Spatial Decision Support System

in site sui
A spatial decision support system (SDSS) is a computer-based system that combines
conventional data, spatially referenced data and information, and decision logic as a tool for
assisting a human decision-maker. It usually includes a user interface for communicating
with the decision-maker. A SDSS does not actually make a decision, but instead assists and
analysing data and presenting processed information in a form that is friendly to the decision-
maker.
Densham (1991) stated that SDSS as – “explicitly designed to provide the user with a
decision-making environment that enables the analysis of geographical information to be
carried out in a flexible manner.”
It is designed to assist the spatial planner with guidance in making land use decisions. A
system which models’ decisions could be used to help identify the most effective decision
path.
Based on the definition there are some characteristics of SDSS:
1. In SDSS, GIS based models are used.
2. A spatial decision support system (SDSS) is an interactive, computer-based system
designed to assist in decision making while solving a semi-structured spatial problem.
3. If there are a lots of options then spatial decision problem help to choose best
alternatives.
4. Based on set of criteria, alternatives are assessed.
5. Some of criteria, quantitative or qualitative.
6. Conflicting group of decision makers.
7. Surrounded by eccentricity.
Multi Criteria Decision Analysis (MCDA):
When we are considering one or more than two criteria or a set of criteria then the decision
analysis is called Multi criteria decision analysis.
Components:
In general, MCDA problems involve six components:
1. Goal or a set of goal that decision maker tries to achieve.
2. Decision makers is involved in this process, they have their conflicting interest.
3. A set of evaluation criteria have objectives and attribute- objectives are the path
through which we accomplish the goal and in order to fulfil the objectives we have to
define attributes.
4. Set of decision alternatives, that is action variable.
5. The set of uncontrollable variables are states in nature.
6. Output will depend between the interaction and criteria.

Framework:
Decision making process can be fulfilled by two path- (I) Alternative focus approach
(II) Value focus approach
In Alternative focus approach we give preferences on the alternatives and in value focus
approach we give preferences on the attributes. First set goal then derived alternatives, then
we define the criteria. After setting the goal, we define the set of criteria then set the
alternatives.
In value focus approach the framework of MCDA:

Problem definition
Problem
Definition

Evaluation Constraints
criteria

Decision Alternatives
Matrix

Decision
makers
preferences

Decision
rules

Sensitivity
analysis

Recommendation

Framework for spatial multicriteria decision

analysis.
Analytical Hierarchy Process (AHP):
The AHP method, suggested by Saaty (1980), has become a popular tool for multi-criteria
decision-making. It supports decision-makers to make the best decision, by reducing complex
decisions to a series of comparative pairs and synthesizing the [Link] disaggregates a
complex decision problem into different hierarchical levels. This method allows quantifying
opinions and transforming them into a coherent decision model.
Steps of AHP method:
i. Define the goal and alternatives.
ii. Define the problem and decision criteria.
iii. Assess the relative value or priority of each decision criterion.
iv. Calculate the weights of the criteria and priorities.
v. Analyse consistency.
Importance:
The AHP is a powerful tool that can be used at home, at school, or at work. It removes bias
from the decision-making process and ensures the decision you make reflects your values and
priorities. The next time you are faced with a difficult decision, consider using the AHP to
make your choice.
Applications:
i. The AHP is a decision-making tool to deal with multi-criteria evaluation and can
be used with integrated GIS spatial analysis for site selection of fire stations.
ii. The application of analytical hierarchy process (AHP) established the sustainable
procurement index with a consistent sensitivity analysis result.

Study area:
Bundelkhand region, spanning between 23˚20′ N to 26˚20′N latitude and 78˚20E to 81˚40’E
longitude, lies at theheart of India and covers up an area of 7 million hectares.
This region comprises thirteen districts: seven in UP andsix in MP. It falls under the semi-
arid and sub-humid climate. The average annual rainfall received by thisregion is only 950
mm, which is insufficient for crop cultivation.

INDIA

Fig: Location of the study area

 Socio economic characteristics:
Vulnerability is term used to describe exposure to hazards and [Link] Bundelkhand
region comprising thirteen districts ofUP and MP has unique socio-economic
[Link] 205 highly backward districts of the country, thisregion consists of 10
districts and is considered as one ofthe most ‘backward regions’ of the country by the Govt.
ofIndia. The socio-economic vulnerability occurs due to poverty, unemployment, educational
attainment, linguistic isolation, and percent of income spent on housing and many more
[Link] and social factors like income, education, employment, community safety,
and social supports can have a big impact on our health and longevity. Our capacity to make
healthy choices, pay for housing and medical care, manage stress, and more is impacted by
these factors.
Criterion selection:
Spatial multi-criteria decision analysis enablesthe user to identify the most suitable
alternatives withrespect to a set of evaluation criteria. In order to assess thesocio-economic
vulnerability risk and also keep in mind the availability of data, we considered the five
criteria, viz. cultivators, agricultural labours,population density, SC/ST population, and
percentage ofliteracy in the Bundelkhand region.

Fig.1 Population density Fig.2 SC/ST Population

Fig.3 Agricultural labours Fig.4 Cultivators

 Fig. 5 Literacy rate

Standardization of criteria map:

Standardization is to enforce a level of consistency or uniformity to certain practices or
operations within the selected environment.
Linear scale transformation is one of the methods used to perform the rescaling of discrete
measurement scales to make the results that have been obtained using verbal scales with
different numbers of response options comparable.
Linear scale transformation is two types,
1. Maximum score
2. Score range
Both have two criteria,
a. Benefit criteria: Benefit criteria those criteria shall be termed as benefit criteria where
higher score indicates more importance or higher performance.
Fig.4 Cultivators
b. Cost criteria : The interrelationship shall be reverse, here higher score indicates poor
performance or less importance.
We mainly use score range method for standardization-
 For cost criteria: Xij=Xjmax-Xij/ Xjmax - Xjmin
 For benefit criteria: Xij = Xij – Xjmin / Xjmax- Xjmin

Population density: There is a distinct level of disparity in this region's population

distribution. While MP Bundelkhand has a much larger area (58%) than UP Bundelkhand
(42%) and a smaller percentage of the total population (47%) than the UP side (53%). Fig.3
If theAgricultural labours
population density increases, the socio-economic vulnerability will increase. So, population
density is benefit [Link], the socio-economic vulnerability will increase. So,
population density is benefit criteria.
 Fig.6 Standardized map of Population density

SC/ST Population:
If the SC/ST Population is increase, the socio-economic vulnerability will increase. So,
SC/ST Population is a benefit criteria.

Fig.7Standardized map of SC/ST Population

Agricultural labours:
Since agricultural labours are directly involved in theagricultural sector for their livelihood,
they are highlyvulnerable socio [Link] percentage ofagricultural labours (28.4%)
in Bundelkhand is higher thanUP (21.8%).If the agricultural labour’s is increase, the socio-
economic vulnerability will increase. So, agricultural labour is a benefit criteria.

Fig.8Standardized map of agricultural labour

Cultivators:
A large number of people in this agriculture-based region areinvolved in the agricultural
[Link] a whole, the districts ofUP Bundelkhand shows comparatively higher density
ofcultivators than the MP part of Bundelkhand. Thehighest density of cultivator was found in
the Tikamgarhdistrict (61 persons/[Link].).If the cultivatorsare increase, the socio-economic
vulnerability will increase. So, cultivators are a benefit criteria.

Fig.9Standardized map of cultivators

Literacy rate:
The literacy rate has been regarded as an important parameter for measuring the
socioeconomic vulnerability of the Bundelkhand region due to its role in enhancing
communities' ability to adapt to natural [Link] the literacy rateis increase, the socio-
economic vulnerability will decrease. So, literacy rate is a costcriteria.

Fig.10Standardized map of Literacy rate

Criterion Weighting:
We consider decision maker’s preferences or knowledge to estimate criterion weights. In
design phase we assign criterion weights. Weights are associated with each criterion.
Weights are values that assign to the criterion which indicates the relative importance of each
criterion.
Weights are calculating in different methods. These are below-
1. Ranking method: It is simplest method, here we depend on their different value.
There are three type of raking method-
a. Rank sum
Rank sum(wi) = n-rj+1/Σ(n-rk+1) where, n = total number of criteria
rj = rank position of “j” criteria
rk = all the criteria under consideration
b. Rank reciprocal
Wj = (1/rj)/Σ(1/rk)

c. Rank exponent
 Wj = (n-rj+1)p/Σ(n-rk+1)p Where, p = probability value

RANK SUM
CRITARIA STRAIGTH
WEIGHT NORMALIZED
CULTIVATOR 1.00 5.00 0.33
AGRICULTURE 2.00 4.00 0.27
POPULATION 3.00 3.00 0.20
LITERACY 4.00 2.00 0.13
TOTAL SC/ST 5.00 1.00 0.07
SUM 15.00 1.00

RANK RESIPROCAL RANK EXPONENT

WEIGHT NORMALIZED WEIGHT NORMALIZED
1.00 0.44 25.00 0.45
0.50 0.22 16.00 0.29
0.33 0.15 9.00 0.16
0.25 0.11 4.00 0.07
0.20 0.09 1.00 0.02
2.28 1.00 55.00 1.00

 Ranking method is simple and easy to use, but it doesn’t consider relative
importance between two criteria.
 However, this method has been criticized for the solid theoretical
foundation.

2. Rating method:
Rating method has two procedures,
a. Ratio estimation procedure
b. Point allocation approach

Assessing Weights by the Ratio Estimation method

Name of the criteria Straight Rank Ratio scale Original weight Normalized weight
Agricultural labours 2 85 2.8333 0.2530
Literacy 5 30 1.0000 0.0893
Population density 1 100 3.3333 0.2976
Total sc/st 3 67 2.2333 0.1994
Cultivators 4 54 1.8000 0.1607
11.2000 1.0000
 Point allocation approach
Name of the Criteria Straightrank Weight
Agricultural labours 25 0.25
Literacy 8 0.08
Population density 45 0.45
 There is no equal interval on SC/ST
Total the ratio estimation 12
procedure. 0.12
3. Pairwise comparison Cultivators
method: 10 0.1
We use AHP method, that’s why we use pairwise comparison
1 method for socio
economic vulnerability mapping in Bundelkhand region.

 First, we have to give criterion weight underlying scale with values from 1 to 9 rate.
 Sum the values in each column.

STEP-1
CRITERIA CULTIVATOR AGRICULTURE POPULATION LITERACY TOTAL SC/ST
CULTIVATOR 1.00 2.00 3.00 4.00 5.00
AGRICULTURE 0.50 1.00 2.00 3.00 4.00
POPULATION 0.33 0.50 1.00 2.00 3.00
LITERACY 0.25 0.33 0.50 1.00 3.00
TOTAL SC/ST 0.20 0.25 0.33 0.33 1.00
SUM 2.28 4.08 6.83 10.33 16.00

STEP-2
CULTIVATO LITERAC TOTAL
R
AGRICULTURE POPULATION
Y SC/ST sum WEIGHT
0.44 0.49 0.44 0.39 0.31 2.07 41
0.22 0.24 0.29 0.29 0.25 1.30 26
0.15 0.12 0.15 0.19 0.19 0.80 16
0.11 0.08 0.07 0.10 0.19 0.55 11
0.09 0.06 0.05 0.03 0.06 0.29 6
1.00 1.00 1.00 1.00 1.00 - 100.00

STEP-3
Consistency vector CONSISTENCY CONSISTENCY
LAMDA
estimation INDEX RATIO
5.179919388
5.194640635
5.161004223
5.12899107 0.032247768 0.02879265
5.05395083
5.055440285
25.64495536
Computation of the criterion weights
 This step involves the following operations,
✓ Sum the values in each column of the pairwise comparison matrix;
✓ Divide each element in the matrix by its column total ( the resulting matrix is

✓ Compute the average of the element in each row of the normalized matrix; that
referred to as the normalized pairwise comparison matrix ); and

is, divide the sum of normalized score for each row by the number of criteria.

The consistency ratio estimation

This operation involves the following steps :

1. Consistency Ratio Estimation : First criterion weight value multiply by first column, then
second criterion value multiply by second column,…………

2. Lamda calculation : Total Consistency Vector / Number of Criterion

3. Consistency Index (CI) Estimation : (Lamda – no. of criteria) / (no. of criteria – 1)

4. Consistency Ratio Estimation : Consistency Index / 1.49 (Random Index value for 10
criteria
ratio value indicatesthat the pairwise comparisons are reasonably consistent.

Weighted overlay analysis:

The socio-economic vulnerability was computed integrating the criteria maps in the AHP-
basedmodel. The pairwise comparison method was employed todetermine the criterion
weights for each criterion considered in this study. Based on the range of values representing
the socio-economic vulnerability, five vulnerabilityzones were demarcated.
The criterion weights for each criterion in this region were determined using the pairwise
comparison method. In this Bundelkhand region, five vulnerability zones were delineated
based on the range of values that represent in the socio-economic vulnerability. Banda is one
of the districts in the very high vulnerability zonwwwe.
Socio-economic Vulnerability Map of Bundelkhand Region