, China Ocean Press, Beijing, pp. 21 1-19.
[5] Chen Mianyun (1988). Stability and regulation problems of grey systems. In: Grey
Systems, China Ocean Press, Beijing, pp. 220-8.
[6] Cheng Biao (1986). The grey control on industrial process. Journal of Hunngshi Col-
lege, 1 , 11-23 (in Chinese).
[7] Deng Julong (1982a). Control problems of Grey Systems, S~lster~u a t ~ dControl Let-
ters, 5,288-94.
[8] Deng Julong (1982b). The least parameters controller of decetralized control system.
Actn A~rtonznticnSirzicn, 1 , 4 9 4 4 (in Chnese).
[9] Deng Julong (1982~).Grey systems control, Jotrrnnl of Htrnzhong Ui7iversit)r of Sci-
ence m d Technology. 3, 9-18 (in Chinese).
[lo] Deng Julong (1983a). Grey System theories and measurement futurics, Future crnd
Development, 3, 20-3 (in Chinese).
[Ill Deng Julong (1983b). Grey fuzzy forecast and control for grain. Journcrl of Hucrthong
Uniilersitj~of Science and Tecl11zolog)r,2, 1-8 (in Chinese).
[I21 Deng Julong (1983~).Group of Grey Systems. Jourtzzal of Huazhorzg Universilj~of Sci-
ence nnd Tecl~t~ology. 4 , 15-24 (in Chinese).
[I31 Deng Julong (1983d). Survey of Grey System. World Science, 7, 1-5 (in Chinese).
[I41 Deng Julong (1984a). The theory and method of socioeconomic Grey Systems. Socicrl
Sciences itz Cl~inn,6 , 47-60 (in Chinese).
[IS] Deng Julong (398413). The differential dynamic model (GM) and its implement in
long period forecasting of grain. Exploratior? ofNnture, 3 , 4 7 4 0 (in Chinese).
1161 Deng Julong (1984~).Grey nonlinear programming. Esplorutioi~of Nature, 3, 47-50
(in Chinese).
1171 Deng Julong (1985a). Grey Systems. Z/?et~gzl~ou Hyclrologj~,1, 1-10; 2, 2-9 (in
Chinese).
[18] Deng Julong (1985b). Relational space of Grey Systems. FLIZZY Mntlletnntics, (Special
Issue of Grey Systems) 2, 1-10 (in Chinese).
[I91 Deng Julong (198%). Generated function of Grey Systems. Fuzzy Mutl~er?~ntics. (Spe-
cial Issue of Grey Systems) 2, 11-22 (in Chinese).
[20] Deng Julong (1985d). The G M model of Grey Systems. FL~ZZJ, (Special
Mcrt/~et~z~tics,
Issue of Grey Systems) 2,23-32 (in Chinese).
[2 I] Deng Julong (1985e). Five kinds of Grey forecasting. Ftrzzj~Matken~ntics.(Special Is-
sue of Grey Systems) 2,33-42 (in Chinese).
 Deng Julorzg

[22] Deng Julong (19850. Grey state decision making. Fuzzy Mathematics, (Special Issue
of Grey Systems) 2,43-50 (in Chinese).
[23] Deng Julong (1985g). Grey system and agriculture. Agricrrltural Science in Slzanxi, 5 ,
34-7; 6,29-33 (in Chinese).
[24] Deng Julong (198521). Grey Systenw: society, economy, National Defence Press, Beij-
ing, (in Chinese).
[25] Deng Julong (1985). Grey Systerns Control, Huzhong University of Science and
Technology Press, Wuhan, (in Chinese).
[26] Deng Julong & Deng Qi (1985). On grey and fuzzy decision of lining building of irri-
gation channels. Systern Science and Con~preliensiveStudies in Agriculture 2, 26-30
(in Chinese).
[27] Deng Julong & Zhou Chaoshun (1985). The feedback control and sufficient criteria
for the stability of grey linear time-invariant systems. Fuzzy Matlzematics, (Special
Issue of Grey Systems) 2, 107-1 1 (in Chinese).
[28] Deng Julong (1986a). Grey Prediction and Grey Decision Making, Huazhong Univer-
sity of Science and Technology Press, Wuhan, (in Chinese).
[29] Deng Julong (1986b). Grey forecasting controller. Exploration of Nature, 4 , 59-68 (in
Chinese).
[30] Deng Julong (1986~).The application of Grey System to quantitative investment
strategy. Znvestn7ent and reform, 3,21-5 (in Chinese).
[31] Deng Julong (1986d). Some essential approachs to latent Grey Systems. Systems En-
gineering Tlzeory and Practice, 1, 66-72 (in Chinese).
[32] Deng Julong (1986e). Essential problems for Grey System theory. Journal of Ximlgfarl
Universitj,, 1,5-27 (in Chinese).
[33] Deng Julong (1986f).Grey Systems. Jour~zal of H~mnngsl~iCollege, 1, 1-10 (in
Chinese).
[341 Deng Julong (1987a). T17e esselztial metl~odsof Grey Systems, Huazhong University of
Science and Technology Press, Wuhan, (in Chinese).
[35] Deng Julong (1987b). Properties of grey forecasting model GM(1,l). Journal of
Huazlronng University of Science and Teckno/ogj),5 , 1-6 (in Chinese).
[36] Deng Julong (1987~).Grey exponential law for accumulative generation. Journal of
Huazl~ongUniversity of Science and Teclznolog~~, 5,7-12 (in Chinese).
[37] Deng Julong (1987d). Grey decision making and its use for the determination of irri-
gation strategy. In: Optinzizatiolz Models Using F~rzzySets anrl Possibility TI7eoiy. eds
J. Kacprzyk & S.A. Orlovski. D. Reidel Publishing Company, Netherlands pp.
447-58.
[38] Deng Julong & Zhou Chaoshun (1987). Sufficient condition for stability of Grey Sys-
tems and its application in the analysis of stability of linear time-varying systems. In-
forn~utionand Corztrol,4,24-7 (in Chinese).
[39] Deng Julong (1988a). Properties of relational space for Grey Systems. In: Grej) Sys-
tem. China Ocean Press, Beijing, pp. 1-13.
[40] Deng Julong (1988b). Properties of generating function for Grey Systems. In: Grey
System, China Ocean Press, Beijing, pp. 4-30.
[41] Deng Julong (1988~).The grey exponential law of AGO. In: Grejj System, China
Ocean Press, Beijing, pp. 13-39.
[42] Deng Julong (1988d). Modelling of the GM model of Grey Systems. In: Grey System,
China Ocean Press, Beijing, pp. 40-53.
[43] Deng Julong (1988e). Grey forecasting model. In: Grej) System, China Ocean Press,
Beijing, pp. 54-69.
1441 Deng Julong (1988f). Properties of the grey forecasting model of GM(1,l). In: Grejr
Systen7, China Ocean Press, Beijing, pp. 79-90.
14.51 Deng Julong (1988h). Five step modelling (FSM) of Grey Systems In: Grejf System,
 Irztrod~rctio~z
to Grey Syslen? Tl~eory

China Ocean Press, Beijing, pp. 91-102.

[46] Deng Julong (19881). Grey strategy of situation, In: Grej~Systenz, China Ocean Press,
Beijing, pp. 103-14.
[47] Deng Julong (1988j). Grey decision making of group. In: Grey System, China Ocean
Press, Beijing, pp. 1 15-29.
[48] Deng Julong (1988k). Grey linear programming. In: Grejt Systen~,China Ocean Press,
Beijing, pp. 130-8.
[49] Deng Julong (19881). Grey forecasting control. In: Grey Syste117,China Ocean Press,
Beiing, pp. 138-49.
[50] Deng Julong (1988m). Process adaptation of GM(1,l). In: Grey System, China Ocean
Press, Beijing, pp. 150-64.
[51] Deng Qi, Feng Qingxi & Deng Julong (1985). Underground water level control sys-
tem and optimal grey decision making (irrigation decision making for the People's
Victory Channel). Acta Autolnaticcc Sinica, 2, 131-6 (in Chinese).
[52] Gong Guoqing (1987). Calculating dropping water level with grey model. Jo~rrrzalof
CESU Youth Scierztist Associatiolz, 1, 81-5 (in Chinese).
[53] Guo Hong (1985). Identification coefficient of relational grade of Grey Systems. FLIZ-
zy Matl1e1natics.2, 55-8 (in Chinese).
[54] Guo Hong (1986a). Controlling for Grey System. Joz~rlzrilof Xiangfa17 Unillersity, 1 ,
28-32 (in Chinese).
[55] Guo Hong (1986b). Grey classification of medical diagnosis. E,vyloration of Nature,
4,69-75 (in Chinese).
[56] Guo Hong (1986~). Grey System theory. I~~terdisciy1i11rtr.y Science, 1, 17-8 (in
Chinese).
[57] Guo Hong (1988). Extension of the norm interval for relational space of Grey System
theory. In: Grey System, China Ocean Press, Beijing, pp. 165-70.
[58] Hu Congwu (1987). The trend bf yield with drought or excessive rain in Guangshan
Country. Systern Sciences and CompreRensive Studies in Agriculture. 1, 39-42 (in
Chinese).
[59] Hu Tao & Huang Xiaoqing (1985). Analysis of factors harming the Korean Pine-
applying Grey System theory to the reasearch on ecosystem. Fuzzy Matlie~~~crtics, 2,
5 1-4 (in Chinese).
1601 Hu Xingdou (1987). Decomposition analysis and stability of Grey Systems. Journal
of Beiji~zgInstitute of Tech~zologjr,5,45-50 (in Chinese).
[61] Huang Jianer (1987). Applications of grey statistics to investment decision making of
forestry. Econornic Problems o f Forestry, 2,44-51 (in Chinese).
[62] Huang jimin (1985). Building of grey forecasting model for grain yield prediction in
Jiangxi Province. Journal of Jicmgxi Normal Universitj~,3,47-52 (in Chinese).
[63] Huo Junjiang (1986a). On the applications of Grey System Theory and methods to
history studies. Social Science 10, 64-8 (in Chinese).
[64] Huo Junjiang (1986b). Grey System methods and investigation. Seeker, 4, 94-8 (in
Chinese).
[65] Jiang Ye et al. (1987). Applying grey group to the devision of the suitable climate
ecocondition Panax Ouinguefoliuln grown. Sj~stenlSciences ancl Con~prehensi~le St~m-
ies in Agriculture, 4, 64-70 (in Chinese).
[66] Jiao Jiujiu (1987). Adjustment and control for hydrogeology system's dynamic. Jour.
17nlof CESU Youtlz Scientist Associcrlion, 1 , 74-8 1 (in Chinese).
[67] Lee Senra (1986). Grey forecasting the freight volume for vehicle of railway. Jo~rrnal
U~ziversitjl,1 , 33-5 (in Chinese).
of Xic~~zg;Pnn
[68] Li Biaolin (1987). Models of grey mutation in ecoagriculture. Sysfelz Sciences nnrl
Conzprel~ensiveStudies in Agric~rlture,3, 39-47 (in Chinese).
[69] Li Biaolin & Du Deshou (1984). Prelitninary investigation of application of Grey Sys-
 Deng Julong

tem in ecological engineering the impact of Ladybird Beetle on Cotton Aphid. Jour-
nal of Hzmazl~oi~g Universitj) of Science and Tecl~i~ology, 3, 1-8 (in Chinese).
[70] Li Biaolin & Deng Julong (1984). The grey model of biological prevention and cure
systems of Aphis gossypii Glover. Esplorntio~lof Nature, 3,44-6 (in Chinese).
[71] Li Xueping & Chen Shaohua (1987). Grey adaptation for linear system and adaptive
control by grey forecast. Acaden~icResearcl~,4, 58-64 (in Chinese).
[72] Li Xueping & Liu Huizhong (1987). On scientific methods in Grey System. Esplorcr-
tion of Nature, 4, 125-30 (in Chinese).
[73] Li Zhongmo (1986). Primary application of the Grey System theory in the study of
earthquake forecast. Joutnnl of Seismologj), 4,27-3 1 (in Chinese).
[74] Li Yaoxian & Zhao Yufu (1987). The application of Grey System theory to
long-range prediction. Acta Meteorologica Sinica, 4,489-94. (in Chinese).
[75] Liu Guangji & Luo Gaorong (1986). Grey System predicting theory applied to estab-
lishing equation for forecasting annual consumption of electrical power. Systeins En-
gineering Theory a~~dprrrctice. 3, 38-41 (in Chinese).
[76] Long Yu (19871. Identifying system's factors with grey relational method. Journal of
Xiangfa~iUniversitj~,1, 51-5 (in Chinese).
[77] Luo Jianjun (1987). Quantitative analysis of agricultural advancing in Shanxi Prov-
ince. Systenl Sciences and Con~preliensi~~e Studies in Agriculture, 1 , 4 3 4 0 (in Chinese).
[78] Luo Jianjun, Wang Xuemeng & Nie Hongsheng (1985). Some instances for applica-
tions of Grey System theory to agriculture in Shanxi Province. Agricultural Science in
Slzanxi, 10,24-6 (in Chinese).
[79] Luo Jianjun & Zhang Bingxiang (1986). An analysis on the grey forecast and the con-
trol of grain prodction in Shanxi Province. Actn Agricrrlturcrl Boreali Siizicn, 1, 66-71
(in Chinese).
[80] Luo Qingcheng (1985). Atentative exploration of the Grey System theory to the agri-
cultural input-output simulation. Systerll Science and Con~prekerzsh~e Stzrdies in Agri-
culture, 3, 24-30 (in Chinese).
[81] Luo Qingcheng (1987). Grey linear programming and its appliance for agricultural
structure optimization. Sjtsteni Sciences and Comnprel~ensiveStudies in Agricult~rre10,
49-55 (in Chinese).
[82] Ma Xianyao & Zhen Ping (1985). The forecasting of creep behaviour for low alloy
steel. Fuzz)) Mc~the~natics, (Special Issue of Grey Systems) 2,85-8 (in Chinese).
[83] Ma Xianyao & Zhen Ping (1986). Research on impact strength of ferritical
apheroidal graphite cast iron by grey system theory. Fuzzj~Mcrtl~en~afics, (Special Is-
sue of Grey System) 2,89-98 (in Chinese).
[84] Ma Zhiying & Shi Shuyan (1988). Application of grey relational grade analysis in
cotton lines evaluation. Jorrrnal of Hebei Agric~rlttcral Uniaersitj). 4, 121-8 (in
Chinese).
[85] Meng Wanrong (1985). Application of Grey System theory to forecasting
technological progress in industry owned by the whole peopl in Hubei Province. FLIZ-
z j ~Mcrtkei~intics,(Special Issue of Grey Systems) 2,97-100 (in Chinese).
[86] Meng Wanrong (1988). Application of the Grey System theory to forecasting the
technical progress in the industries of Hubei Province. In: Grejj Sj~steiii,China Ocean
Press, Beijing, pp. 175-9.
[87] Tan Chengye (1986). Correlation degree and fuzzy cluster analysis on strong
earthquake precursors. Journul of Seisniological Resecrrch, 4,409-17 (in Chinese).
r881 Tan Chengye, Ruan Shengna & Yang Guifeng (1986). Discussions on the establish-
ment of the seismological grey model. Jo~rrnnlof Seis~~~ologiccrl Research, 6, 685-92
(in Chinese).
r891 Wang Aicheng & Yu Min (1988). Applications of grey forecasting method to diseases
prediction. Jo~nrznlof Epideinio/ogjr of Cltina. 1 , 4 9 4 2 (in Chinese).
 [90] Wang Min & Jiao Jiujiu (1987). Application studies of hydrogeology with Grey Sys-
tem theory. Jour~~al of CESU Youth Scientist Association, 1, 65-74 (in Chinese).
[91] Wang Mingxi & Liu Wan (1985). Long period forecasting of first frost by Grey Sys-
tem theory. Fuzzy Matl~emcrtics,(Special Issue of Grey Systems) 2, 59-66 (in
Chinese).
[92] Wang Qingyin (1987). Situation analysis of the development of machine building
industry in the City of Handan. Systems E n g h l e e r i ~ ~ g - T I and
I Practice, 2,,43-6
(in Chinese).
[93] Wang Shumin (1985). The grey state strategy of multitarget for agricultural division.
Ftlzzy Matl~en~atics, (Special Issue of Grey Systems) 2, 67-76 (in Chinese).
[94] Wang Ting (1985). Grey functions and relational grade. Fuzzy Mathenirrtics, (Spe-
cial Issue of Grey Systems) 2,77-80 (in Chinese).
[95] Wang Xuemeng & Luo Jianjun (1986). Grey dynamic inodel and its analysis of the
circle. Systenis Scie~lces aricl Conzprel~ensi~~e Studies in Agriculture, 1, 23-6 (in
Chinese).
[96] Wang Zhixue (1988). The grey calculating analysis oof Grand Jujube output in Fup-
ing County. Journal of Hebei Forestuj College, 1,24-32 (in Chinese).
[97] Wei Zhongzhe & Yang Zheba (1986). The macro control model for the teachers of
university. Applicc~tionsof System Engi~~eeri~ig, 1,32-9 (in Chinese).
[98] Xia Jun (1987). Estimating influences by the environment for exploiting the water
sources. Social Science Revie~vof Journal of Wuhm?Hydrology C I I Z ~Power Ziutit~~tc..
1,97-104 (in Chinese).
[99] Xin Jianjun et al. (1987). Applying the Grey System theory to the evaluation lantl
quality. Systenl Sciences and Comnprehensi~reStudies in Agriculture. 4 , 39-45 (in
Chinese).
[I001 Xiong Jiling (1988). Analysis of potentials of the load of judicial system of China.
In: Grey S y s t e ~ ~China
i, Ocean Press, Beijing, pp. 195-210.
[loll Xu Dianxiong (1986). Analysis of GM (1,l) model assembly language algorithm.
Jou171aIof H~mligslliCollege, 1, 25-3 1 (in Chinese).
[I021 Xu Shaoyan (1988). Analysis of GM (1,N) forecasting model and its applications.
In: Grey Syste~ii,China Ocean Press, Beijing, pp. 180-94.
[lo31 Yan Wenbing & Chang Lianfang (1986). Forecasting and analysis for the average
population incoms of farmers of Shanxi Province in 1986. Prediction of E2~penrlit~ire
Tendency, 6,42-4 (in Chinese).
[I041 Yan Wenbing & Luo Jianjun (1986). An analysis of dominent economy and a selec-
tion of the most economic model on the region. Sjlsten~Scie~lcescr~ldCon?prehe~uille
Studies in Agriculture, 1, 60-6 (in Chinese).
[I051 Yi Desheng (1987). Grey inodel (GM) and technician forecast. Esyloration of No-
ture. 4 , 75-8 1 (in Chinese).
[I061 Yi Desheng (1987). Grey series forecasting. Foreccrstblg, 4, 14-7 (in Chinese).
[I071 Yiao Jianlnin (1987). The problem for determining the accuracy of the grey forecast
value. S~rsteni Sciences and Con~prelie~ui~le 3, 18-22 (in
Studies A Ag1.ic~11tur.e.
Chinese).
[I081 Yu Liyun (1986). The grey forecast for urban domestic wastes. E I I I ~ ~ ~ O IPol1~1- IIII~III
ti011n~zdControl, 5, 7-9 (in Chinese).
[I091 Yuan Jiazu (1988). The grey prediction model of the Populus sp growth increment.
Jo~rrrialofHebei Forestry College, 1, 15-23 (in Chinese).
11101 Yang Yunhua & Wang Ting (1985). Measurement and operation of grey concept.
Aizzy Mothenirrtics, (Special Issue of Grey Systems) 2, 81-4 (in Chinese).
[I 111 Zeng Guoning (1987). Applications of Grey System theory and methods to estimate
the business benefits. De~~elopnient St~iclies,4, 54-6 (in Chinese).
[I121 Zhan Yihui (1982). The synthetic controllability of Grey System. Jo~rrnril of
 6 , 11-6 (in Chinese).
Hirtrzliong Utziilersity of Science arid Tecl~nolog~~,
[I131 Zhang Qifan (1982). A preliminary study of the whitening of the Grey System. Joarr-
iznl of Htrazhorig U~iiversitj~ of Science nlid Tecl~nology,3, 19-22 (in Chinese).
[I141 Zhang Bingxiang & Luo Jianjun (1985). Grey dynamic model (GM) with applica-
tion to grain prediction. Systerii Engineering, 4,36-7 (in Chinese).
[I151 Zhang Huimin (1986). A grey forecast of tendency to dry from 1985 to 2000 in
Luoyang Area. Systeni Sciences and Conzprekensive Studies in Agriculture, 1, 36-40
(in Chinese).
[I161 Zhang Dianqing (1986). Grey relational analysis of the shape function and the level
of body quality for youngsters and children. Sport Science in Guizhou Proi~ince,2,
1-5 (in Chinese).
[I171 Zhang Guangzheng et al. (1987). System analysis and forecast of farmer's income
and its construction with Grey System theory. System Sciences and Compreheiisii~e
Studies in Agriculture, 1, 57-70 (in Chinese).
[118] Zhang Jizhong et al. (1987). Quantitative research in classification of countryside
economy in Shanxi Province. Systeni Sciences and Coniprehensiile Studies in Agricul-
ture, 1 , 17-23 (in Chinese).
[I 191 Zhao Songnian (1987). On decision of macrosysterns. Esploration of Nature, 4,
1 19-28 (in Chinese).
[I201 Zhao Yunsheng (1987a). Grey modelling for the parameters of ventilating dudt.
Jo~wntllof CESU Youth Scierztist Associtrtior~.1 , 55-65 (in Chinese).
[I211 Zhao Yunsheng (1987b). Theory of establishment of grey models and its uses in cal-
culation of ventilation variables. Nor?ferrousMetals, 4, 18-25 (in Chinese).
[122] Zhou Ru (1986). The estimation of satisfactory soil constitution by means of grey
clustering in North of the Huihe River. Ap~~licotioti of'Sj~steinEngiiieeri~ig,1, 40-52
(in Chinese).