COMPREHENSIVE METHODS OF SYSTEM ANALYSIS,
INFERENCE , SYNTHESIS AND MODEL SETS FOR
STUDYING THE SOCIO - ECONOMIC - ECOSYSTEM
Professor Qifan Wang
School of Management,Fudan University,
Shanghai, 200433,People's Republic of China
ABSTRACT
The author explores a comprehensive’ methods of systems
analysis, inference and ‘synthesis and model sets for
studying complex system. These methodologies and model sets
can be used in studying the development strategy and
planning of socio-economic-ecosystem. It has be successful
in the study of Pudong New Economic Zone of Shanghai.
INTRODUCTION
1. The Characteristics of a Complex System
Since the socio-economic-tech-ecosystem is a high-
nonlinearity, high order, large scale, multi level and
multi-loop feedback complicated system, it is impossible to
use traditional methodology or linear analysis to deal with
a .complicated system, nor can it be settled effectively by
only a simple new theory and method of systems science.
2. Theory and Methodology
Based on the wide study of theories and applications in
1980's the author synthetically applies economic theory,
management science, economic mathematics and system
dynamics, .and draws the equintessence of other theories of
systems science, to put forward these methodologies’ and
model sets. These are comprehensive theories, methods and
means, which include the combination of qualitative and
quantitative analyses, systems thinking, synthesis and
inference, system forecasting, planning and optimization,
etc. ( may be intellectual), ard also can use to research
non-linear and complex time-change large system. In
shorthly, that is a combination of qualitative and
quantitative, system thinking, synthesis, inference methods
and means.
DESCRIPTION OF THE THEORY AND METHODOLOGY OF THE APPROACH
1. The Theoretical Foundation of the Approach
(1) Methodology
554 SYSTEM DYNAMICS '93
Its methodology is system methodology which put the research
object in system form, and insist on the view of systemetic,
dialectical, developmental and historical materialist.
(2) Basic Theory
It is based on economic theory, management theory, economic
mathematics, system dynamics, non-linear system theory,
large-scale system theory and system theory which is being
formed.
(3) Applied Technics
It uses computer sciences, simulation, the science and
technic relative to the research object.
2. Description of a System
(1) We basicly use the high-order non-linear random partial
differential equations, which are a set of evolutional
equations with time-space structure, for the part which can
be quantified among real system. Because of the difficulty
in using mathematic and simulation language, we simplify
those equations into deterministic differential equations in
model-setting, and use the noise-function to describe the
effects of uncertainty random facters in real system.
(2) We apply partial quantitative, partial qualitative or
qualitative method to deal with non virtuous-structure which
can not be describled by differential equations and other
mathematical functions in modeling. We use approximately
virtuous structure in stead of bad-structure making bad-
structure to be changed into relative virtuous-structure, or
combine qualitative and quantitative methods to quantified
some parts of qualitative problems. We use qualitative
method to deal with some parts of un-quantitative and un-
partial-quantified problems.
3. The Composition of the Approach
Thought, method and frame of "systemetic method,
comprehensive model sets "has been proposed in
introduction, and the same time the approach also includes a
series of "hardware". After it has been set, it is a
synthetical system which contains model base, data base, and
man-machine interface, etc. The appoach cah be classified
two categories by the.research object -- macro and micro,
but it has the same thinking, method and frame. As
following, we simply introduce model base which we have put
greater efforts and have many results.
(1) Comprehensive Model Base with Multi-Models
SYSTEM DYNAMICS '93 555
(a) Various common used quantitative models such as
econometric model, system dynamic model, industrial
correlation analysis. model, dynamic input-output model, time
series model, economic control theory model, etc.
(b) Combination model of econometrics, mathematic economics
and system dynamics.
(c) Integrated model with combination of industrial
correlation analysis, dynamic input-output and system
dynamics.
(d) Combination model of non-equilibrium self-organized
theory and system dynamics.
(e) Coordinative use of optimization theory model and some
of above-mentioned models.
(2) Technique One of Simplifying Modeling
For a research object, to set up a key model with some sub-
models and combine to use those models according to the
needs of research. This is a good, saving and quick
technical way, according to several decades international
experiences and our one decade practices, as well as
combining our. national conditions in modeling.
(3) Technique Two of Simplifying Modeling
To set up generic substructure (about 10) for some kinds of
system, then, according to requirements of modeling, to
assemble and form a suitable models. This technic can be
used with (2).-
(4) Classifying Models According to Order and Number of |
Equation
Table 1
examples
large | >75 | >1500 !SDNM of the U.S., 200 orders,
H inumber eqution over 4000
Key model of Chinese S.D. national
middle; 20- 70} 500- 1500 !model series, 50 orders, number of
H H tequation about 1000
556 SYSTEM DYNAMICS '93.
Models can be classified into three kinds -- large, middle
and small by orders-and number of equations. As shown in
table (1), it should be selected by purpose and requirement
of modeling.
(5) Classifying Models According to Span of. Operating Time
Table 2
time span { years } examples | main suitable model
short-ter 1 <1 tquartly and ‘tecomometrics,time
‘ tyearly {series, input
H \forecast i-output etc.
; 1<T<5 short term ithe same as above
i iforecast and}
t tplan H
middle-term 5<T<20; national teconometrics
turban and '(<10yrs) dynamic
egional long! input-output
erm forecastisystem dynamic &
tits combination
long-term {long 120<T<50inational,cityimainly use system
t ' tregion long- ‘dynamics and draw
H H iterm forecast iother method to
t ' tand plan tform an inte-
' H t tgrated model
jultra 1 T>50 {ito study the ithe same as
tlong H Western econ-iabove
H t tomic. longwave}!
t i ecological '
t t environment, |
t 1 population, |
: H inatural re- {|
H i tsources, ‘
H H isocial H
H ' iculture and {
t H ithoughts,etc |
We found outhow to select a reasonable running time span of
a model for a specific purpose and problem is still a
confusion for many decision maker in our country. According
to the feature and dynamic mechanism and behavior of socio-
economic system, we proposed three modeling time spans ~~
short-term, middle-term and long-term as shown in table 2.
Of course, the classification is rough and not quite
accurate, and has proper overlapping each other.
We hold that we should at least use middle-term model to
SYSTEM DYNAMICS '93, 557
study economic cycle in our country and Western business
cycle, kuznez cycle, as well as city and regiona] developing
strategy and plan. For urban and regional development
strategy and plan, it should have long-term, or even ultra-
long-term. model to study its long-term restricting and
driving factors. As for national and world socio- economic
development long-term development strategy and plan,
involving such as the Western economic long wave,
population, natural resource, energy, ecological
environment, change of. social culture and thought and its
effects, etc., only the ultra-long-term model can. be
qualified.
We also hold that , for.. improving decision, it is
indispensable to establish a series of coordinate short-
term, middle-term and long-term models for a nation,a
region anda large city, and further to set up an approach
of "“systemetic method, comprehensive model sets" as this
paper introduced.
4. The Main Functions of the Approach
(1) It can be used to forecast and plan in short-term,
middle-term and long-term for several kinds of socio-
economic system, and to systemetically and synthetically
study middle-term and long-term development strategies.
(2) It can be used to synthetically study complicated,
high order and multi-department large system in macro- and
micro-level.
(3) The approach owns: well-open and adaptability itself.
It is convenience to use. various data, information,
experiences and knowledge, and also to draw and merge the
equintessence of other disciplines and science theories,
making the approach itself. to be improved and consummated
unceasingly.
(4) The approach can be used as a tool to develop the
combination effects and creativity among decision maker,
experts and modeling personnel, making the organization and
system (from nation to enterprise) managed by decision maker
to. quickly. become a learning organization possessed well
dynamic -structure, self-study, self-adjustment, strong
adaptability and full of energy.
(5) The relative errors of quantitative analyses of
Shanghai Pudong development from different mothods are less
than 5-10%, after using the approach.
SOME TECHNICAL PROBLEMS ABOUT MODELING AND THE APPROACH
ESTABLISHMENT
558 SYSTEM DYNAMICS '93
For these issues, such as definition of system boundary,
description of system. treatment of endogenous and
exogenous varialbes, estimation of parameter, sieving and
technical treating of data, . test of model, etc., the author
does not mention verbosely, since it belongs to bssic
knowledge of space and time. Please refer to the reference.
PROSPECTS OF PNEA IN 2000
-- An Example of the Application of the Approach
This paragraph studies the blueprint of Pudong New Economic
Area(PNEA) of Shanghai in 2000 by means of our approach .The
development. of Shanghai are influenced by its main factors
such as transportation ,population and technology etc.Under
different conditions, this paper puts forward two different
development strategies. PNEA should plays a great role in
rebuiding Shanghai's economic functions.
1.Background
With no exceptions that the development of the Gities all
experience the birth,growth,culmination and the
decline.Shanghai is now just on the track of this regulation
its further development. It's of crucial importance to study
the development. demand.
The development of PNEA or the East. Shanghai .New Economic
Zone will be not only an important part for. Shanghai's
economic growth,but also a great event for China's
modernization. The background of the development. of PDNA
can be. viewed from
various aspects as follows.
First,as the largest industrial and commercial center of
China and the Major contributor to the national
revenue,Shanghai has .played .a.decisive role in China's
economy over the last four decades.Shanghai itself has also
developed an industrial system. which is. comprehensive
integrated and of a relatively high technology by
nature.However,it appears that some problems have. occurred
in recent years which have unfortunately. hampered Shanghai's
continuous economic progress.
(1).Its tertiary industry has sustained a slow growth
rate,making its urban function imperfect.Shanghai is still
an industrial city,but,as is widely recognized,it should
become an economic,trade and financial center of China and
even one of Southeast. Asia.
(2).The ‘superiority of leading industries in Shanghai is
declining.The competitiveness of Shanghai's products is
weakening in both the domestic and international
SYSTEM DYNAMICS '93 559
‘markets,which caused substantial drop of sales and. trade
benefit.
(3).The provisions of raw mateial and energy for industries
in Shanghai were not regular.Her share both in the domestic
market and in the export trade was declining in recect
years.
Now the question are raised in the light of background
analysis:
(1).What kind of development strategy shall we adopt?
(2).What kind of structure shall PNEA have?
(3).What is the GNP growth rate to be proprietely assigned?
(4).What is the proper population size for the PNEA?
(5).What the maximum capacity of transportation &
communication should be for the development.
2.Modeling Conceptualization
Our approach is used in the following analyses:
(1).The macro relationship among the flows of manpower,raw
materials,funds,information,technology and ecosystem of the
whole system.
(2).The dynamic analysis of the coordination and equilibrium
between the east and west regions of Shanghai. incorporated
in the sysyem .
(3).The dynamic analysis of the adjustment of industry
structure and investment structure.
(4).The dynamic analysis of endogenous and exogenous factors
that limit the generation and development of the system.
(5).The dynamic analysis of the development and modification
of the western and eastern parts of the system as well as
their influence and effects on domestic socio-economic
development.
The models are developed according to industrial structure
and inter-industrial relations:
(1).Classification and definition of various industries and
their inter-relationship will be presented in terms of
correlation coefficients ,the later will be obtained both
from experiences gained by other countries or it can be
produced by Delphi method.
(2).Spacially ,it can be devided into two parts that have
interactions and inter-flows.
3.Model Framework
To study the PNEA, we created our model made of two parts
,the estern part and western parts.The basic structures of
these two areas are similar .From this model,we analyze the
transportation ,capital distribution ratio,the flows of
population of future of Shanghai.The basic structure of each
560 SYSTEM DYNAMICS '93
part includes industries,population,GNP,transportation and
technology.
Here are some basic assumptions: (for plan I)
(1).The initial simulation time begins in 1988.All the price
indexes are based on 1988.
(2).The immediate consumption coefficients are constant.
(3).The population influx of Shanghai is 2 million.
(4).The total fixed capital of 2000 is 52.85 billion yuan.
(5).The productivity of 1990 is 8637 yuan per person which
will be doubled every five years since 1990.
4, The Results of Base-run
The GNP of Shanghai is 65.55 billion in 1988 while 200
billion in 2000.The average growth rate during that period
is 9.74% for plan I and 7.35% for plan II with 153.5 billion
GNP in 2000.The GDP of PNEA in 2000 are detailed below:
Table 3. GNP in PDEA in 2000
Indexes
GNP in 2000 !
First: Industry H
Secondary Industry! 248.5 130 1 48. 54.0
Tertiary Industry 256.3 107 + 50.1 44.6
5. Policy Recommendation
Based on the above simulation and policy tests ,we conclude:
(1).According.. to the current condition,the goal we set in
2000 will be achieved.
(2).The transportation capability is main factor limiting
the development .It needs a great attention.
(3).Investment distribution ratio between industries and two
areas are effective policy orientation that can adjust the
industrial structure. We confess that it is difficult to
have a breakthrough within a short-limited period. However,
if we try to assimilate this obstacle in the development.
process itself,it will be encouraging.For example,the high
demand for transportaiton resulted from the development of
secondary industry may be overcome by emphasizing the
development of tertiary industry.This will release the
stress on the transportation capability. Furthermore,the
tertiary industry calls for relatively few labor
force which will be obtained locally.
SYSTEM DYNAMICS '93, 561
(4). The population in~ Shanghai is a vital aspect for
control. This potential crisis needs consideration:for the
decision makers.
Here we prospose:
(1).Increase the investment ratio in tertiary industry .Try
to build Shanghai from a industrial city to a
multifunctional central modern city.
(2).Solve the transportation problems to expediate the
economic development. :
(3).The coordinated development between the eastern and
western parts should be adjusted.None of ‘them can be
prejudiced.
In all, the development of the western and the eastern parts
should be coordinated and complemented from each other to
ensure further development in Shanghai and realize its goal
of an open, multifunctional and modern metropolis.
REFERENCES
Forrester Jay W. 1969.Urban Dynamics,The MIT Press
Levy John M. 1985. Urban and Metropolitan Economics,McGraw-
Hill Book Co.
Johnston J.1972. Econometric Methods,McGraw-Hill Kogakusha
rLtd, 1972
Senge,Peter M. 1990. The Fifth Discipline :The Art and
Practice of the Learning Organization, Doubleday.
Wang, Qifan -1987.Theory and Applicaiton of System
Dynamics,;New Times Press
Wang,Qifan .1988.System Dynamics,Tsinhua University Press
562 SYSTEM DYNAMICS '93
