See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/341568751

GIS-based multi criteria evaluation techniques in site suitability analysis: Case

of potential future development site in Gasabo District of Kigali city.

Chapter · June 2014

CITATIONS READS

0 159

1 author:

Maxime Rushemuka
United Nations University (UNU)
3 PUBLICATIONS 0 CITATIONS

SEE PROFILE

All content following this page was uploaded by Maxime Rushemuka on 22 May 2020.

The user has requested enhancement of the downloaded file.

 i

UNIVERSITY OF RWANDA (UR)

HUYE CAMPUS
COLLEGE OF SCIENCES AND TECHNOLOGY
SCHOOL OF PURE AND APPLIED SCIENCE
DEPARTMENT OF GEOGRAPHY

GIS-BASED MULTI CRITERIA EVALUATION TECHNIQUES IN SITE SUITABILITY

ANALYSIS: Case of Potential Future Development Site in Gasabo District of Kigali city.

Dissertation submitted in partial fulfillment of the Academic requirements for the award of
Bachelor’s Degree with honors in Geography, Option of Environmental Management.

Presented by RUSHEMUKA N. Maxime

Supervisor: Mr. Theodomir Mugiraneza Huye, June, 2014

 ii

Declaration

This is to declare that the dissertation entitled ‘‘GIS-based multicriteria evaluation technique in
site suitability analysis: case study of housing site suitability analysis in Gasabo district of Kigali
city’’ Submitted by Mr. Rushemuka N. Maxime towards partial fulfillment of Bachelor’s Degree
of Science in Geography, Option of Environmental Management under the supervision of Mr.
Theodomir Mugiraneza is based on the original study; information in this document has been
obtained and presented in accordance with academic rules and ethical conduct. I also declare
that, as required by these rules and conducts, I have fully cited and referenced all materials and
results that are not original to this work.

Rushemuka N. Maxime,

Signature…………………………………………..Date……………………………….

Supervisor: Theodomir Mugiraneza,

Signature…………………………………………..Date………………………………..
 iii

Dedication

To

My entire family

I dedicate this work

 iv

Abstract

Site suitability analysis for housing purpose is a complex process that often involves multiple,
conflicting and different nature objectives. This study entitled ‘‘GIS-based Multicriteria
Evaluation Technique for Site Suitability Analysis: case study of housing site suitability analysis
in Gasabo District of Kigali city’’ have the intention of demonstrating the efficiency of Gis-
based Multicriteria Evaluation techniques in the urban planning domain in general and in site
suitability analysis. Using the existing data from different sources and based on environments
rules and other housing regulations in Rwanda, objectives that guide the housing suitability
assessment have been identified and classified in two groups, restrictions and suitability factors.
AHP has been choose as the MCE technique to be used in calculating the weight of each
suitability factor according to its importance; the resulting weighted factors map have been
combined with the all restrictions map using map algebra tool in ArcGis10.1 environment to
produce the final suitability map indicating the degree of suitability (extremely suitable, high
suitable, suitable, moderate suitable, less suitable and restricted) across the study area. Through
the present study GIS-based multi criteria evaluation is revealed to be an effective tool in site
suitability analysis in general and in housing site suitability analysis in particular that we can
recommend Kigali city and Gasabo district; GIS-based MCE techniques can be used as a tool of
decision making and planning in Rwanda because it is an easy to understand and use approach. It
can also be used in hazard management, landslide and other suitability analysis needed by Kigali
city. Some factors used in housing suitability assessment were not incorporated in this study due
to time, means and lack of appropriate data constraints; therefore further researches in the this
domain in Kigali city are still needed.

Key words: GIS, Multi Criteria Evaluation, site suitability analysis, housing site
 v

Acknowledgments

This research has been possible thanks to the contribution of different people that I want to
express my sincere gratitude. First, I would like to thank my supervisor Mr. Theodomir
Mugiraneza for assisting me during the whole period of the present research; your remarks were
constructive and appreciated.

I also want to thank the entire geography department staff at former N.U.R for their guidance
and hints during my research.

Last but not least I want to deeply thank my classmates for their encouragement and making a
good environment for research, among them I want to thank especially Mr. Tuyisingize J. Felix
and Mr. Andre Ruramira who highly motivated me during my research period.
 vi

Table of Contents
Declaration .................................................................................................................................................... ii
Dedication .................................................................................................................................................... iii
Abstract ........................................................................................................................................................ iv
Acknowledgments......................................................................................................................................... v
List of figures ............................................................................................................................................. viii
List of Tables ............................................................................................................................................... ix
List of acronyms ........................................................................................................................................... x
List of Appendices ....................................................................................................................................... xi
CHAPTER 1. INTRODUCTION ................................................................................................................ 1
1.1. Study background .............................................................................................................................. 1
1.2. Problem statement .............................................................................................................................. 3
1.3. Research objectives ............................................................................................................................ 5
1.4. Research hypothesis and research questions ...................................................................................... 5
CHAPTER 2. LITERATURE REVIEW ...................................................................................................... 6
2.1. Housing development issues in Kigali city ........................................................................................ 6
2.2. Geographical information systems (gis) ............................................................................................ 6
2.3. GIS and urban planning ..................................................................................................................... 7
2.4. Site search procedures using standard gis map overlays ................................................................... 8
2.5. Multi criteria decision making (MCDM) ........................................................................................... 9
2.6. Spatial multi criteria decision making ............................................................................................. 10
2.7. GIS and multi criteria decision analysis .......................................................................................... 11
2.8. Analytical hierarchy process ............................................................................................................ 12
2.8.1. Pair-wise comparisons method ................................................................................................. 12
2.8.2. Identifying Criteria .................................................................................................................... 13
CHAPTER 3. MATERIAL AND METHODS ........................................................................................... 15
3.1. Study area description ...................................................................................................................... 15
3.2. Data and materials............................................................................................................................ 16
3.3. Methods............................................................................................................................................ 17
3.3.1. Restriction map (constraints) .................................................................................................... 18
3.3.2. Suitability Factors map ............................................................................................................. 19
3.3.3. Allocation of Weight to factors ................................................................................................. 19
 vii

CHAPTER 4. RESULTS AND DISCUSSION .......................................................................................... 24

4.1. Permissible area for potential future development against restricted areas ..................................... 24
4.2. Suitability factors maps.................................................................................................................... 31
CHAPTER 5. GENERAL CONCLUSION AND RECOMMENDATIONS ............................................. 37
5.1. General conclusion........................................................................................................................... 37
5.2. Recommendations ............................................................................................................................ 38
6. REFERENCES ....................................................................................................................................... 39
7. APPENDICES ........................................................................................................................................ 41
 viii

List of figures

Fig1. GIS and urban planning…………………………………………………8

Fig2. Decision flowchart for spatial multicriteria analysis……………………11

Fig3. The location of Gasabo district………………………………………….15

Fig4. Used housing site suitability analysis procedure………………………..17

Fig5. Lake constrained area…………………………………………………....24

Fig6. Airport constrained area…………………………………………………25

Fig7. Defense constrained area………………………………………………...26

Fig8. River constrained area…………………………………………………...27

Fig9. Industry constrained area…………………………………………………28

Fig10. Wetland constrained area………………………………………………..29

Fig11. Final Restriction map……………………………………………………30

Fig12. Elevation factor map…………………………………………………….31

Fig13. Slope factor map…………………………………………………………32

Fig14. Road proximity factor map………………………………………………33

Fig15. Land use/ cover suitability classes………………………………………..34

Fig16. Housing suitability map………………………………………………….35

 ix

List of Tables

Table1. Research hypothesis and research questions……………………………5

Table2. Nine point weighting scale for pairwise comparison…………………..20

Table3. Pairwise comparison matrix……………………………………………22

Table 4. Calculation of the consistency vector………………………………….22

Table5. Random index ………………………………………………………….23

Table6. Suitability classes statistics…………………………………………….36

 x

List of acronyms

AHP: Analytical Hierarchy Process

CGIS-NUR: Centre for Geographic Information and Remote Sensing –National University of
Rwanda

DEM: Digital Elevation Model

GIS: Geographic Information System

LUDMP: Land use and Development Master Plan

MCA: MultiCriteria Analysis

MCDA: MultiCriteria Decision Analysis

MCDM: Multicriteria Decision Making

MCE: MultiCriteria Evaluation

MCET : MultiCriteria Evaluation Technique

REMA : Rwanda Environnemental Management Authority

RNLUMP: Rwanda National Land Use Master Plan

RNRA: Rwanda Natural Resources Authority

 xi

List of Appendices

Appendix1. Restrictions Model……………………………………………41

Appendix2. Factors Suitability Model……………………………………..42

 1

CHAPTER 1. INTRODUCTION

1.1. Study background

For the first time ever, the majority of the world's population lives in a city, and this proportion
continues to grow. By 1990, less than 40% of the global population lived in a city, but as of
2010, more than half of all people live in an urban area; by 2030, 6 out of every 10 people will
live in a city and by 2050, this proportion will increase to 7 out of 10 people (UNFPA, 2007).
Almost all urban population growth in the next 30 years will occur in cities of developing
countries include in Africa; between 1995 and 2005, the urban population of developing
countries grew by an average of 1.2 million people per week, or around 165 000 people every
day. By the middle of the 21st century, it is estimated that the urban population of these counties
will more than double, increasing from 2.5 billion in 2009 to almost 5.2 billion in 2050 (WHO,
2013).

The territorial expansion of cities will necessarily affect environmental outcomes. Since many
cities are situated at the heart of rich agricultural areas or other lands rich in biodiversity, the
extension of the urban perimeter evidently cuts further into available productive land and
encroaches upon important ecosystems (UNFPA, 2007).Unplanned urbanization may lead to
main problems like inadequate solid waste disposal services, lack of adequate public water
supply, traffic congestion, water logging, air pollution, noise pollution, hill cutting and even
social conflicts (Alam, et al., 2006).

Spatial decision problems typically involve a large set of feasible alternatives and multiple,
conflicting and incommensurate evaluation criteria. The alternatives are often evaluated by a
number of individuals (decision-makers, managers, stakeholders, interest groups). The
individuals are typically characterized by unique preferences with respect to the relative
importance of criteria on the basis of which the alternatives are evaluated. Accordingly, many
spatial decision problems give rise to the GIS-based multicriteria decision analysis (GIS-
MCDA).
 2

The Multi Criteria Decision Analysis (MCDA) approach is one of the most widely used GIS-
based decision rules; the method is often applied in land use/suitability analysis, site selection,
and resource evaluation problems (Malczewski 1999). The primary reason for its popularity is
that the method is very easy to implement within the GIS environment using map algebra
operations and cartographic modeling, the method is also easy-to-understand and intuitively
appealing to decision makers (Berry 1993)

The availability and use of geographic digital data and decision-making tools have increased the
development of geographic analyses that can assist in decision making and land-use planning.
Multi-criteria evaluation (MCE) analysis (the implementation of decision-making rules to
identify and enable the combination of many criteria, in the form of GIS layers, into a single
map) and Geographic Information Systems (GIS) are two examples of tools that aid in the
development of geographic data and maps for different purposes, such as selecting land for
housing uses.

The Analytic Hierarchy Process (AHP) used in Multi Criteria Evaluation (MCE), introduced by
Saaty (1980) is an analytical tool that enables planners to explicitly rank tangible and intangible
criteria against each other for the purpose of selecting priorities. Localization problems have also
been treated with a more intensive use of GIS. Steve (1991) uses GIS to evaluate various
alternatives for nuclear waste sites, Kumar and Biswas (2012), Shattri et al (2006), all of them
used the AHP in GIS-based multicriteria Evaluation Techniques to find new potential housing
sites.

It is in this context that, this research intends to show how progress in computing sciences,
including Geographical Information Systems (GIS) associated with problem solving approaches
can help planners handle the multi-objectivity nature and complexity of the housing site
suitability analysis process.
 3

1.2. Problem statement

The 2012 population census record show that the number of population in Kigali city increased
rapidly from 765,325 in 2002 to 1,135,428 in 2012 with annual growth 4%, which increased
1.4% more than the annual population growth for Rwanda (NISR, 2012). Today, a large part
of the urban areas in Kigali are unplanned, under serviced and growing haphazardly(Kigali
city, 2013). As the city grows, more land will be needed to satisfy further growth of urban
population in the future. In this context it is very important to find housing suitable areas for
urban development to avoid further undesirable urban growth, but also to relocate the population
living in unsuitable areas in Kigali city.

So far, there have been some efforts in urban planning by the country and by Kigali city in
particular that helped Kigali guide its urban development. Kigali city managed to put in place in
2010 a Land Use and Development Master Plan (LUDMP), the City as well as the whole country
territory has actually completed parcel land registration programme, in 2007, the first long
term conceptual Master Plan was prepared for Kigali to guide long term planning in the
City. All districts of Kigali city have their detailed physical plans that guide land utilization
strategy.

Those existing planning materials are very important in guiding the growth of Kigali city; But,
they are not enough detailed, as they only provide an overview and strategy of the land
utilization. They combine also many themes that make it difficult to give details on a particular
land use like housing sites suitability as it is the case for the present research. Another problem is
that those land use plans have been established based on environmental regulations and policies
in Rwanda, when any regulation changes it would be necessary to readjust those existing plans to
meet the recent regulations and policies; this would be a big task, as those plans are designed by
consultants and it will be necessary to call on them. Kigali city needs a new simple to use, update
and understand tool that can be used in site suitability analysis. A housing suitability assessment
in Gasabo district can be needed either by government institutions as a basis for the decision
making or private property developers who need to know the most suitable sites for developing
new houses properties.
 4

Selecting the location for housing sites is a complex process involving not only technical
requirement, but also physical, economical, social, environmental and political requirements that
may result in conflicting objectives; such complexities necessitate a Multi Criteria Analysis
(MCA). The MCA in housing site suitability analysis process involves the simultaneous use of
several decision support tools such as Geographical Information System (GIS) and Multi Criteria
Evaluation Techniques (MCET) using analytical hierarchy process (AHP).

The Analytic Hierarchy Process (AHP) introduced by Saaty (1980) is an analytical tool that
enables planners to explicitly rank tangible and intangible criteria against each other for the
purpose of selecting priorities. The AHP as a multi-objective, multicriteria decision-making
approach uses a pair-wise criteria comparison to arrive at a scale of preferences among sets of
alternatives, and has been widely used in the housing site suitability analysis. Integration of GIS
and the AHP can be a powerful tool to solve the housing site suitability analysis problem. A
review of literature reveals numerous successful applications of GIS and AHP in the housing site
suitability analysis process (Kumar and Biswas 2012, Shattri et al 2006, Steve 1991).

The intention of this research is to demonstrate how Progress in computing sciences, including
Geographical Information Systems (GIS) combined with the AHP in Multi Criteria Decision
Analysis (MCDA) approach can help planners handle the complexity of housing site suitability
analysis process.
 5

1.3. Research objectives

The overall objective of this research is to present a GIS-based approach for land suitability
assessment. It is presented by means of a case study: the selection of suitable housing site in
Gasabo district of Kigali city.

To meet the overall objective of this study some specific objectives were set:

1. Reviewing planning concepts and existing planning guidelines in housing development

sector globally and by Kigali city in particular
2. To develop a Multicriteria Evaluation Technique (MCET) used in this study in the
determination of suitable land for housing
3. To produce a final suitability map for potential sites for housing purpose in Gasabo
district.
4. To formulate some recommendations based on the findings of the study.

1.4. Research hypothesis and research questions

Table1. Research hypothesis and research questions
1. Kigali city faces challenges in What are the structure and guidelines used in
organizing a planning framework for housing site suitability analysis by Kigali city,
multicriteria problems in urban Gasabo district?
development

2. Gis based Multicriteria evaluation What are different Multicriteria evaluation

techniques are an efficient tool in site techniques?
suitability analysis How those MCE techniques are applied in
housing development sector?

3. Land suitability differ from an area to Is the land use suitability similar across the
another study area, What is the housing suitability level
across the study area (Gasabo District)?
 6

CHAPTER 2. LITERATURE REVIEW

2.1. Housing development issues in Kigali city

Kigali faces an acute housing shortage in the formal housing market and there is a huge existing
demand as well as the growing needs that require to be addressed in term of housing provision.
(Kigali City, 2013)

35% of the entire city land area is occupied by steep slopes of more than 20% on which
urban development is restricted by Rwanda National Land use Development Master Plan
(RNLUMP), 21% of which are not yet build upon and 14% located within urban areas,
mostly encroached by unplanned settlements. As per (RNLUMP) Kigali City land area falls
under medium to high risk soil erosion zone and soil present on more than 5% slope is
susceptible to heavy erosion. Land degradation is linked to widespread land clearance and over
cultivation. Heavy soil erosion reduces soil fertility and heavy silt load depletes river and
wetland habitats. In many parts of the city the developments are happening along the steep slopes
of Gatsata, Gisozi, Remera, Kicukiro etc. The city lacks sustainable slope management and
erosion control policies (Kigali city, 2013).

2.2. Geographical information systems (gis)

A geographic information system (GIS) can be defined as an organized collection of software

and geographic data that allow efficient storage, analysis, and presentation of spatially explicit
and geographically referenced information (Clarke, 1997).

Traditional methods of processing such data have been extremely labor intensive, such as
manually digitizing a map from an aerial photograph and then adding information (attribute)
regarding the studied phenomena. A GIS provides a powerful analytical tool that can be used to
create and link spatial and descriptive data for problem solving, spatial modeling and
presentation of results in tables or maps (Armstrong and Densham, 1990)

Geographic information systems (GIS) have emerged as useful computer-based tools for spatial
description and manipulation. Although often described as a decision support system, there have
been some disputes regarding whether the GIS decision support capabilities are sufficient.
 7

Since current GIS do not provide decision-making modules that reason a decision and are
primarily based on manual techniques and human judgments for problem solving, the individual
should have the decision rules in place before GIS can be utilized (Shattri et al, 2006).

2.3. GIS and urban planning

Planners require solutions that address day-to-day work needs while also fostering the ability to
effectively predict and respond to chronic urban problems and future market fluctuation. The
success of planners in combating chronic urban problems is largely determined by their ability to
utilize effective tools and planning support systems that allow them to make informed decisions
based on actionable intelligence (Esri, 2011). Today, planners utilize GIS around the world in a
variety of applications. Actually GIS is being used as a platform to help planners reach their
goals of creating livable communities and improving the overall quality of life while protecting
the environment and promoting economic development. GIS tools can provide the necessary
planning platform for visualization, modeling, analysis, and collaboration (Shattri et al, 2006).

Urban planning is one of the main applications of GIS. Urban planners use GIS both as a spatial
database and as an analysis and modeling tool (Fig1). Urban planning involves many functions,
scales, sectors, and stages. In general, the functions of urban planning can be classified into
general administration, development control, plan making, and strategic planning. General
administration and development control are relatively routine planning activities, whereas plan
making and non-routine strategic planning are undertaken much less frequently (A G-o Yeh,
1999).
 8

Fig1. GIS and urban planning (Source: A G-o Yeh, 1999)

The scale of the planning area covered can range from a whole city, to a sub-region of a city, a
district, or a street block. The most frequently involved sectors of urban planning are land use,
transport, housing, land development, and environment. At each scale of planning there are
different stages: the determination of planning objectives; the analysis of existing situations
modeling and projection; development of planning options; selection of planning options; plan
implementation; and plan evaluation, monitoring, and feedback. Different functions, scales,
sectors, and stages of urban planning make different uses of GIS. The use of the data
management, visualization, spatial analysis, and modeling components of GIS varies according
to different functions of urban planning (Bardon et al, 1984)

2.4. Site search procedures using standard gis map overlays

The application of digital map overlays for area screening purposes is a classic example of the
applied use of one subset of tools from the GIS. In the use of map overlay routines to search a
specified region of interest for areas suitable for a specified use, some sitting factors are
considered; numeric and qualitative criteria are applied to these sitting factors in an exploratory
investigation of the search area. It is noted here, however, that that such overlay procedures can
do little more than identify areas which simultaneously satisfy all the specified criteria (Steve,
1991).
 9

Overlays are ideal for area screening using deterministically defined sitting criteria but have
certain limitations when dealing with information of a non-deterministic nature. These are
summarized by Janssen and Rietveld (1990) cited by (Berry, 1987) as follows:

(1) Digital map overlays are difficult to comprehend when more than 4 or 5 factors are involved;

(2) Most GIS overlay procedures do not allow for the fact that variables may not be equally
important;

(3) When mapping variables for overlay analyses the problem arises of how the threshold values
used there in are defined (clearly, the outcome of an area screening exercise depends strongly on
the choice of threshold values); and

(4) The use of threshold values to map continuous variables, such as population density, on a
nominal basis will inevitably lead to substantial loss of information.

There is nothing in an analysis of this kind that informs the user which individual site(s) offer the
most promising characteristics for development. It is suggested that additional procedures, based
here on multi-criteria evaluation (MCE) techniques, are required to evaluate the level of
suitability of sites falling within the feasible areas identified using standard GIS overlay
procedures (Steve, 1991).

2.5. Multi criteria decision making (MCDM)

The techniques adopted in the various approaches of decision analysis are called multi criteria
decision methods (MCDM). These methods incorporate explicit statements of preferences of
decision-makers. Such preferences are represented by various quantities, weighting scheme,
constraints, goal, utilities, and other parameters. They analyze and support decision through
formal analysis of alternative options, their attribute, evaluation criteria, goals or objectives, and
constraints. MCDM used to solve various site selection problems. However, they assume
homogeneity within the study area, which is unrealistic for site selection problems (Malczewski,
1999)

The choice of the MCDA method is very important since it has a significant effect on the final
outcome. MCDA characteristics and properties should be compatible with the specific nature of
 10

the decision problem. For example, some MCDA techniques efficiently handle a continuous set
of alternatives and criteria belonging to the same domain (e.g. economic). Other MCDA methods
can only consider a small set of discrete alternatives but are more efficient to handle
heterogeneous criteria (Florent et al. 2001). If there is a conflict between the various actors, they
can negotiate the subjective parameters, like the weights associated with each criterion before
adopting a common set of values. It is also possible to repeat the MCDA process and thus select,
for each different group of stakeholders, a solution that is adapted to its specific needs. MCDA
results can be mapped in order to display the spatial extent of the best areas or index of land
suitability. The negotiating parties can then discuss and compare the results by overlaying these
maps, which are in fact geographical representations of their own set of preferences
(Malczewski, 1999).

2.6. Spatial multi criteria decision making

Spatial multi criteria decision problems typically involve a set of geographically-defined

alternatives (events) from which a choice of one or more alternatives is made with respect to a
given set of evaluation criteria. Spatial multi-criteria analysis is vastly different from
conventional MCDM techniques due to inclusion of an explicit geographic component. In
contrast to conventional MCDM analysis, spatial multi-criteria analysis requires information on
criterion values and the geographical locations of alternatives in addition to the decision makers’
preferences with respect to a set of evaluation criteria. This means analysis results depend not
only on the geographical distribution of attributes, but also on the value judgments involved in
the decision making process. Therefore, two considerations are of paramount importance for
spatial multi-criteria decision analysis:

(1) The GIS component (e.g., data acquisition, storage, retrieval, manipulation, and analysis
capability); and

(2) The MCDM analysis component (e.g., aggregation of spatial data and decision makers’
preferences into discrete decision alternatives) (Steve, 1991). The major elements involved in
spatial multi-criteria analysis are shown in Figure 2.
 11

Fig2. Decision flowchart for spatial multicriteria analysis (Source: A G-o Yeh, 1999)

2.7. GIS and multi criteria decision analysis

Spatial decision problems typically involve a large set of feasible alternatives and multiple,
conflicting and incommensurate evaluation criteria (Shattri et al, 2006). The alternatives are
often evaluated by a number of individuals (decision-makers, managers, stakeholders, interest
groups). The individuals are typically characterized by unique preferences with respect to the
relative importance of criteria on the basis of which the alternatives are evaluated (Clarke, 1997).
Accordingly, many spatial decision problems give rise to the GIS-based multicriteria decision
analysis (GIS-MCDA).

These two distinctive areas of research, GIS and MCDA, can benefit from each other
(Malczewski 1999). On the one hand, GIS techniques and procedures have an important role to
play in analyzing decision problems. Indeed, GIS is often recognized ‘as a decision support
system involving the integration of spatially referenced data in a problem solving environment’.
On the other hand, MCDA provides a rich collection of techniques and procedures for
structuring decision problems, and designing, evaluating and prioritizing alternative decisions
(Esri, 2011). At the most rudimentary level, GIS-MCDA can be thought of as a process that
transforms and combines geographical data and value judgments (the decision-maker’s
 12

preferences) to obtain information for decision making. It is in the context of the synergetic
capabilities of GIS and MCDA that one can see the benefit for advancing theoretical and applied
research on GIS-MCDA

2.8. Analytical hierarchy process

The most frequently raised problem in MCDM is how to establish weights for a set of activities
according to importance. Location decisions such as the ranking of alternative communities are
representative multi-criteria decisions that require prioritizing multiple criteria. Saaty (1980) has
shown that this weighting of activities in MCDM can be dealt with using a theory of
measurement in a hierarchical structure. The analytic hierarchy process (AHP) is a
comprehensive, logical and structural framework, which allows improving the understanding of
complex decisions by decomposing the problem in a hierarchical structure.

The incorporation of all relevant decision criteria and their pair-wise comparison allows the
decision maker to determine the trade-offs among objectives. Such multi-criteria decision
problems are typical for housing sites selection. The AHP allows decision-makers to model a
complex problem in a hierarchical structure showing the relationship of the goal, objectives
(criteria), sub-objectives, and alternatives. Uncertainties and other influencing factors can also be
included. It is not only supports decision makers by enabling them to structure complexity and
exercise judgment, but also allows them to incorporate both objective and subjective
considerations in the decision process (Shattri et al, 2006).

2.8.1. Pair-wise comparisons method

The Pair-wise comparisons method was developed by Saaty (1980) in the context of the
Analytical Hierarchy Process (AHP). This method involves pair-wise comparisons to create a
ratio matrix. As input, it takes the pair-wise comparisons of the parameters and produces their
relative weights as output.
 13

2.8.2. Identifying Criteria

The development of urban residential land use is influenced by numerous factors. These include
physical, socio-economic and environmental quality and amenities. (Chapin and Kaiser, 1978).

The first step that was taken in this analysis was to collect all of the data that would be needed to
meet all of the criteria. Criteria were selected to evaluate potential housing sites and to support
decisions concerning the location of additional housing areas. The criteria must be identified and
include factors and constraints. The criteria were selected on the light of Different literatures
(Shattri et al, 2006; Berry, 1987; Steve, 1991; Malczewski, 1999) and local planning guidelines
like Rwanda National Land Use development Master Plan (RNLUMP) and Gasabo District
detailed physical plan (2013). These factors include:

Topography

Topography factors affect the land use planning and the important factors associated with
topography include elevation and steep slopes. From the master plan policies, considered that the
sites on or near cliffs is not suitable for housing development also we have to avoid the high
elevation area because the planning in these areas costs a lot to the government, particularly
supplying the mountains area by facilities like roads, water supply, electricity, and so on, are
much more costly in comparisons with the flat areas. Among the physical factors that are
commonly studied in residential site suitability analysis. Areas with exceeding 10 % are usually
not suitable for residential development (Chapin and Kaiser 1978). The idlest areas for housing
residential use are areas with 2-6 % slopes. However, in master plan in Kigali city the slopes
exceeding 20% is not suitable for residential development.

From the goal of safety

The presence physical hazard reduces the suitability of a site. For selecting safe housing sites and
a void the risks deriving from water. The risks here can arise from flooding in the rainwater
season whether from rivers, wetlands, or low-lying land subject. Each of them would be buffered
according to the severity of the hazard. In addition the location must be far at least 1 km far from
the military camps.
 14

From the goal of minimization of the cost of urban development reducing mobility

Road accessibility is one of the important parameters for urban development as it provides
linkage between the settlements The distance to existing urban areas is important because the
significantly impact moving costs, so the roads are an important factor in housing development
because their presence indicates human activity. The locations must be adjacent to built up areas
(existing neighborhood), in the low-density population areas, within 1-5 km from the main and
secondary roads. (A G-o Yeh, 1999).

From the goal of safeguarding nature areas

• To safeguarding nature areas: No building sites are allowed within the ground water basin.

• The locations must be far at least 3 km from the airport. The exclusion zones may be based on
noise from airport operation, areas affected by aircraft landing pattern, and areas that would
interfere with airport radar. (Shattri, et al, 2006)

The environment aspect

• Housing areas should be located at least 200 meters away from industry areas (Berry, 1987).

• The location must be far at least 5 km from the waste water treatment station.
 15

CHAPTER 3. MATERIAL AND METHODS

3.1. Study area description

The study area, Gasabo District, is one of the three districts of Kigali city; other districts of
Kigali city are Kicukiro and Nyarugenge. Gasabo is the largest of all of the Districts,
divided into 15 sectors with a total area covering 429.3 sq km; this is approximately
60% of the total area of Kigali city. It extends between latitude 1o46 - 1o59 and longitude
29o59 - 29o16 (Figure3)

The Gasabo District forms the northern region of the city of Kigali. It is bounded by
Lake Muhazi, Gicumbi and Rulindo Districts to the north. Kicukiro and Nyarugenge
District border it to the south. To the east and west side it is bounded by Rwamagana
and Rulindo Districts respectively.

Fig3. The location of Gasabo district

 16

The Northern part of the Gasabo District has steep undulating terrain, 40% of the land
has slopes above 20%. Approximately 6% of the land has steep slopes above 40%, The
Southern sectors of the District offer relatively flatter terrain which is more suitable for
large scale urban development. 7% of land within the District is delineated as wetlands
by Rwanda Environmental Management Authority (REMA). The wetlands divide the land
into smaller parcels suitable for developments. Provisions of sustainable and an efficient road
and infrastructure network across the wetlands and steep slopes are challenging. Apart
from the wetlands there is a large network of natural storm water drainage channels,

3.2. Data and materials

To be completed this research has been based on different published materials found in different
libraries such as former NUR Main Library, CGIS-NUR Library. Other information has come
from, reports, internet web pages. Those literatures were exploited to support the statistics,
statements in the research and to get a clear understand of different concepts.

In order to develop Site suitability map for Housing development, different shape files produced
and/ or updated by CGIS-NUR have been used as the input of GIS. A Digital Elevation Model
(DEM) of resolution 10 has also been used to deduct in Slope maps. A series of orhtophotos of
the study (obtained from the Rwanda Natural Resources Authority) area have been used to verify
the land use type provided by the land use shape file.

Down is the list of Data used in the research.

Shape files: Administrative boundaries, Rivers, Wetland, Land use, Roads

Raster files: DEM of resolution 10, Orhtophotos of resolution 5

 17

3.3. Methods

During the map-making process, multiple geographic layers are aggregated to produce maps that
show the appropriateness of the land for housing purpose. While the application of MCE analysis
in GIS is a technical procedure requiring parameters defining different objectives, input from
knowledgeable organizations like REMA and Kigali City in selecting and prioritizing the criteria
to be used in the GIS analysis to produce the final map have been considered. Fig.4 describes the
process we used to generate the map of housing site suitability analysis.

Define the objectives

Identify the criteria

Non-GIS
Constraints
procedure
Factors

Weight the factors

Map the criteria

GIS
Procedure

Run the MCE module in

the GIS Program

Standardized criteria
Set criteria weights
Run the module

Fig4. Used housing site suitability analysis procedure (source: López et al, 2011)
 18

The general formula used to produce a Suitability map is:

Suitability Map = Σ [factor map (cn) * weight (wn) * constraint (b0/1)]

Where, cn = standardised raster cell, wn = weight derived from AHP pair-wise comparison, and
b0/1 = Boolean map with values 0 or 1 (Source: Saaty, 1980)

As one of the Multi-criteria Evaluation Technique, the AHP was used; it is a procedure of
evaluating several criteria that are used to determine an objective.

This study is about identifying potential Sites for Urban Development Using GIS Based Multi
Criteria Evaluation Technique, following the procedure in fig4, (1) on the base of different
literatures we had to identify the necessary parameters that can influence the suitability of sitting
a housing site, (2) those parameters were classified in two categories namely constrains (that will
be used to produce restrictions maps) and factors (used to produce a suitability map regarding a
particular factor or factor maps), (3)after obtaining a restriction map and a factors suitability
map, we combined both of them considering the weight for each factor to produce a final
suitability map.

3.3.1. Restriction map (constraints)

In order to delimit the restricted zone and the permissible zone for housing, the constraints layers
in Gis environment (ArcGis 10.1) were buffered according to the severity of the hazard it may
present. From the goal of safety and environmental protection a 10m Buffer from rivers, 20m
from wetlands, 50m from lake (as described by Rwanda organic law), 200m from industries,
3km from airport, 1km from military zones. After buffering those shape files, they have been
converted to raster data sets to facilitate the overlay operation. By raster calculation tool in
ArcGis 10.1, all restriction maps have been combined and the final product was a restriction map
for the entire study area considering or the restriction criteria. The final product is a raster file
with values 1 representing permissible area for housing site sitting and 0 that represent restricted
areas. The ArcGis 10.1 model used is presented in Appendix1.
 19

3.3.2. Suitability Factors map

Factors are those parameters that influence the suitability of a site for a precise use, they are not
direct constraints for housing purpose but they are criteria used to determine the degree of
suitability of a location compared to others regarding a precise factor of suitability. For the
present study, factors that were considered are: Slope, Elevation, Road proximity and Land use
type.

From a DEM raster of the study area we extracted a slope and elevation raster. A street shape file
of the study area has been converted to raster to facilitate further calculations and analysis, a land
use type shape file has also been converted to raster data model. Finally, all those five raster have
been classified in 5 classes each representing the degree of suitability. Factor classes have been
defined as: 5 extremely suitable, 4 High suitable, 3 suitable, 2 Moderate suitable, 1 Less suitable,
restricted unsuitable. The ArcGis 10.1 model used to generate the suitability factor map is
presented in Appendix2.

3.3.3. Allocation of Weight to factors

For suitability analysis it is necessary to give some score to each of the criteria as per their
suitability for urban development. For this purpose the pairwise comparison matrix using Saaty's
(1980) nine-point weighing scale was applied (table 2).
 20

Table2. Nine point weighting scale for pairwise comparison (Source: Saaty, 1980)

Intensity of importance Definition

1 Equal importance
2 Equal to moderate importance
3 Moderate importance
4 Moderate to strong importance
5 Strong importance
6 Strong to very strong importance
7 Very strong importance
8 Very strong to extremely strong importance
9 Extremely importance

Computation of the criterion weights

To develop a pairwise comparison matrix different criteria are required to create a ratio
matrix. These pairwise comparisons are taken as input and relative weights are produced as an
output.

After the formation of pairwise comparison matrix, computation of the criterion weights has
been done (Table3). The computation involves the following operations:

a) Finding the sum of the values in each column of the pairwise comparison matrix.

b) Division of each element in the matrix by its column total (the resulting matrix is referred to
as normalized pairwise comparison matrix).

c) Computation of average of elements in each row of the normalized matrix, i.e. dividing the
sum of normalized scores of each row by the number of criteria. These averages provide an
estimate of the relative weights of the criteria being compared.

It should be noted that for preventing bias thought criteria weighting the consistency ratio
(CR) was used.
 21

Estimation of the consistency ratio

The next step is to calculate a consistency ratio (CR) to measure how consistent the judgments
have been relative to large samples of purely random judgments. The AHP deals with
consistency explicitly because in making paired comparisons, just as in thinking, people do not
have the intrinsic logical ability to always be consistent (Saaty, 1994).

For estimating consistency, it involves the following operations:

a) Determination of the weighted sum vector by multiplying matrix of comparisons on the right
by the vector of priorities to get a new column vector. Then divide first component of new
column vector by the first component of priorities vector, the second component of new
column vector by the second component of priorities vector, and so on. Finally, sum these
values over the rows (table4).

b) Determination of consistency vector by dividing the weighted sum vector by the

criterion weights. Once the consistency vector is calculated it is required to compute
values for two more terms, i.e. lambda (λ) and the consistency index (CI). The value for
lambda is simply the average value of the consistency vector. The calculation of CI is based on
the observation that λ is always greater than or equal to the number of criteria under
consideration(n) for positive, reciprocal matrices and λ = n, if the pairwise comparison
matrix is consistent matrix. Accordingly, λ-n can be considered as a measure of the degree of
inconsistency.

This measure can be normalized as follows: CI = (λ-n) / (n-1) the term CI, referred to as
consistency index, provides a measure of departure from consistency. To determine the
goodness of C.I., the analytical hierarchy process compares it by random index (R.I.) and the
result is what we call consistency ratio (C.R.), which can be defined as:

CR = CI/RI

Random index is the consistency index of a randomly generated pairwise comparison

matrix of order 1 to 10 obtained by approximating random indices using a sample size of
500 (Saaty, 2000). Table (5) shows the value of R.I. sorted by the order of matrix.
 22

The consistency ratio (CR) is designed in such a way that if CR < 0.10, the ratio indicates a
reasonable level of consistency in the pairwise comparisons; if, however, CR > 0.10, then
the values of the ratio are indicative of inconsistent judgments. In such cases one should
reconsider and revise the original values in the pairwise comparison matrix.

Table3. Pairwise comparison matrix

Criteria Pairwise comparison matrix Normalized Pairwise Computation of

comparison matrix criterion weights
Slope Elevation Road Land (a) (b) (c) (d)
(a) (b) Proximity use
(c) (d)
Slope 1 3 4 4 0.55 0.64 0.36 0.48 0.5
Elevation 1/3 1 3 3 0.18 0.21 0.27 0.36 0.26
Road 1/4 1/3 1 1/3 0.14 0.07 0.09 0.04 0.09
Proximity
Land use 1/4 1/3 3 1 0.14 0.07 0.27 0.12 0.15
Total 1.83 4.66 11 8.33 1 1 1 1 1

Table 4. Calculation of the consistency vector

Criterion Weighted sum vector Consistency vector

Slope [(1)×(0.5)+(3)×(0.26)+(4)×(0.09)+(4)×(0.15)]=2.24 2.24/0.5=4.48
Elevation [(1/3)×(0.5)+(1)×(0.26)+(3)×(0.09)+(3)×(0.15)]=1.146 1.14/0.26=4.38

Road proximity [(1/4)×(0.5)+(1/3)×(0.26)+(1)×(0.09)+(1/3)×(0.15)]=0.35 0.35/0.09=3.9

Land use [(1/4)×(0.5)+(1/3)×(0.26)+(3)×(0.09)+(1)×(0.15)]=0.63 0.63/0.15=4.2

 23

Table5. Random Index (Source: Kumar and Biswas, 2012)

Order Matrix Random Index

1 0.0
2 0.0
3 0.58
4 0.9
5 0.12
6 1.24
7 1.32
8 1.41
9 1.45
10 1.49

It is required to check whether our comparisons are consistent. Table 3 shows the determination
of weighted sum vector and consistency vector.

Calculation of lambda (λ) = (4.48+4.38+3.9+ 4.2)/4 = 4.24

Note: Lambda (λ) is the average of consistency vector.

Condition 1: λ should be equal or greater than the number of criteria under consideration.

The value calculated above satisfies this condition.

Calculation of consistency index (CI)

CI = (λ – n)/ (n-1) = (4.24-4)/ (4-1) = 0.08

Calculation of consistency ratio (CR),

CR = CI/RI = 0.08 / 0.9 (Since RI= 0.9 for n = 4) = 0.09

Condition 2: Consistency ratio CR (=0.09) <0.10 indicated a reasonable level of consistency in

the pairwise comparisons. Therefore, the values obtained satisfy the noted conditions, which
denote that the weights obtained are agreeable.
 24

CHAPTER 4. RESULTS AND DISCUSSION

4.1. Permissible area for potential future development against restricted areas

Following the procedure described in section 3.3.1, permissible and restricted zones regarding a
constrain factor have been identified as follow:

Fig5. Lake constrained area

To the northeastern border of Gasabo district in Rutunga and Gikomero sectors, there is Muhazi
Lake; fig5 shows the prohibited zone for housing from the lake.
 25

Fig6. Airport constrained area

The Kigali international airport has been buffered to show the restricted zone for housing around
the airport (fig6) allowing the proper activities of aviation. Some zones of Kimironko, Remera,
Ndera and Kimihurura sectors are within the restricted zone for housing regarding the airport
constraint.
 26

Fig7. Defense constrained area

Parts of Kinyinya and Bumbogo sectors are within the constrained area for housing because of
their proximity to a military facility or defense area. (fig7)
 27

Fig8. River constrained area

As shown in fig8, the establishment of a housing site should be out of the zones in blue
indicating the river restricted zones.
 28

Fig9. Industry constrained area

Apart from Jali, Kacyiru, Kimihurura, Remera sectors, other sectors of Gasabo district before
establishing a housing site should respect the zones in green showing the industries constrained
areas (fig9).
 29

Fig10. Wetland constrained area

Every sector of Gasabo district comprises at least one wetland. Fig10 shows the restricted zones
for developing a housing site from wetlands.
 30

Fig11. Final Restriction map

Fig11, combines fig5-10 in one and shows all the restrictions to housing development within
Gasabo district.
 31

4.2. Suitability factors maps

Following the procedure described in section 3.3.2, suitability level (degree of suitability) across
the study area regarding a suitability factor has been identified. The final restrictions identified in
section 4.1 have been removed to the study area, therefore the suitability factor maps were
produced using the mask of non-restricted zones.

Fig12. Elevation factor map

 32

Housing development needs a less elevated surface to facilitate the distribution of infrastructures
and preparation of terrain. That is why; zones of Jali, Nduba, Bumbogo, Rutunga, Gikomero and
Rusororo with relatively higher elevation are considered to be less suitable for housing
development (Fig12).

Fig13. Slope factor map

Jali sector is known for its remarkable steep slopes therefore a big part of the sector is classifies
as less suitable for housing; Parts of Nduba, Bumbogo, Rutunga and Gikomero are also less
suitable regarding the slope factor. Kacyiru, Kinyinya, Kimironko, Remera are classified as
suitable. (fig13)
 33

Fig14. Road proximity factor map

A zone of 10 meters from a road has been classified as restricted for housing development. In
fig14, the suitability decreases as we get far from a road.
 34

fig15. Land use/ cover suitability classes

Already developed areas were classified as restricted for new housing development, this explain
the reason why, in Kacyyiru, Kimironko, Remera, Kimihurura, Kinyinya and Gisozi sectors
which are already developed are within the restricted zones. Reserve site, open space and parks
have also been classifies as restricted. (fig15)
 35

Fig16. Housing suitability map

As described in section 3.3, restrictions (constraints) maps have been combined with the factors
map using the map algebra tool in ArcGis to produce the final housing site suitability map
indicating the class (degree) of suitability for housing development across the study area (fig16).
 36

Table6. Suitability classes statistics

Suitability Class Code Suitability Class Total Area in sq km

5 Extremely Suitable 17
4 High Suitable 40.5
3 Suitable 107
2 Moderate Suitable 15
1 Less Suitable 0.000502
Total All potential site for 179.5
Future Development

After obtaining suitability classes in section 4.2, statistics showing the area occupied by each
class in the study area has been determined in table6. In the whole area of Gasabo district which
is 429.3 sq kilometers the land available for potential future urban development calculated as the
sum of all suitability classes is 179.5 sq kilometers; the remaining 249.8 sq kilometers are
occupied by 83.5 sq kilometers of built up area and 166.3 sq kilometers are restricted for urban
development due to constrains described in section 4.1 that make it not appropriate for urban
development.
 37

CHAPTER 5. GENERAL CONCLUSION AND RECOMMENDATIONS

5.1. General conclusion

Concluding, housing sites in Gasabo district as well as in Kigali city are still needed and planners
who are in charge of finding that confront the multi-objectivity and complexity of the site
suitability analysis. GIS-based multi criteria evaluation is revealed to be an efficient tool in site
suitability analysis in general and in housing site suitability analysis in particular as
demonstrated in this study.

The production of a housing suitability map has been produced based on the guidelines used in
urban planning policies and laws by Kigali city, Gasabo district. The results reveal that the
suitability across the study area differs from a place to another

This approach has helped combining multiple and often conflicting objectives determining the
housing suitability for land in one map. A restriction map for each housing constraint has been
produced (rivers, wetlands, lakes, airport, defense, industries) and after a final restriction map
showing permissible in contrast of restricted zone for housing has been produced. Factor
suitability maps showing the degrees of suitability across the study area regarding a given factor
have been produced (slope, elevation, road proximity and land use). Finally, the ‘‘all restrictions
map’’ and the ‘‘all suitability factors map’’ have been combined to generate the final housing
suitability index (classes) across the study area showing the degree of suitability for housing on
the entire surface of the study area.

The AHP used in MCE has been usefully to allow the weighted overlay of factors of different
importance and nature (qualitative ones like land use and quantitative ones like road proximity,
elevation and slope) which is impracticable using the simple map overlays.

This research contributes in the study of the use of GIS-based multicriteria approach in urban
planning in Rwanda and may serve as the basis for further researches but, we recognize that
some factors and necessary consideration from different stakeholders are not applied in the
present study due to time and means limitations and lack of some data
 38

5.2. Recommendations

Based on the findings of the present research, recommendations on the use of GIS-based multi
criteria evaluation in site suitability analysis and on further researches in this domain are
formulated as follow:

1. We recommend to urban planning units in Rwanda in general and to Gasabo district in

particular incorporating in their urban planning framework the new tool, approach (GIS-
based multicriteria evaluation) to efficiently deal with the multi objectivity and
complexity of the site suitability analysis process.
2. GIS- based MCE techniques are easy to use and understand and may help decision
makers, in Rwanda implementing the district master plans
3. We recommend to Kigali city and to Gasabo district to develop GIS- based MCE
framework in their different domain of planning like in disaster management, land slide
control etc.
4. Finally, this research has demonstrated the usefulness of GIS-based MCE in site
suitability analysis but, further researches in this domain in Kigali city are still needed as
some factors of suitability were not considered during the present study due to time,
means and lack of data limitations.
 39

6. REFERENCES

A G-O YEH (1999), Urban planning and GIS, Geographical information system, Second edition,
Volume2 Management issues and applications, 62, pp877-879.

Alam, Md. J. B., Jobair Bin Alam, M., Habibur Rahman, M., Khan, S. K. and Munna, G. M.,
(2006). Unplanned urbanization: Assessment through calculation of environmental degradation
index. Int. J. Environ. Sci. Tech., 3(2), 119-130.

Armstrong M P, Densham P J 1990 Database organisation strategies for spatial decision support
systems. International Journal of Geographical Information Systems4: 3–20.

Bardon K S, Elliott C J, Stothers N 1984 Computer applications in local authority planning

departments 1984: a review. Birmingham, Department of Planning and Landscape, City of
Birmingham Polytechnic.

Berry J K 1987 Fundamental operations in computer-assisted map analysis. International Journal

of Geographical Information Systems1: 119–36.

Berry J K 1993 Cartographic modeling: The analytical capabilities of GIS. In: Good child M,
Parks B and Steyaert L (eds) Environmental Modeling with GIS. Oxford, Oxford University
Press: 58-74

Chapin , F.s.,and Kaiser, E.J., 1978 Urban landuse planning. Urbana, (Chicago: Illinois).

Clarke, K.C. 1997. Getting Started with Geographic Information Systems. Prentice Hall, Upper
Saddle River, N J.

ESRI, Environmental Systems Research Institute, Inc. (2011), GIS for Urban and Regional
Planning, Redlands, California, USA.

Florent Joerin , Marius Thériault & Andre Musy (2001) Using GIS and outranking multicriteria
analysis for land-use suitability assessment, International Journal of Geographical Information
Science, 15:2, 153-174, DOI:10.1080/13658810051030487.

Kigali City - Analysis, benchmarking and vision report. 2013.

 40

Kumar M, Biswas V. 2012. Identification of Potential Sites for Urban Development Using GIS
Based Multi Criteria Evaluation Technique. A Case Study of Shimla Municipal Area, Shimla
District, Himachal Pradesh, India. Journal of Settlements and Spatial Planning, vol. 4, no. 1
(2013) 45-51.

López-Marrero, T.; González-Toro, A.; Heartsill-Scalley, T.; Hermansen-Báez, L.A. 2011.

Multi-Criteria Evaluation and Geographic Information Systems for Land-Use Planning and
Decision Making. [Guide]. Gainesville, FL: USDA Forest Service, Southern Research Station. 8
p.

Malczewski, Jacek (1999). GIS and Multicriteria Decision Analysis. New York: John Wiley &
Sons, Inc.

National Institute of Statistics of Rwanda (NISR), 2012. Population and housing cencus,
provisional results

Saaty, T. L. (2000), Fundamentals of Decision Making and Priority Theory, RWS

Publications Pittsburgh.

Saaty, T. L. (1994), Fundamentals of Decision Making and Priority Theory with the Analytic
Hierarchy Process, RWS Publications Pittsburgh.

Saaty, T.L., (1980). The Analytic Hierarchy Process. McGraw-Hill, New York, 20-25.

Shattri B M, Nordin B A, Rashid S. 2006. GIS Based Multicriteria Approaches to Housing Site
Suitability Assessment. Shaping the Change XXIII FIG Congress. Munich, Germany.

Steve Carter. 1991. Site search and multicriteria evaluation, Planning Outlook, 34:1, 27-36

UNFPA, United Nations Population Fund, 2007, State of world population, New York, USA.

University Press).

WHO, World Health Organisation, 2013, Urban Environment state, New York, USA.
 41

7. APPENDICES
Appendix1. Restrictions Model
 42

Appendix2. Suitability Factors Model

View publication stats