About GAIA GAIA Case Studies Global GIS Agenda 21 Country Data Model Database

China:
Sustainable Urban Development - Guilin Case





Contents









Basis

    Agenda 21: Chapter 7 : Sustainable Human Settlements
    Habitat Agenda









Key Words

    GIS
    Decision analysis
    Urban land use
    Water pollution
    Air pollution
    Urban traffic









Overview

By the year 2000 more than half of the world's population will live in urban areas. A largely urbanized world implies that sustainable development will depend very largely on the capacity of urban and metropolitan areas to manage the production and consumption patterns and the transport and waste disposal systems needed to preserve the environment[2]. Since the UN conference on Environment and Development in 1992 and Habitat II in 1996, sustainable urban development has been widely concerned in the world. In China, with the rapid economic growth, a great movement of urbanization is spreading out in the country with about a quarter of population and one fifth land in the world. It's clear, urban study for China is urgent and very important for the future urbanizing world.

This case study is abstracted from the project of urban planning in Guilin city. This city is praised with two "crowns", that is, "international scenic tourism city", "national historic cultural city", for its city scenes blending with mountains, lakes and rivers, and its historic and cultural remains for two thousands years ago. Especially, the well known magical Lijiang river flows through this city, and the most picturesque area along the river is located at the downstream of the city. However, Guilin city and natural environment along Lijiang river are endangered by the rapid urbanization and industrialization.

Water with mountains forms the beautiful scenes in Guilin areas. In spite of water quality in Lijiang river still remains good condition, the future of Lijiang river is not very optimistic. Lijiang river is the main water resource of the city and the final river to carry the discharges from the city. And rainfall is the main resource of the river. With the economic growth and population growth in Guilin, water amount in Lijiang river has got less and less, and water quality has descent tendency. Among lots of influencing factors on environment, industry development is an elementary one, and it can be controlled by city government with development plan, policies and rules.

Usually, urban planning is revised every 5 years, and it must be carried out in city construction according to national laws. Therefore, urban planning is a key step and a very important chance towards sustainable urban development. This case study will have decision analysis on industry structure optimization, and location optimization for industry and residents in the process of urban planning. Industry structure optimization will be carried out towards less environment pollution, less water consumption, less energy consumption and more industry income, within the environmental capacity, and keeping proper growth speed in the situation of rapid economic growth in south China. Both environment and traffic are two major urban problems. Different human being's activity type and scale in different areas will have different influence on city environment quality and traffic condition Location optimization will be done to have a rational distribution of various industry sector and residents on urban areas, in order to get better environment quality and traffic condition.



OBJECTIVES

This case study aims at to set up a decision analysis framework of economic analysis and location optimization for urban planning.









Topography


(1) Height Above Sea Level:

    Average height above sea level (counties excluded ): 150 m
    The maximum height(counties included ): 1,778 m.


(2)Hydrology

    Total amount of water resources(counties included, 1994): 8300 million cu.m
    Total length of rivers (counties included): 692.97 km
    Total water area (counties included): 3971.0 sq. km

    Major rivers:
    River Name Length(km): counties included counties excluded
    Lijiang 116.0 40.00
    Taohuajiang 19.00
    Lianfengjiang 34.67
    Ningyuanhe 2.00
    Xiaodongjiang 6.00
    Nanxihe 9.00
    Yijiang 85.05
    Longshenghe 63.15
    Yulonghe 37.00
    Dayuanhe 26.70


(3) Vegetation

    The area of green space in developed urban district (1994): 14.11 sq.km
    The vegetation rate of green space in developed urban district(1994): 32.6%
    The vegetation rate of woodlands(counties included): 30.9%

    The area of woodlands (counties included):
      Year Area
      1990 1,051
      1991 1,124
      1992 1,197
      1993 1,270
      1994 1,296


(4) Industry:

    Gross industrial output value in 1994( million RMB)
    counties included counties excluded
    light industrial 3,992.31 3,560.03
    heavy industrial 4,500.08 4,008.99
    TOTAL 8,492.39 7,569.02

    The sum of the production of following 8 industry sectors:
      machinery industry
      food processing
      chemical industry
      medical industry
      textile industry
      rubber products
      electronic & instruments industry
      metal smelting & pressing
    is over 81% of total industry production.


(5) Infrastructure(1994)

    1) Transportation

      The total length of paved roads(counties included) : 1,699 km
      The total length of paved roads(urban district) : 311 km
      The total area of paved roads(urban district): 284 sq. km

      Passenger traffic(1,000 persons):
        counties included counties excluded
        Railway 3,010 3,010
        Highway 12,600 3,150
        Air 710 710
        Waterway 980 850

      Freight traffic(1,000 tons):
        counties included counties excluded
        Railway 1,050 1,050
        Highway 4,800 4,320

      Number of civil motors: 40549

    2) Telecommunications facilities(1994)
      counties included counties excluded
      urban switchboards capacity: 135,530 75,550
      number of urban telephones: 63,967 60,678
      rural switchboards capacity: 2,332
      number of rural telephones: 1,021


(6) Agricultural (counties included, 1994)

    Cultivated land(hectares): 63,520
    Output of Grain(tons): 341,266









Climate


1) External Air Temperature (counties excluded)
    The mean air temperature of 35 years: 18.75 degrees C;
    The maximum (Aug. 13, 1953): 39.4 degrees C.
    The minimum (Jan. 12, 1955) : - 4.9 degrees C.
    Monthly distribution (1994)
      Month degrees C
      Jan. 9.2
      Feb. 8.7
      Mar. 11.9
      Apr. 20.1
      May 24.9
      Jun. 26.1
      Jul. 27.3
      Aug. 26.9
      Sep. 24.4
      Oct. 18.6
      Nov. 17.2
      Dec. 11.5
      average 18.9

2) Wind Speed(counties excluded)
    The mean wind speed of 33 years: 2.57 m per minute
    The maximum (Aug. 11, 1983): 29.5 m per minute

3) Relative Humidity(counties excluded)
    The mean relative humidity of 35 years: 75.85 percentages

4) Total Daylight(counties excluded)
    The average of annual total daylight of 34 years: 1553.09 hours per year.
      Year total daylight(hours per year)
      1990 1509
      1991 1339
      1992 1668
      1993 1420
      1994 1888

5) Solar Radiation(counties excluded)
    The annual solar radiation is 244 million Joules per sq. metres

6) Rainfall(counties excluded)
    The mean total rainfall of 35 years: 1894.39 mm.

    Monthly distribution in 1994 (mm):
      Jan. 21.8
      Feb. 113.8
      Mar. 127.6
      Apr. 223.1
      May 280.3
      Jun. 601.8
      Jul. 252.1
      Aug. 330.6
      Sep. 41.2
      Oct. 172.9
      Nov. 15.0
      Dec. 86.1
      TOTAL 2266.3









Demography


1) Population size
    Population (counties included, at the end of 1994): 1,282,007
    Population (counties excluded, at the end of 1994): 550,493

2) Age Structure
    percentage of age group(%,1990)
    0-- 9 17.86
    10--19 19.15
    20--29 20.61
    30--39 15.85
    40--49 10.20
    50--59 7.97
    60--69 4.98
    70--79 2.51
    80--89 0.80
    90-- 0.07

3) Ethnic Mix (end of 1994)
    Nationality composition at the end of 1994 (%)
    Counties included Counties excluded
    Han nationality 91.45 92.90
    Minority nationality 8.55 7.10
    Zhuang nationality 5.32 3.28
    Yao nationality 1.78 1.08
    Hui nationality 0.99 1.96

4) Number of Households(at the end of 1994)
    Counties included: 313,025
    Counties excluded: 141,475

5) Average Household Size
    urban residential household size
    1990 3.50
    1991 3.33
    1992 3.32
    1993 3.22
    1994 3.05

6) Number of Sq. m of Living Space per Inhabitant
    The number of sq. m of living space per inhabitant in urban district is 6.34 in 1993 and 6.81 in 1994.


Economic Characteristics:


1) Number of Jobs (farmer excluded, 1994)
    counties included: 239,041
    counties excluded: 209,000

2) Employment by Industry/Sector (counties included, 1994)
    Number Percentage(%)
    Primary industry 369,700 52.4
    Secondary industry 162,300 23.0
    Tertiary industry 173,600 24.6
    TOTAL 705,600 100.0

3) Per Capita Income (1994)
    The per capita income of urban residential: 4,674.01 RMB
    The per capita income of peasant: 1,499.75 RMB

4) Economic Growth Rate
    GDP growth rate(%)
    Year counties included counties excluded
    1990 3.73 2.48
    1991 13.19 15.60
    1992 17.80 18.11
    1993 22.05 23.35
    1994 13.41 10.13

5) GDP and its composition(1994)
    Counties included:
      GDP: 5,195 million RMB
      Composition:
        Percentage(%)
        Primary industry 16.7
        Secondary industry 44.4
        Tertiary industry 38.9

    Counties excluded:
      GDP: 3,881 million RMB
      Composition:
        Percentage(%)
        Primary industry 4.7
        Secondary industry 52.2
        Tertiary industry 43.1

6) Tourism

    Income of tourism (1994): 89 million RMB

    Number of international tourists:
      Year Number
      1985 336,435
      1986 356,853
      1987 500,306
      1988 479,378
      1989 306,583
      1990 484,929
      1991 429,113
      1992 508,638
      1993 439,111
      1994 354,465

    Sum of international tourists (1973~1994): 5,220,000









Current Status of Urban Land Use


1) Total Area
    Total area of developed urban district: 46.3 sq.km(1994)

2) Urban Land Use Composition
    Urban land use composition (developed urban district, 1993)
      Type Percentage (%)
      Residential 32.81
      Public Faculties 9.90
      Industrial 32.55
      Warehouse 7.03
      Transportation 6.51
      Roads&Square 3.65
      Municipal utilities 1.30
      Green space 6.25

3) Per Capita Urban Land
      Per capita urban land(1993)
      Type per capita area (sq.m)
      Residential 23.46
      Public Faculties 7.06
      Industrial 23.23
      Warehouse 5.02
      Transportation 4.65
      Roads&Square 2.60
      Municipal utilities 0.93
      Green space 4.80
      TOTAL 71.38

4) Current Status of Urban Industrial Land Use


5) Current Status of Urban Residential Land Use










Problem Description I (Industry structure optimization)


For the property of Guilin city defined as "international scenic tourism city" and "national historic cultural city" by the state council, and Lijiang river being major object to be protected in this area, the development of the city must keep consistent with it. Regulating industry structure of Guilin is a very important measure for both environment protection and economic development.

Lijiang river is the main water resource of Guilin city and the final river to carry the discharges from the city. Guilin city is the only city on the upstream of Lijiang river. So, water environment in Lijiang river is sensitive to the development of Guilin, especially, industry scale, industry discharge and industry water consumption. In the situation of rapid economic growth in south China, development in Guilin is unavoidable, but, its development must be well planned towards less industry discharge, less industry water consumption, less industry energy consumption and more industry income.

On the other hand, it is necessary to regulate industry structure in order to get stable economic growth and further to support environment protection, because tourism income in Guilin is not stable, and industry has not well developed and organized. Guilin has the most of industry types, covering 34 sectors of 39 sectors in the state standards for industry sector, but the overall scale of industry is small; There are many industry enterprise up to 771, but their scale are small and distribute in many urban areas, i.e. the industry production per employee is even smaller than the average value of the country; The ratio of technical progress contribution to industry growth only is 27.7%, less than 35%, the average value of the country, and there are a few of famous brand of products.

In this case study, industry of Guilin is classed into 9 sectors, they are,
    electronic industry
    machinery industry
    textile industry
    food processing
    chemical & medical industry
    rubber products
    metal smelting & pressing
    others with less pollution
    others with heavy pollution

The objectives of optimization are,
    1. maximum of industry revenue
    2. minimum of industry environment pollution
    3. minimum of industry water consumption
    4. minimum of industry energy consumption

The main constraints are,
    1. Total industry discharge should be less than a planed value which depend on environment capacity, treatment ability and the residents discharge amount
    2. Economic system mechanism (input-output relations)
    3. Overall industry growth speed
    4. Special policies for some industry sector

The industry structure in year 2000, 2010, 2020 will be generated with a mathematical programming model together with a interactive model testing, knowledge refining and problem solving process.









Industry Structure Optimization


For the features of environment requirements and industry development problems Industry structure optimization model is constructed as the followed. It is implemented with a mathematical programming model.

There are two levels of objectives in the model, they are,
    1. maximum of industry revenue
    2. minimum of industry discharge, include
      2.1 waste water amount
      2.2 COD amount
      2.3 SO2 amount
    3. minimum of industry water consumption
    4. minimum of industry energy consumption
Preference on objects are represented by weights. And the sum of weights in same level should be equal to 1. The object values in year 2000, 2010, and 2020 are in consideration. In order to make every object value in same numerical level, the ratio of calculated value for an object to its planed value(or forecasted value) is taken as an object value in the model.


The constraints in the model consist of the following parts.

(1) Environment capacity
COD is selected as controlled index for water environment quality, and SO2 is selected as control index for atmosphere quality. The sum of COD amount discharged by every industry sector should be less than planned industry discharge amount. The sum of SO2 amount discharged by every industry sector should be less than planned industry discharge amount on SO2 also. The planned industry discharge amount is obtained according to natural environment capacity, treatment amount, and residents discharge.

(2) Input-output relationship among industry sectors
Industry structure regulation process must obey to economic systems laws. Input-output analysis is applied to describe the interaction relationship among industry sectors in development process. In principal, supply should be balance with demands. In industry structure optimization model, the production of each industry sector should fall into a possible demand range which is related with the market ratio of products of each sectors in Guilin market and outside market. Demands include consumption, investment, and exports.

(3) Industry growth speed
In industry structure regulation process, some industry sector will growth with higher speed, some with lower speed, even negative growth, but the overall growth speed should be as planned in socio-economic development long term plan.

(4) Special policies for special conditions
For examples, metal smelting and pressing should have only little growth because Guilin has few mineral resources; "others with heavy pollution" should shrink; "others with less pollution" should have lower growth speed to make mandatory industry sectors well developed to form scale economy.


After constructing model, it is can be utilized to carry out industry structure optimization together with an interactive model testing, knowledge refining and problem solving process. At the beginning, knowledge contained in the model often is incomplete, inconsistent, and even incorrect. The programming model testing process can be viewed as a process to acquire new knowledge, to find inconsistent knowledge in model. This process make decision maker, analysts and domain experts to refine knowledge in problem solving situation, and may make them get new ideas and insight about the problem to be solved. Some new constraints, playing the role of "heuristic rules" in optimization, may be constructed based upon experts' experience, to lead the solution directly close to practice, to make search limited in a small space. This process integrates experts' experience, intuition, and mathematical model to search the optimal alternatives for industry structure.









Problem Description II (Location optimization)


For the topography characters of Guilin city naturally divided by hills, rivers and lakes, the city layout in urban planning appears in style of multi-centers and desecrated clusters. Each cluster is a relatively isolated area with integrated functions. Guilin city is divided into seven clusters, they are, North cluster, Center cluster, South_Center, East cluster, South cluster, South-East cluster, West cluster, as shown in the following figure.


Environment and traffic are two major urban problems. It is quite clear, different human being's activity type and scale in different clusters will have different influence on the environment quality of Lijiang river and Guilin city, and have different influence on traffic condition which has influence on environment also. This case study try to have a decision analysis on "cluster land use alternative" which will define activity type and scale in every cluster in year 2000, 2010 and 2020, to have better environment quality in Lijiang river and Guilin city, to get better traffic condition. It is an approach to combine urban planning and industry development plan together.

The human being's activity type and scale in a cluster can be represented by industry sectors and their production that will happen in the cluster, and how many residents who will live in the cluster. The following industry sectors will be in the consideration,
    electronic industry
    machinery industry
    textile industry
    food processing
    chemical & medical industry
    rubber products
    metal smelting & pressing
    others with less pollution
    others with heavy pollution
and the following tertiary industry sectors:
    commerce
    1st hierarchy (but exclude commerce )
    2nd hierarchy
    3rd hierarchy
    4th hierarchy
The 1st hierarchy of tertiary industry includes transportation, communication, commerce, and warehouse; the 2nd includes finance, insurance, real estate, consulting service, and citizen service; the 3rd includes science, education, cultural, health, and sport; The 4th includes government, military, and police.


The decision analysis process has four parts:

(1) Environment impacts assessment
For a cluster land use alternative, the concentration of COD, SS, and Oil in Lijiang river, and the concentration of SO2 and TSP in Guilin city will be evaluated.

(2) Traffic evaluation
For a cluster land use alternative, a kind of relative turnover value of freight traffic, passenger traffic for employment, shopping and sightseeing will be evaluated, and the entropy of those relative traffic which reflect the balance degree of traffic amount distributed in urban areas, will be evaluated also.

(3) Alternatives generated with optimization algorithms
A set of cluster land use alternatives will be generated with optimization techniques. The objects of optimization are better environment quality and traffic condition, and the main constraints are: the environment quality evaluated for an alternative should meet the standards in environmental planning; the sum of each industry sector production in a cluster in 2002 should be less than the ultimate overall industry production defined in urban planning, which reflects the land capacity for human being's activities; the sum of sector production in a cluster should be equal to the sector production in industry development plan or socio-economic development plan, the number of resident is similar.

(4) Interactive decision making
The basic set of optimal alternatives of location optimization generated with optimization techniques usually is obtained on some assumptions, such as, decision maker's preference on objects, assumptions on the future situation and problem modeling, or simplified presentation of problems and so on. In fact, only a part of decision objects are put in models, for the reasons of difficulties on optimization for nonlinear equations, difficulties on quantification of some factors, and other human being factors. So, these alternatives frequently are not real optimal solutions, or are local optimal solutions. It will be close to the requirements of decision makers to have an interactive decision analysis process, in which decision makers may have a review on the alternatives, they may have a comparison on those alternatives to find problems, and to get insight on urban land use problem, they may present some new alternatives based upon basic set of alternatives. At last, they select one or some satisfied alternatives.









Environment Impacts Assessment


There are five rivers in Guilin city, they are Lijiang, Xiaodongjiang, Taohuajiang, Ningyuanhe, Nanxihe. They form a complex river network in the city, as shown in the following figure. There are five waste water treatment plants in Guilin city. Some discharged waste water are sent to these plants, and some directly flow into rivers for the limited capacity of treatment and municipal facilities. We suppose several discharge points for the discharged water directly flowing into rivers. The monitored points (or sections of river flow), discharge points are shown in the figure also.


The concentration of COD, SS, Oil are taken as controlled indexes for water quality in Lijiang river. One dimension water quality model is used to evaluate the concentration of COD, SS and Oil in rivers.

Mullet-Box air quality assessment model is applied for atmosphere environment evaluation in Guilin city. For the purpose, the city is divided into six boxes shown in the following figure. Both industry waste gas discharge and residential discharge are in consideration. The discharge amount is evaluated according to the production of each industry sector and number of residents in future years.









Traffic Evaluation


In traffic evaluation, relative turnover value of traffic and the entropy of traffic distribution are calculated. Turnover value reflects transportation amount of a "cluster land use alternative"; entropy reflects the balance degree of traffic distribution. A larger value of entropy means traffic distribute equally among clusters and there will be few heavy traffic problems.

Freight traffic among economical sectors, passenger traffic for employment, shopping, and sightseeing are main traffic within city. Because there are no statistical data for traffic within city, it is a difficlut duty to have an evaluation for it. Based upon some assumption, a method using information provided by input-output table to calculate the ratio of freight traffic amount from one industry sector to another to overall freight traffic amount within a city, called relative freight traffic, is presented. The method can be applied to evaluate relative freight traffic between any pair of clusters for a "cluster land use alternative".

Passenger traffic for employment is evaluated based on the assumption of that employee will live in the cluster close to his (her) working place as close as possible. Passenger traffic for shopping is evaluated according to residents distribution on clusters, employment distribution, and the attraction size of each commerce center which depends on its location and scale. The Passenger traffic for sightseeing is distributed among scenic spots, center of the city and public traffic center.

The relative turnover value can be calculated from traffic amount and average distance between clusters. The entropy are calculated from traffic between clusters directly.









Location Optimization


Because environment and traffic are two major urban problems and they are heavily influenced by industry and residents distribution on urban areas, location optimization is based upon environment impacts assessment and traffic evaluation for "cluster land use alternative". Location optimization model is constructed as the following,

Objects:

    1. maximum of environmental quality
      1.1 minimum of water pollution
        1.1.1 COD concentration on monitor points
        1.1.2 area source pollution of COD in clusters
        1.1.3 SS concentration on monitor points
        1.1.4 area source pollution of SS in clusters
        1.1.5 OIL concentration on monitor points
        1.1.6 area source pollution of OIL in clusters
        1.1.7 area source pollution of waste water in clusters
      1.2 minimum of air pollution
        1.2.1 SO2 concentration on monitor points
        1.2.2 TSP concentration on monitor points
    2. maximum of good traffic condition
      2.1 minimum of total turnover value of traffic
        2.1.1 freight traffic
        2.1.2 passenger traffic for employment
        2.1.3 passenger traffic for shopping
        2.1.4 passenger traffic for sightseeing
      2.2 maximum of entropy value of traffic
        2.2.1 freight traffic
        2.2.2 passenger traffic for employment
        2.2.3 passenger traffic for shopping
        2.2.4 passenger traffic for sightseeing

The objects in the model are quite complex. AHP is used to deal with the decision maker's preference on these objects.


Constraints

    1. Environmental standards in environment planning for monitor points, or some important points in city to be protected in emphasis, like scenic spots, historical cultural remains, should be satisfied.

    2. The sum of each industry sector production within the same cluster in 2002 should be less than the ultimate overall industry production defined in urban planning, and the number of residents in a cluster should be less than ultimate scale defined, which reflects the land capacity for human being's activities.

    3. Requirements of clusters for industry sectors, like, any manufacturing and process industry should be out of center cluster in year 2000, should be satisfied.

    4. The sum of production for same sector in every cluster should be equal to the sector production in industry development plan or socio-economic development plan, the number of resident is similar.

    5. The consistence requirements of land use alternative, like, the production of a sector in a cluster should vary in same direction, increase or decrease, should be satisfied.


In the optimization model, both objects and constraints contain complex nonlinear function, it's difficult to solve it with traditional optimization algorithms. We are trying to use Genetic Algorithms to solve the problem. It is still in experiments.









Interactive Discrete Multicriteria Decision Making


As stated in problem description, the set of optimal alternatives obtained with optimization techniques usually are not real optimal alternatives. One way to get better solution based upon the basic set of alternatives is to use discrete multicriteria decision making method. The following is the sketch of the decision analysis procedure for Guilin case.

Step1. adopt interactive industry structure optimization procedure to produce a set of alternatives,

      Alts1={A[1],A[2],...,A[n]}
      A[i] = (X[i],Y[i],Z[i],W1[i],R[i],E[i])

X[i]represents the ith solution of the programming model; A(i) represents the ith alternative X(i) and its attributes which consist with (1) the effects of the alternative X(i) represented by Y(i), the related values of system state variables, and Z(i), goal variables; (2) the assumption for the system and its environment in future, they include, the preference on multiobjects represented by W1(i), system structure and parameters represented by R(i), systems environment parameters represented by E(i). go to step 2;

Step 2. for each alternative of industry structure optimization, adopt location optimization model to generate a set of alternatives

      Alts2={B[i,j] | i=1,...n, j=1,...,m}
      B[i,j] = (XX[i,j],P[i,j],YY[i,j],ZZ[i,j],W2[i,j],RR[i,j],EE[i,j])
B[i,j] is the jth optimal location alternative and its attributes, correlated with A[i].

      XX[i,j] = {xx(c,s,t)}
      P[i,j] = popu(c,t)
xx(c,s,t) is the production of industry sector s in cluster c in year t, and popu(c,t) is the population live in cluster c in year t.
The meaning of other variables in the definition of B[i,j] are similar with ones in A[i].go to step 3;

Step 3. presents some new indexes which have not been considered in optimization models. They are represented by I[i,j]. For each alternative XX[i,j], evaluate its extended index I[i,j] to form a alternatives-attributes table ALTS(0). Let k=0,

      ALTS(k) = { D[i,j] | i=1,...,n, j=1,...,m }
      D[i,j] = ( Alternative[i,j], Attributes[i,j] )
      Alternative[i,j] =( XX[i,j], P[i,j] )
      Attributes[i,j] = (Effects[i,j], Preference[i,j], Assumption[i,j])
      Effects[i,j] =(YY[i,j],Y[i],ZZ[i,j],Z[i], I[i,j] )
      Preference[i,j] = (W2[i,j],W1[i])
      Assumptions[i,j] = (X[i],RR[i,j],R[i],EE[i,j],E[i])
go to step4;

Step 4. for each alternative in ALTS(k), review and further analyze the alternative and its attributes by decision maker; go to step 5;

Step 5. make comparison analysis for all alternatives, and rank the alternatives, according to interested indexes; go to step 6;

Step 6. if decision maker generate a new alternative N, calculate it attributes with related models, let ALTS(k+1) = ALTS(k)+ {N}, k=k+1, go to step 4; otherwise, go to step7;

Step7. select all nondominated alternatives to form NDALTS, and let ALTS(k+1)=NDALTS, k=k+1; go to step 8;

Step 8. present a reference point, and use the point to find out an optimal alternative, go to step 9;

Step 9. review the optimal alternative, and compare it with others in ALTS(k); If it is a satisfied solution, output the alternative, and stop; otherwise go to step 4.

In the procedure, the system interface to decision maker is very important, it must be well designed.









Supporting Information

[1] Economics and Social Statistical Yearbook of Guilin, 1996
[2] Economics and Social Statistical Yearbook of Guilin, 1995
[3] Economics and Social Statistical Yearbook of Guangxi, 1995
[4] Economics and Social Statistical Yearbook of China, 1996
[5] Statistical Yearbook of Chinese Cities, 1995
[6] Report For Compendium of Guilin Urban Planning, 1995









RETURN






Copyright 1995-2002 by:   ESS   Environmental Software and Services GmbH AUSTRIA