Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 198 (2017) 1128 – 1136

Urban Transitions Conference, Shanghai, September 2016

Land suitability analysis of urban growth in Seremban Malaysia,

using GIS based Analytical Hierarchy Process.
Maher M Aburasa, Sabrina HO Abdullah* a, Mohammad F Ramli a , Zulfa H Asha’aria
a
Faculty of Environmental Studies, Universiti Putra Malaysia Serdang 43400, Selangor Darul Ehsan, Malaysia

Abstract

One of the critical issues in urban planning is the determination of appropriate locations for urban growth in marginal areas
adjacent to large-scale development. This study aims to use a Geographic Information System (GIS) and Analytical Hierarchy
Process (AHP) to choose the best locations of urban growth in Seremban, Malaysia. Various social, economic, environmental,
utilities, and physical factors were used to generate a final land suitability map (LSI) including elevation, slope, soil texture,
population density, land cover, distance to roads, highways, railways, powerlines, streams, industrial, residential, commercial,
and educational areas. The findings of the study reveal that highly suitable areas for urban growth in Seremban reach up to 48%
of its total area, while unsuitable and less suitable areas make up 35%. These findings confirm that unsuitable and less suitable
areas such as forests and agricultural lands cover approximately half of Seremban, which means that these techniques can help
save the ecosystem in marginal areas as well as create future sustainable development.
©©2017
2017TheThe Authors.
Authors. Published
Published by Elsevier
by Elsevier Ltd. is an open access article under the CC BY-NC-ND license
Ltd. This
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the Urban Transitions Conference.
Peer-review under responsibility of the organizing committee of the Urban Transitions Conference
Keywords: Land suitability; urban growth; GIS; Analytical Hierarchy process.

1. Introduction

The suitability analysis for urban growth is considered one of the most important and effective techniques for
identifying the best urban growth locations. This technique employs different types of criteria and weights [1]. Land
suitability evaluation involves the selection of suitable locations of development via mapping of the suitability index
of a specific area [2]. GIS techniques have become a significant tool for controlling and monitoring changes in urban
development and their impact on ecosystems [3]. Land suitability analysis based on GIS environments is a process
that aims to identify the best locations of development while considering environmental sustainability [4].
In setting the importance of the criteria used and computing the weights of factors, GIS tools must be integrated
with other methods to improve the results of land suitability analysis. The integration of GIS tools and multi-criteria

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the Urban Transitions Conference
doi:10.1016/j.proeng.2017.07.155
 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136 1129

decision analysis is a powerful approach for evaluating land suitability [5]. The concept of sustainability generally
leads to improved suitability analysis, which is a complex operation due to various types of factors and criteria that
must be considered in the process [6].
AHP, which was developed by Saaty [7], is currently one of the important techniques for analyzing land
suitability. AHP is categorized under the multi-criteria decision analysis approach and is an effective technique that
helps planners and decision makers to analyze all data before arriving at a final decision for future land-use changes
[5, 8]. AHP has been integrated with GIS tools to identify the importance of the criteria used and to calculate
weights by using a scale of importance and the opinion of experts [9].
AHP is commonly applied to identify the weights of influencing factors on urban growth on the basis of the
analysis functions of GIS. AHP is also a structured approach that can be used for complex cases of making decisions
that include competing criteria [10]. The weights of factors in AHP can commonly be identified by using driven
knowledge and driven data. The weights of factors can also be calculated by using a questionnaire given to
specialists who have considerable experience in the field of urban growth and can then be determined by using
pairwise comparison method to measure their relative importance vis-à-vis one another [5, 9-11].
The main challenge in applying this model is that AHP needs the right experts with the widest knowledge and
experience in the fields of suitability analysis and application to judge the factors in terms of their importance and
weights [5]. Identifying the relative weights of the factors used in land suitability analysis is generally difficult.
Thus, the use of a technique that has a powerful capability to identify the weights is important. AHP is one of the
significant techniques used in analyzing issues related to spatial nature [10].
This study aims to integrate GIS techniques with an Analytical Hierarchy Process (AHP) in order to develop a
final land suitability map based on both the driving forces of urban growth and the importance of each criterion
according to expert opinion.

2. Materials and method

2.1. Study area

Seremban city is the largest district in the Negeri Sembilan State (Figure 1) and is the state’s capital. It occupies a
total land area of approximately 935.78 sq. km and includes the districts of Seremban town, Setul, Labu, Rasah,
Ampangan, Rantau, Pantai, and Lenggeng. Seremban is located approximately 20 km from Putrajaya, the national
capital of Malaysia, and 67 km from Kuala Lumpur, the economic center of Malaysia. The population of Seremban
numbers more than 500,000 people and is expected to increase to 1,000,000 in 2020 [7]. Seremban city was selected
as the study area because it is the biggest city in the Negeri Sembilan state, the economic center of the Negeri
Sembilan state, located near the main developed areas in Malaysia, such as Kuala Lumpur, Putrajaya, and Selangor,
an extension of the urban mass of Kuala Lumpur, and the future center for urban development.

Fig. 1. Overview of the study area.

1130 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136

2.2. Data collection and processing

This study utilized land-use maps of 1984, 1990, 2000, and 2010 from the Department of Agriculture of
Malaysia. These land-use maps were extracted from Spot 2, 4, and 5 images, with a 10 m spatial resolution of Spot
2.4 and 2.5 m of Spot 5. All Spot images were registered and geo-corrected with ground control points using a
Global Positioning System (GPS) and were classified using image-enhancement techniques. A supervised
classification method was used to group and extract all clipped images into land-use categories. Field data were
collected using GPS to assess the accuracy of classification by comparing the classified images with GPS points
from the field for each type of land use. The accuracy assessment values reached acceptable Kappa index values.
This means that the image classification is acceptable. Based on the Anderson scheme, the acceptable Kappa index
value for the accuracy assessment should be higher than 0.85 [8]. The data used to extract main and sub-factors for
land suitability analysis is presented in Table 1.
The accuracy assessment method was applied to the land-use maps of 2000 and 2010 to ensure that the
classification accuracy of the land-use maps generated by the Malaysian Department of Agriculture reached
acceptable values. This study assumed that if the values of the overall accuracy and kappa coefficient of the land-use
maps of 2000 and 2010 were acceptable, then all the land use maps used in this study would also be acceptable.
High-resolution Google maps were used to conduct the accuracy assessment of the 2000 and 2010 maps. Random
samples of each class were used on the basis of the distribution and density of each class in the study area. This
method was easily conducted by using the create random point tool in the ArcGIS 10.2 software. The sample
number of each class was identified on the basis of the stratified random sample. The minimum number of samples
of each class should be 50, except for water, which could be less [12]. This study randomly selected 340 points for
verification. Fifty-five errors were identified. The total accuracy values of the land-use maps of 2000 and 2010 were
92% and 91%, respectively. The kappa coefficients of the land-use maps of 2000 and 2010 were 0.90 and 0.89,
respectively. Thus, the classification of the land-use maps met the study requirements.

Table 1. Data used in the analysis.

Materials Sours Type of data Scale
Land use maps of 1984, 1990, Department of Agriculture, Grid 10
2000, and 2010. Malaysia
Population census Department of Statistics, Statistics 1984-1990-
Malaysia 2000-2010
Topographic map of 2012 Department of Surveying and Map 1:25
Mapping, Malaysia(JUPEM)
Digital Elevation Department of Surveying and Grid 10
Model(DEM) Mapping, Malaysia(JUPEM)

2.3. GIS Based AHP

In this part of the research, previous studies and expert opinions were used to identify the significant factors that
influence the determination of appropriate sites for urban growth (Figure 2). All maps were generated based on
density, distance, elevation and all criteria affecting the suitability of urban growth [4, 9, 10]. Euclidean distance,
reclassification, conversion, union, raster calculator, and model builder are the major GIS tools that were used in this
part of the research. For suitability analysis using GIS-AHP, it is important to assign scores to each of the factors
based on their suitability for urban growth. To do this, a pairwise comparison matrix using Saaty's nine-level scale
for identifying relative weights was used. The calculation of factor weights was applied after the formation of the
pairwise comparison matrix. The next step involves the computation of Consistency Ratio (CR), which is used to
measure the consistency between the experts’ opinions. The acceptable Consistency Ratio (CR) should be CR <
0.10, which refers to a reasonable level of consistency in the pairwise comparisons. In contrast, CR > 0.10 refers to
ratio values that are indicative of inconsistent judgments.
 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136 1131

Land use maps, soil maps, topographic

maps, digital elevation model

Socio-economic criteria Utilities Physical area Environment

- Population Density - Distance to roads -Slope -Distance to stream

- Distance to commercial -Distance to highway -Elevation -Land cover
-Distance to Educational -Distance to railway -Soil texture
-Distance to residential -Distance to powerline
-Distance to industrial

Developing pairwise comparison matrix

Expert’s
Calculation of the influencing factors weight using AHP opinion

Using GIS tools: Overlay, and raster calculator

Creation of final land suitability map of urban growth in

Seremban

Fig. 2. Stepwise of GIS-AHP for developing final land suitability map of urban growt

3. Results and discussion

3.1. Identification of influencing factors:

In this part of the research, the previous studies were used to identify the significant factors that influence the
determination of appropriate sites for urban growth [4, 17-20]. Additionally, the local level of influencing factors
was determined based on statements gained from local experts working in related Government authorities such as
the Town Planning Departments, Local Councils, and Federal Planning Departments. Four main criteria were
chosen to achieve the aim of this study. Socio-economic, environment, utilities, and physical factors were the main
criteria used in this research. Then, the sub-factors derived from each main criterion were identified (Figure 3).
1132 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136
 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136 1133

Fig. 3. Factors used in GIS-AHP analysis: (a) slope (b) elevation (c) soil texture (d) population density (e) distance to commercial area (f)
distance to educational area (g) distance to residential area (h) distance to industrial area (i) distance to roads (j) distance to highway (k) distance
to railway (l) distance to powerline (m) distance to stream (n) land cover.

3.2. Development of the pairwise comparison matrix

A pairwise comparison matrix was developed with the support of the main and sub-factors. After the
development of the ratio matrix, each main and sub-factor was normalized. Then, the relative weights were
calculated for each factor using the pairwise comparison method (Table 2).

Table 2. Relative weights of main and sub-factors based on the pairwise comparison matrix, normalized pairwise comparison matrix, and
computation of main criterion weights.
Weight of physical factors 0.55 Weight of socio-economic Weight of utilities factors 0.16 Weight of environmental
factors 0.24 factors 0.05
Sub-Factor Weight Sub-Factor Weight Sub-Factor Weight Sub-Factor Weight
Slope 0.72 Population 0.42 Distance to 0.58 Distance to 0.83
Density Road Stream
Elevation 0.20 Distance to 0.12 Distance to 0.25 Land cover 0.17
Commercial Highway
Soil texture 0.08 Distance to 0.09 Distance to 0.12 - -
Educational Railway
- - Distance to 0.32 Distance to 0.05 - -
Residential Powerline
- - Distance to 0.05 - - - -
Industrial
Total 1 Total 1 Total 1 Total 1

3.3. Calculation of Consistency Ratio

The CR is important for identifying whether or not the study’s comparisons are consistent. Condition 1: λ must
be equal or greater than the number of factors used. The value of λ in this study = 4.1, which means that it satisfies
this condition. Computation of consistency index (CI) is done using equation (1):
1134 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136

‫ ܫ‬ൌ ሺߣ െ ݊ሻሺ݊ െ ͳሻሺͳሻ

‫ ܫܥ‬ൌ ሺͶǤͳ െ ͶሻሺͶ െ ͳሻ ൌ ͲǤͲ͵3

Computation of consistency ratio (CR):

ͲǤͲ͵͵
‫ ܴܥ‬ൌ  ൌ ͲǤͶ ‫ ͲͲͳ כ‬ൌ ͶΨሺʹሻ
ͲǤͻ
Condition 2: Consistency ratio, CR (0.04) <0.10, refers to the reliable level of consistency in the pairwise
comparisons. Thus, the CR value meets the requirement of condition 2, indicating that the weights obtained are
accepted. The CR values for the sub-factors of physical, socio-economic, utilities, and environmental criteria are
3.10, 5.16, 4.10, and 0, respectively.

3.4. Generation of final land suitability map

All fourteen sub-factor shapefiles were converted into raster format. Therefore, a score can be identified for each
pixel [21]. Next, all sub-factor maps were combined and overlaid and a final location suitability map was generated
(Figure 4) using the following formula:

Land suitability map=[( slope*0.72+elevation*0.20+soil*0.08)*0.55]+[(population density*0.42+distance to

commercial area*0.12+distance to educational area*0.09+distance to residential area*0.32+distance to industrial
area*0.05)*0.24]+[(distance to roads*0.58+distance to highway*0.25+distance to railway*0.12+distance to
powerline*0.05)*0.16]+[( distance to stream*0.83+land cover*0.17)*0.05].

Fig. 4. Final site suitability map.

The final site suitability map reveals that Seremban can be divided into five suitable categories. The findings
show that suitable and less suitable lands cover about 35.19% of the whole area (Table 3) while highly and
 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136 1135

extremely suitable lands for urban growth covers 48.26% of the area. These results call attention to the need for
proper and sustainable urban planning in order to save Seremban’s ecosystem in the long term.

Table 3. Area under various suitability categories.

Suitability categories Area in km 2 Area in %
Non-suitable 131.92 13.86
Low suitability 203.01 21.33
Moderate suitability 157.55 16.55
High suitability 249.88 26.25
Very high suitability 209.51 22.01
Total 951.87 100

4. Conclusion

This study conducted a Land Suitability Analysis (LSA) to determine the best locations for urban growth in
Seremban using an integrated GIS-AHP model. The results confirm that the GIS-AHP model is a useful technique
for environmental management and urban planning. In addition, the findings illustrate that wide strips of land in
Seremban are unsuitable or less suited for urban growth, which means that there is a need to save the ecosystem in
marginal areas due to the extent of urbanization in developed areas.

Acknowledgements

The study presented here is the part of research project funded by Universiti Putra Malaysia (UPM).

References

[1] K.W. Alexander,, M. Benjamin, and O.P. Grephas, Urban landuse suitability assessment using geoinformation techniques for Kisumu
municipality in Kenya. Int. J. Research and Reviews in Applied Sciences, 2012. 13.
[2] F. M. Joerin, Thériault, and A. Musy, Using GIS and outranking multicriteria analysis for land-use suitability assessment. International
Journal of Geographical information science, 2001. 15(2): p. 153-174.
[3] A.Rosli. Sustainable Urban Forestry Potential Based Quantitative And Qualitative Measurement Using Geospatial Technique. in IOP
Conference Series: Earth and Environmental Science. 2014. IOP Publishing.
[4] S. Park, Prediction and comparison of urban growth by land suitability index mapping using GIS and RS in South Korea. Landscape and
urban planning, 2011. 99(2): p. 104-114.
[5] M. Bagheri, W.N.A. Sulaiman, and N. Vaghefi, Application of geographic information system technique and analytical hierarchy process
model for land-use suitability analysis on coastal area. Journal of Coastal Conservation, 2013. 17(1): p. 1-10.
[6] T.T.Duc, Using GIS and AHP technique for land-use suitability analysis. in International Symposium on Geoinformatics for Spatial
Infrastructure Development in Earth and Allied Sciences. 2006.
[7] T.L. Saaty. The analytic hierarchy process: planning, priority setting, resources allocation. New York: McGraw, 1980.
[8] T.Nguyen Hieu et al., Application of GIS in Land-Use Planning. A case study in the Coastal Mekong Delta of Vietnam. 2006
[9] M. Mohammad, A. Sahebgharani, and E. Malekipour, Urban Growth Simulation Through Cellular Automata (CA), Anaiytic Hierarchy
Process (AHP) and GIS; Case Study of 8th and 12th Municipal Districts of Isfahan. 2013.
[10] M. Javadian, H. Shamskooshki, and M. Momeni, Application of sustainable urban development in environmental suitability analysis of
educational land use by using AHP and GIS in Tehran. Procedia Engineering, 2011. 21: p. 72-80.
[11] E. Zebardas., Application of analytic hierarchy process in urban and regional planning. HONAR-HA-YE-ZIBA, 2002. 10(13): p. 2.
[12] R.G. Congalton., A review of assessing the accuracy of classifications of remotely sensed data. Remote sensing of environment, 1991. 37(1):
p. 35-46.
[13] DOSM, Statistics Yearbook. Department of Statistics Malaysia, 2011: p. 367.
[14] J.R. Anderson., A land use and land cover classification system for use with remote sensor data. Vol. 964. 1976: US Government Printing
Office.
[15] R.B. Thapa, and Y. Murayama, Drivers of urban growth in the Kathmandu Valley, Nepal: examining the efficacy of the analytic hierarchy
process. Applied Geography, 2010. 30(1): p. 70-83.
1136 Maher M. Aburas et al. / Procedia Engineering 198 (2017) 1128 – 1136

[16] M.M.Aburas, et al., A Review of Land Suitability Analysis for Urban Growth by using the GIS-Based Analytic Hierarchy Process. Asian
Journal of Applied Sciences (ISSN: 2321–0893), 2015. 3(06).
[17] S. Pourebrahim, M. Hadipour, and M.B. Mokhtar, Integration of spatial suitability analysis for land use planning in coastal areas; case of
Kuala Langat District, Selangor, Malaysia. Landscape and Urban Planning, 2011. 101(1): p. 84-97. [18] K. Jain and Y.V. Subbaiah, Site
suitability analysis for urban development using GIS. Journal of Applied Sciences, 2007. 7(18): p. 2576-2583.
[18] M. Kumar and V.R. Shaikh, Site Suitability Analysis for Urban Development Using GIS Based Multicriteria Evaluation Technique. Journal
of the Indian Society of Remote Sensing, 2013. 41(2): p. 417-424.
[19] H.Zabihi et al., Land suitability procedure for sustainable citrus planning using the application of the analytical network process approach
and GIS. Computers and Electronics in Agriculture, 2015. 117: p. 114-126.
[20] A.A. Alsharif and B. Pradhan, Urban sprawl analysis of Tripoli Metropolitan city (Libya) using remote sensing data and multivariate logistic
regression model. Journal of the Indian Society of Remote Sensing, 2014. 42(1): p. 149-163.
[21] K. Jain and Y.V. Subbaiah, Site suitability analysis for urban development using GIS. Journal of Applied Sciences, 2007. 7(18): p. 2576-
2583.