Available online at www.sciencedirect.

com

ScienceDirect
Available online at www.sciencedirect.com
Procedia Computer Science 00 (2018) 000–000
www.elsevier.com/locate/procedia
ScienceDirect
Procedia Computer Science 124 (2017) 69–76

4th Information Systems International Conference 2017, ISICO 2017, 6-8 November 2017, Bali,
Indonesia

Spatial data utilization for location pattern analysis

Dyah Lestari Widaningruma,*, Isti Surjandaria, Aniati Murni Arymurthyb
a
Department of Industrial Engineering, Faculty of Engineering, Universitas Indonesia, Kampus Baru UI Depok, Depok 16424, Indonesia
b
Fakultas Ilmu Komputer, Universitas Indonesia, Kampus Baru UI Depok, Depok 16424, Indonesia

Abstract

Retail industry growth in Indonesia is considered significant and will continue, related to Indonesian lifestyle changes. One of the
important things to consider is location determination. The complexity of the location determination problem for the development
of a business requires a comprehensive decision support system, using relevant data and appropriate technology so that it can
accommodate a variety of data formats and information. This paper described the spatial pattern analysis approach to discover
patterns of fast food restaurants. ArcGIS Spatial Autocorrelation (Moran’s I) tool was used to find evidence that the fast food
restaurant distribution in Jakarta, as the biggest city in Indonesia, has a clustered pattern and were not the result of random
chance. The clustered pattern was investigated using Kernel Density Estimation (KDE) to identify the clustered area of fast food
restaurant.

© 2018 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the scientific committee of the 4th Information Systems International Conference 2017.

Keywords: Spatial data; GIS; cluster; fast food restaurant

1. Introduction

Food and beverage industry in Indonesia grew significantly. It becoming one of the key drivers in 2015 which
includes huge investment in the long term [1] and the highest investment value [2]. It has become one of the favorite
industry for many banks in Indonesia in the disbursement of loans in 2016 as it is considered as a promising sector
(addition to infrastructure sector) [3]. This information was in accordance with performance reports submitted by the
Ministry of Industry in 2015 [2], that in 2015 food and beverage industry sector grew 7.54%, the second highest in
the non-oil manufacturing sector (with an average growth of 5.04%).

* Corresponding author. Tel.: +62-21-7888-8805; fax: +62-21-7888-5656.

E-mail address: dyah.lestari61@ui.ac.id

1877-0509 © 2018 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the scientific committee of the 4th Information Systems International Conference 2017.

1877-0509 © 2018 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the scientific committee of the 4th Information Systems International Conference 2017
10.1016/j.procs.2017.12.131
70 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76
2 Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000

Indonesian food consumption pattern has been changed, especially in the large city. This phenomenon has been
affected by the shift of social demography status [4], as well as increased the income of Indonesian people [5] which
also automatically increase the purchasing power of people, including the purchasing power of consumers in the
food service outlets [6]. The Industrial Ministry of Indonesia explains that Trade, Hotel, and Restaurant sector has
14% contribution to Gross Domestic Product (GDP) [7] and the growth rate of GDP for this sector at constant prices
in percent was 5.03 in 2016 [8]. Food service subsector was projected to have 9.1% growth in 2013-2017 [9].
This article is focused on spatial data utilization for location pattern analysis, by using Geographical Information
System. Geographical information has involved with retail business processes including facilities management and
market analysis, to understand the growth possibilities [10]. The ability of GIS to compile and analyze spatial data
provides an overview of the company, including opportunities and challenges. These abilities are important not only
for corporate sustainability but also for the surrounding environment.

2. Spatial Data Utilization

In the early days of the development of geographic information systems, Clarke [11] has been reviewing changes
in the methodology used for location study, as well as to emphasize the importance of considering modeling
procedures were adequate to be able to investigate the company's growth strategy. Many organizations are
committed to assessing retail locations more deeply than traditional method (assisted with a checklist). It is creating
job opportunities in the field of geography. An important thing to consider in location determination are the methods
or approaches or tools for the decision-making process. To obtain the best decision that can support the company's
growth, the methods or approaches or tools should have the ability to accommodate the variety of necessary data for
the consideration. There is much evidence from research results about the capability of Geographic Information
System as a methods or approaches or tools to compile a variety of data types, including spatial and attribute data.
Therefore it can be a comprehensive media to manage, visualize, analyze, and communicate a study, including the
decision about location determination [12, 13]. Retail location analysis is an important part of the retail outlet
location selection, that can be made through Geographic Information System, combining with spatial statistics [13,
14]; integer programming model [15]; Multi-Criteria Decision Making (MCDM) [16-19].

2.1. Spatial data analysis

The spatial information has been powerful tools to gain a comprehensive analysis, both to understand the
behavior and for a specific purpose, such as for determining the location. Johnston [20] conclude that there were two
fundamental questions of geographical research, such as “(1) Are there relationships between phenomena in various
locations?”, and “(2) are places different in terms of the phenomena present there?”.
The early development of GIS at the late 1960s was mostly in North America, USA, and the United Kingdom,
but in accordance with the change of the ownership of data, the spread of GIS implementation in the world are more
prevail [21]. Anselin [22] has formulated various ideas include spatial analysis and social sciences. He has
concluded that the development of theoretical and empirical studies in social sciences have been increased the
interest in spatial analysis, thus interaction between theory, data analysis, and computation is required. The further
study has emphasized the use of spatial analysis approach with multi-disciplinary interactions. But the growth of
knowledge is hampered by the unavailability of data and the lack of a support system for the application of
advanced spatial analysis for the social sciences [23].
Gregory and Healey [24] introduced the history of GIS and critically evaluate how it would affect the historical
practice of geography. Generally, it happens three changes, (1) GIS software more affordable and easier to use, (2)
meta data are increasingly used, (3) even though the capabilities of GIS software for visualizing spatial data and
spatiotemporal growing, but in conceptual and modeling operation of GIS has slowed, particularly due to the
inconsistency of data availability. But now researchers have a higher opportunity to obtain spatial data, provided by
national or international organizations. For example, land use data obtained from “Geospasial untuk Negeri”
(http://tanahair.indonesia.go.id/), an Indonesian official website that provides geospatial data. Additional data are
obtained from OpenStreetMap Indonesia (http://openstreetmap.id/data-openstreetmap-indonesia/), a collaborative
project to create a free editable map of the world or other crowdsourcing such as google map
 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76 71
Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000 3

(https://www.google.co.id/maps/). Satellite image data extraction can be performed using data from open sources
access such as Google Earth (https://www.google.com/earth/) or from USGS, Landsat gallery
(http://landsat.usgs.gov/).
GIS applications in the 1990s began to grow not only in the fundamental area but also in applied geographical,
specifically in decision-making processes in business such as branch location analysis [25]. Heywood et al. [26]
divided the utilization of GIS into several areas, such as for socioeconomic/government (health, transport planning,
urban management, etc), defence agencies (target site identification, intelligence data integration, etc), commerce
and business (market share analysis, target marketing, retail site location, etc), Utilities (network management,
telecommunications, etc), and environmental management (landfill site selection, resource management, etc).
Multiple uses of GIS becomes easier with the ModelBuilder as a programming tool to build a complex or an
automatic task [27]. For an example, Horn [28] build a model of the demand-responsive passenger to design a
transportation system.
Various studies related to spatial data has been done, but there are still many opportunities for development. As
has been done by Allaway et al. [29], which found that at the initial stage, the distance will affect the success of a
new program of the retailer, but then it will be influenced by other store attributes and other marketing tools. This
analysis can be developed with variable competitors in the surrounding area and the need for differentiation between
the behavioral dimension and psychographic. Zhu et al. [30] produced a flexible architecture for spatial analysis of
data linkage and integrated services in an integrated urban area in China. Spatial Data linkage acts as middleware for
data spatiotemporal, transient analysis service using the methods and parameters of semantic metadata and
computing components to help users to operate and understand the analytical method available, by integrating 11
layers on a network that consists of 51 sources of data, such as statistical data of households, population, and
economic status. Further research is needed to reconstruct automatically with contextual information in publicly
available data services in general.
Panori et al. [31] developed static spatial microsimulation models that combine census data from small areas
with social survey data, for analysis of income distribution and poverty in Athens. From the results of the simulation
model that generates micro data set from a small area to see the socioeconomic changes between the period 2006
and 2011. The developed method has proved to be an alternative for estimating the variables of geographic
distribution when the spatial data on a micro level is not available, it can be an effective solution for the analysis of
spatial and socio-economic impacts on urban areas, national social policies (for 'what-if' scenario analysis).
GIS has become a comprehensive method for the decision-making process, by using various data. As has been
done by Makri and Folkesson [32] who used the shortest distance network application and the actual driving time in
ArcGIS for analysis of land use and traveling; Gudes et al. [33] developed public participation approach to assist the
government in developing the local master plan, with the support of GIS and Web; Luo [34] utilized GIS-based
approach to detect floating catchment shortage of medical services in a given region; Mazinga [35] illustrated GIS-
based techniques that are useful in the research methodology of social development today, to find spatial patterns
that indicate the level of patient satisfaction in the healthcare industry, where it reinforces the argument of capability
comprehensive methodology, which involves not only economic and social factors but also factors that have
relevance to spatial demographic data or other data that describes the geographical conditions of a particular region;
Lai et al. [36] concluded that GIS and AHP approach were considered very useful for spatial analysis of complex
data, which utilizes network data, population density, costs incurred, by using network analysis, buffer analysis, as
well as AHP method for weighting each criterion that must be met.

2.2. Spatial data for location analysis

One of spatial analysis approach for location analysis is catchment area i.e. using buffer analysis and Voronoi
diagram [25, 34, 37-40]. For example, Geertman et al. [40] has used an analytical tool (flow map) and catchment
area to relocate the ambulance facilities, so its distribution becomes more optimal. Dolega et al. [37] produced the
concept and estimation of empirical catchment area of a retail store or a shopping center, and serves as well as test
model of the network of retail centers on a national scale, with the analysis of spatial interaction between potential
customers with a hierarchical network of retail centers to estimate the probability of customers choose store certain
and the development of area catchment based on the size of retail centers, the proportion of units of entertainment,
72 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76
4 Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000

store / main unit (anchor), a trip to the retail center (cost, time, and distance), network street, walking distance. It has
not included the qualitative factors that may affect customers in choosing a retail such as the level of cleanliness,
safety, or operating hours; necessary comparative approach other than gravity models to model the area with a high
level of competition; have not considered the impact of population movements on the model; do not consider
external factors that are going on conditions / certain time, for example, is currently developing an online sales.
Others spatial analysis have been conducting, as has been done by Murray [41] who perform location modeling
approaches using GIS; Uelmen [42] analyzed the siting wind turbine based on the parameters water depth, wind
speed, distance to the coastline, shipping routes, and urban areas; Abramovich [43] performed an analysis to
determine the location using GIS methods based on maximal distance, travel time, and network analysis, using tools
and extensions such as ArcGIS Spatial Analyst and network analyst to determine the location of the fire station; Al-
Marwani [44] performed spatial and socioeconomic data analysis using GIS to assist understanding of the causal
links that can be used to predict the cost of real estate; Cui and Mahoney [45] integrated Google Map APIs and
Internet GIS into an online survey instrument to collect data related to the movement of the spatial reference to use
boats to demonstrate how GIS can be used for the Internet survey tool.

3. Methods

Fast food restaurant has become the object of this study. The coverage is within Jakarta, as the biggest city in
Indonesia. Spatial data used in the study was obtained from various sources. This is due to the limited information
from a single dataset [46].

Data Layer
Preparation

Conversion Spatial
Data Spatial
Autocorrelation

Geoprocessing

Kernel Density
Estimation
Interpretation

Fig. 1. The analytical procedures

The analytical procedures as outlined in Fig. 1. are consist of data layer preparation and conversion spatial data in
the early step. The borderline layer of Jakarta is in UTM Projected Coordinates System GCS_WGS_1984, obtained
from BPS-Statistics Indonesia. The fast food restaurant locations were geocoded based on their street address, found
on the website for each brand. The next step was geoprocessing, consist of Spatial Autocorrelation and Kernel
Density Estimation; following by interpretation the result.

4. Results and discussion

Fast food restaurants discussed in this research consist of top 5 global brand name in chained fast food in
Indonesia, a source from Euromonitor International – 2016 in Minister of Agriculture and Agri-Food [9], i.e. KFC,
Es Teler 77, McDonald’s, A&W, and Hokben. These five brands are selected based on the number of outlets in
Jakarta so that any pattern is easier to learn. The distribution of these outlets is visualized using ArcGIS® Desktop
10.0 for Windows®, as can be seen in Fig. 2. (a).
The Moran’s I function has applied to find the clustered pattern of values across the study area, using a set of fast
food restaurant locations [47]. Fig. 2. (b) describe the fast food grid distribution, result from grid creation and spatial
 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76 73
Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000 5

join operation, used for Moran’s I calculation. From both figures, it appears that there is a tendency that the location
of the restaurant is clustered, and tends to spread more in the center of the city, with the number of outlets that are
large in one grid, compared to the grid on the perimeter outside Jakarta.

(a) (b)
Legend
KFC
Es Teller 77
Mc Donald's
A&W
HokBen

Jakarta
CITY
West Jakarta
Central Jakarta
South Jakarta
East Jakarta
North Jakarta
Thousand Island

Fig. 2. (a) The distribution of 5 global brand name fast food restaurant in Indonesia; (b) Fast food grid distribution.

Spatial autocorrelation report in Fig. 3. (a) and Global Moran’s I Summary in Fig. 3. (b), the Moran’s I index
was 0.673, with z-score 6.21, show that the outlet distribution significantly clustered, with only less than 1% that
this clustered pattern was the result of random chance (based on the result of spatial autocorrelation – Moran’s I
using spatial statistics tools for analyzing patterns of ArcGIS® Desktop 10.0 for Windows®). This shows that there
is a certain rule that encourages food service providers in determining the location of the restaurant. The proximity
of various brands within a certain coverage, indicating that the inter-brand co-location. But to be sure, there needs to
be further study. Co-location study is also important to do to prove the existence of a certain rule related to the
existence of this fast food restaurant.

(a) (b)

Fig. 3. (a) Spatial Autocorrelation Report; (b) Global Moran’s I Summary.

Estimation using feature class of fast food restaurant’s location in Jakarta is conducted using kernel density
method. Fig. 4. (a) shows kernel density surface, based on the number of fast food restaurant around Jakarta and
74 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76
6 Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000

distribute them smoothly, so it provides average surface estimation [48]. Kernel density estimation allows us to
observe both the centrality and agglomeration of existing outlets [49].

(a) (b)

Legend
1
2
3-4
5-6

7-9

Kernel Density Estimation

VALUE
0.003374136 - 0.003374136
0.003374136 - 0.003374136
0.003374136 - 0.003374137
0.003374137 - 0.003374137
0.003374137 - 0.003374137
0.003374137 - 0.003374137 Legend
0.003374137 - 0.003374138
Standard Deviational Ellips
0.003374138 - 0.003374138
Fast Food Restaurant's Grid
0.003374138 - 0.003374138

Fig. 4. (a) Kernel density estimation of fast food restaurant in Jakarta;

(b) Standard deviational ellipse of fast food restaurant in Jakarta.

The location of fast food restaurant in Jakarta tends to have a directional trend, as can be seen in Fig. 4. (b).,
called as a standard deviational ellipse. This trend, which reflects the orientation of location distribution, is
calculated from the standard deviations of the X and Y coordinates separately from the mean center [47].

5. Conclusion

The complexity of the location determination problem for the development of a business requires a
comprehensive decision support system, using relevant data and appropriate technology so that it can accommodate
a variety of data formats and information. Geographic Information System has proved its ability to support this
necessity. In this research, GIS has used for compiling various data, as well as for analyzing data. Fast food
restaurant in Jakarta as the object of this study has significant differences with a random pattern. These locations,
which represented the locations of 5 global brand name of fast food restaurants in Jakarta, were clustered.
It was also found that the density of the fast food restaurants (based on 5 brands data) was in the center of the
region. The outlet's distribution also proved statistically has a specific direction, i.e. northeast – southwest trend.
This still needs further investigation, to find its relationship with another region outside Jakarta.

References

[1] Indonesia Finance Today. (2015). Investasi Sektor Makanan Topang Pertumbuhan Industri pada 2015. Available:
http://kemenperin.go.id/artikel/10726/Investasi-Sektor-Makanan-Topang-Pertumbuhan-Industri-pada-2015
[2] Kementerian Perindustrian RI, "Laporan Kinerja Kementerian Perindustrian Tahun 2015," Biro Perencanaan, Jakarta, Indonesia, 2016.
[3] R. A. Arfandi and I. Oeij. (2016). Indonesian Banks Look at Food & Beverage Sector for 2016 Credit Growth.
[4] Minister of Agriculture and Agri-Food, "The Indonesian Consumer: Behaviour, Attitudes, and Perceptions Toward Food Product," Her
Majesty the Queen in Right of Canada, Ottawa 2011.
[5] J. Dyck, A. Woolverton, and F. Rangkuti, "Indonesia’s Modern Food Retail Sector: Interaction With Changing Food Consumption and
Trade Patterns," U.S. Department of Agriculture, Economic Research Service 2012.
[6] Minister of Agriculture and Agri-Food, "Indonesia: The Foodservice Industry," Her Majesty the Queen in Right of Canada, Ottawa 2013.
[7] Kementerian Perindustrian RI, "Industri Dalam Angka ", Kementerian Perindustrian, Edisi XI Jakarta: Pusat Data dan Informasi, 2011.
[8] Kementerian Perindustrian RI, "Laporan Analisis Perkembangan Industri Edisi Oktober 2016," Kementerian Perindustrian RI, Edisi
Oktober 2016 Jakarta, 2016.
 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76 75
Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000 7

[9] Minister of Agriculture and Agri-Food, "Foodservice Profile - Indonesia," Her Majesty the Queen in Right of Canada, Ottawa, Canada
2016.
[10] ESRI. (2007). GIS Best Practices: GIS for Retail Business. Available: http://www.esri.com/library/bestpractices/retail-business.pdf
[11] G. Clarke, "Changing methods of location planning for retail companies," GeoJournal, vol. 45, pp. 289-298, 1998.
[12] G. M. E. Hartoyo, Modul pelatihan sistem informasi geografis (SIG) tingkat dasar, 2010.
[13] N. Duggal, "Retail Location Analysis: A Case Study of Burger King & McDonald’s in Portage & Summit Counties, Ohio," degree of
Masters of Arts, College of Arts, Kent State University, 2007.
[14] R. Chen, Y. Hsu, Y. Ye, and C. Huang, "Prediction of Convenience Store Location based on Support Vector Machines," International
Journal of Digital Content Technology and its Applications, vol. 6, pp. 248-255, 2012.
[15] İ. Önden, C. G. Şen, and A. Şen, "Integration of Integer Programming with GIS Analyzing Abilities for Determining the Convenience
Levels of Retail Stores," Procedia - Social and Behavioral Sciences, vol. 62, pp. 1144-1149, 2012.
[16] A. Rikalovic, I. Cosic, and D. Lazarevic, "GIS Based Multi-criteria Analysis for Industrial Site Selection," Procedia Engineering, vol. 69,
pp. 1054-1063, 2014.
[17] A. Ohri and P. K. Singh, "GIS based environmental decision support system for municipal landfill site selection," Management of
Environmental Quality: An International Journal, vol. 24, pp. 583 - 598, 2013.
[18] G. Tavares, Z. Zsigraiova, and V. Semiao, "Multi-criteria GIS-based siting of an incineration plant for municipal solid waste," Waste
Manag, vol. 31, pp. 1960-72, Sep-Oct 2011.
[19] G. P. Panagopoulos, G. D. Bathrellos, H. D. Skilodimou, and F. A. Martsouka, "Mapping Urban Water Demands Using Multi-Criteria
Analysis and GIS," Water Resources Management, vol. 26, pp. 1347-1363, 2012.
[20] R. J. Johnston, Multivariate Statistical Analysis in Geography. England: Longman Scientific & Technical, 1986.
[21] J. T. Coppock and D. W. Rhind, "The History of GIS," in Geographical Information Systems: Principles and Applications vol. Principles,
D. J. Maguire, M. F. Goodchild, and D. W. Rhind, Eds., ed England: Longman Scientific &Technical, 1991.
[22] L. Anselin, "The future of spatial analysis in the social sciences," Geographic information sciences, vol. 5, pp. 67-76, 1999.
[23] M. F. Goodchild, L. Anselin, R. P. Appelbaum, and B. H. Harthorn, "Toward spatially integrated social science," International Regional
Science Review, vol. 23, pp. 139-159, 2000.
[24] I. N. Gregory and R. G. Healey, "Historical GIS: structuring, mapping and analysing geographies of the past," Progress in Human
Geography, vol. 31, pp. 638-653, 2007.
[25] J. R. Beaumont, "GIS and Market Analysis," in Geographical Information Systems: Principles and Applications vol. Applications, D. J.
Maguire, M. F. Goodchild, and D. W. Rhind, Eds., ed England: Longman Scientific & Technical, 1991, pp. 139-151.
[26] I. Heywood, S. Cornelius, and S. Carver, An Introduction to Geographical Information Systems: Pearson Education, 2011.
[27] D. W. Allen and J. M. Coffey, GIS Tutorial 3: Advanced Workbook. California: ESRI Press, 2011.
[28] M. E. T. Horn, "Modelling and Assessment of Demand-Responsive Passenger Transport Services," in Applied GIS and Spatial Analysis, J.
Stillwell and G. Clarke, Eds., ed England: John Wiley & Sons, Ltd., 2004, pp. 171-194.
[29] A. W. Allaway, D. Berkowitz, and G. D’Souza, "Spatial diffusion of a new loyalty program through a retail market," Journal of Retailing,
vol. 79, pp. 137-151, 2003.
[30] X. Zhu, B. She, W. Guo, S. Bao, and D. Chen, "Integrating Spatial Data Linkage and Analysis Services in a Geoportal for China Urban
Research," Transactions in GIS, vol. 19, pp. 107-128, 2015.
[31] A. Panori, D. Ballas, and Y. Psycharis, "SimAthens: A spatial microsimulation approach to the estimation and analysis of small area income
distributions and poverty rates in the city of Athens, Greece," Computers, Environment and Urban Systems, 2016.
[32] M.-C. Makri and C. Folkesson, "Accessibility measures for analyses of land use and travelling with geographical information systems,"
Department of Technology and Society, Lund Institute of Technology, Lund University & Department of Spatial Planning, University of
Karlskrona/Ronneby, Sweden, 1999.
[33] O. Gudes, E. Stern, and T. Svoray, "Geographical information systems on the web for public participation in planning," Geoinformatics,
2004.
[34] W. Luo, "Using GIS-Based Floating Catchment Method to Assess Areas with Shortage of Physicians," Health & Place, vol. 10, pp. 1-11,
2004.
[35] Gideon Mazinga, "A geospatial approach for new perspectives on satisfaction with health services in Malawi," Degree of Doctor of
Philosophy, School of Science Technology, Loma Linda University, 2008.
[36] W. Lai, L. Han-Lun, L. Qi, C. Jhing-Yi, and C. Yi-Jiao, "Study and implementation of fire sites planning based on GIS and AHP," in
Procedia Engineering, 2011, pp. 486-495.
[37] L. Dolega, M. Pavlis, and A. Singleton, "Estimating attractiveness, hierarchy and catchment area extents for a national set of retail centre
agglomerations," Journal of Retailing and Consumer Services, vol. 28, pp. 78-90, 2016.
[38] I. Lee and C. Torpelund-Bruin, "Geographic knowledge discovery from Web Map segmentation through generalized Voronoi diagrams,"
Expert Systems with Applications, vol. 39, pp. 9376-9388, 2012.
[39] D. L. Widaningrum, "A GIS – Based Approach for Catchment Area Analysis of Convenience Store," Procedia Computer Science, vol. 72,
pp. 511-518, 2015.
[40] S. Geertman, T. d. Jong, C. Wessels, and J. Bleeker, "The Relocation of Ambulance Facilities in Central Rotterdam," in Applied GIS and
Spatial Analysis, J. Stillwell and G. Clarke, Eds., ed England: John Wiley & Sons, Ltd., 2004, pp. 215-232.
[41] A. T. Murray, "Advances in location modeling: GIS linkages and contributions," Journal of geographical systems, vol. 12, pp. 335-354,
2010.
[42] B. H. Uelmen. (2010, Using Geographic Information Systems to Analyze Suitable Locations for Water Wind Turbine Farms in Lake
Michigan.
[43] A. A. Abramovich, "Using GIS to Assist Location and Site Selection Decisions," University of Washington, 2012.
[44] H. A. Al-Marwani, "An approach to modeling and forecasting real estate residential property market," 2014.
[45] Y. Cui and E. Mahoney, "Employing Internet GIS Surveys to Characterize Recreational Boating Travel Patterns," Transactions in GIS, vol.
19, pp. 42-62, 2015.
76 Dyah Lestari WIDaningrum et al. / Procedia Computer Science 124 (2017) 69–76
8 Dyah Lestari Widaningrum et al./ Procedia Computer Science 00 (2018) 000–000

[46] A. Ennis, L. Chen, C. D. Nugent, G. Ioannidis, and A. Stan, "High-level geospatial information discovery and fusion for geocoded
multimedia," International Journal of Pervasive Computing and Communications, vol. 9, pp. 367-382, 2013.
[47] D. W. Allen, GIS Tutorial 2: Spatial Analysis Workbook. California: ESRI Press, 2016.
[48] W. L. Gorr and K. S. Kurland, GIS Tutorial 1: Basic Workbook. California: ESRI Press, 2013.
[49] T. Reigadinha, P. Godinho, and J. Dias, "Portuguese food retailers – Exploring three classic theories of retail location," Journal of Retailing
and Consumer Services, vol. 34, pp. 102-116, 2017.