Advanced GIS – 2023; 3(1); 14-21

Advanced GIS
http://publish.mersin.edu.tr/index.php/agis/index
e-ISSN:2822-7026

Landslide susceptibility analysis with multi criteria decision methods; a case study of Taşova
Melike Öcül*1 , Aziz Şişman2
1Ondokuz Mayis University, Faculty of Engineering, Department of Geomatics, Samsun, Turkey

Keywords ABSTRACT
MCDM Natural or artificial based disasters threaten to humans. In order to minimize the loss of life
FUCOM, and property that may occur after a disaster, various studies are carried out. One of these
AHP, studies is disaster risk maps. To produce disaster risk maps, the criteria which affecting the
GIS disasters should be determined. In order to evaluate more than one criterion Multi-Criteria
Landslide Susceptibility Decision Making Methods (MCDM) and Geographical Information Systems (GIS) are used.
MCDM methods are used both to weight criteria and to rank among alternatives. In the current
study Taşova district of Amasya province is used as study area and two different Landslide
Susceptibility Maps were produced based on Analytical Hierarchy Process (AHP) and Full
Consistency Method (FUCOM) for this region. A total of twelve criteria were determined for
production of risk map and raster data was produced by performing various spatial analyzes
Research Article for the current criteria. Two different landslide susceptibility maps were obtained by giving
Received: 15/01/2023 criterion weights to the generated raster data. It was observed that, Risk-free area, low risk
Revised: 24/03/2023 area and high risk area rates are almost equal, but medium risk area and risk area rates are
Accepted: 28/03/2023 different in two different weighting methods.
Published: 06/04/2023

1. Introduction these risk maps is landslide susceptibility analysis.

(Gökkaya, 2014; Cankaya, 2016; Acar, 2019)
Nowadays, studies are carried out to produce Considering the damage and losses caused,
disaster risk maps in order to reduce the loss of life and landslides, which are in the second place after
property in disasters. The studies are aimed at pre- earthquakes in Turkey, have caused many losses of life
disaster vulnerability analysis and post-disaster and property until today (Acar, 2019). According to the
evacuation resistance analysis. When these objectives Ergünay (2007), since the beginning of the 20th century
are combined, disaster risk maps will emerge. There are a total of 89500 house affected by landslides and rock
many known disaster types and multiple criteria falls in Turkey which is 14% of all natural disasters. A
affecting these disasters. It is very difficult to evaluate landslide is a situation where rock, soil or pieces of land
these criteria at the same time. Therefore, Multi-Criteria shift or move noticeably down the slope due to gravity or
Decision Making Methods are needed. Multi-criteria external factors such as earthquakes, heavy rains.
decision-making (MCDM) is a method that enables the (Disaster Management Dictionary). Although a landslide
selection of the best choice among the criteria applied is a natural disaster, the human factor also triggers it.
simultaneously and more than once (Zahedi, 1986; Examples of human factors such as unknowingly felling
Ishizaka & Labib, 2009; Kabak et al., 2018; Arslankaya & trees, unauthorized mining, inadequate retaining walls
Göraltay, 2019; Boyacı, 2020). Before evaluating the on the roadside. Therefore, it allows to determine the
criteria, spatial analysis of the data should be done. After places with landslide risk and to act carefully in those
the spatial analysis of the data is done with Geographic areas. Thus, the loss of life and property is minimized.
Information System (GIS) or Remote Sensing (RS) Different kind of MCDM techniques were used in
methods, weights are assigned to the criteria with a landslide studies, Analytical Hierarchy Process (AHP) is
determined criterion weighting method. The criterion one of the most used method. Landslides are affected
with the highest weight will affect the risk map more, many factors, and all factors have different weight on
while the criterion with the least weight will affect the landslide. In the current study, it was focused to
risk map less. Thus, more results that are reliable will be determine weight of the factors which affect landslides
obtained. using two different methods which are Full Consistency
The most common risk analyzes in the literature Method (FUCOM) and the AHP method. To achieve this,
review are landslide susceptibility analysis, flood risk, two different maps were produced based on both
earthquake, forest fires, and tsunami risk maps. One of

* Corresponding Author Cite this article

*([email protected]) ORCID ID 0000–0001–9949–0115 Öcül, M., & Şişman, A., (2023). Landslide susceptibility analysis with multi criteria decision methods; a
([email protected]) ORCID ID 0000–0001–6936–5209 case study of Taşova. Advanced GIS, 3(1), 14-21.
 Advanced GIS – 2023; 3(1); 14-21

methods of landslide susceptibility analysis for the The study was carried out in in Taşova district of
Taşova district of Amasya province. Amasya province. Taşova is located on 40° 46' 36" north
latitude and 36° 13' 12" east longitude (Figure 1). The
2. Material and Method District has an area of 1051 km2. The lowest altitude is
170 m where Karlık Stream meets Yeşilırmak. The
In this section, the study area described, AHP and highest altitude is Cami Hill, located in the South of
FUCOM methods are shortly explained, and criteria for Esençay Village, 1956 m. Regional landslides have been
landslide susceptibility analysis are determined. observed during times of heavy rainfall.

2.1. The study area

Figure1. Study area

2,4,6,8 Intermediate values
2.2. Analytical hierarchy process (AHP)
Table 2. Pairwise Comparison Matrix (aij) (Ci: Criterion,
The AHP method was developed by Thomas Saaty in i: 1, 2…n)
1980 that provides a basis for comparing decision- C1 C2 C3 …. Cn
making criteria in a mathematical structure by creating a C1 a11 a12 a13 … a1n
hierarchical structure. C2 a21 a22 a23 … a2n
Organizing goals, attributes, issues, and C3 a31 a32 a33 … a3n
stakeholders in a hierarchy serves two purposes; first …. …. …. …. … ….
provides an overview of the complex relationship vessels C4 an1 an2 an3 … ann
inherent in the situation, second helps the decision
maker to assess whether the problems at all levels are of
The third step is to determine the weights of the
the same magnitude, so that they can accurately compare
criteria. In the pairwise comparison matrix, the sum of
these homogeneous elements (Saaty, 1994).
each column is taken and divided by each element in the
In the first stage of the method, a hierarchical model
column and matrix B is obtained. If we divide the row
is created which shows the relations between the aim,
sum of matrix B by the number of criteria, that is, if the
criteria and alternatives to be obtained by taking expert
arithmetic average of the row is taken, the weights of
opinion for the solution of the problem.
each criterion will be found (Equation 1).
In the second stage, each criterion is compared with
𝑛=1
other criteria and values are assigned according to the
𝑊𝑖 = ∑ 𝑏𝑖,𝑗 (1)
importance scale in Table.1 prepared by Saaty. In line 𝑗=1
with these values, the degree of importance of the criteria 𝑖 = 1, 2, 3 … 𝑛; 𝑗 = 1, 2, 3 … 𝑛
to each other is filled according to the pairwise
comparison matrix in Table 2. With these values, nxn The consistency ratio (CR) of the measures is
dimensional pairwise comparison matrix is created for n calculated in the last step. If the consistency ratio (CR) is
criteria. less than 0.1, the comparisons are consistent; else, the
Each criterion is evaluated mutually, ignoring the comparisons are inconsistent (Equation 2-4). That is, the
other criteria. Evaluation of the criteria in this way is closer the CR is to zero, the higher the consistency of the
advantageous in cases where the number of criteria is decision matrix (Jian-Zhong et al., 2008). No matter how
high (Yılmaz, 2010). mathematically consistent the AHP has in itself, the
Table 1. Significance Scale by Saaty realism of the results will depend on the consistency of
Importance Values Value Definitions the judgment of the decision maker in the one-to-one
comparison between the criteria (Yılmaz, 2010).
1 Equal Importance
3 A little more important [𝐶𝑖𝑗 ]𝑛𝑥1 = [𝑎𝑖𝑗 ]𝑛𝑥𝑛 𝑥 [𝑤𝑖𝑗 ]𝑛𝑥1 (2)
5 Quite Important
[𝑑𝑖𝑗 ]𝑛𝑥1 = [𝐶𝑖𝑗 ]𝑛𝑥1 / [𝑤𝑖𝑗 ]𝑛𝑥1 (3)
7 Very Important
9 Highly Important

15 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21

∑𝑛𝑗=1 𝑑𝑖 (4) 𝜆: 𝐵𝑎se value

𝜆=
𝑛
𝑎𝑖𝑗 : Pairwise comparison matrix The basic value (λ) is calculated with the help of
𝑤𝑖𝑗 : 𝑊eight vector of criteria matrices in Equation 2 and Equation 3. Finally, the
randomness indicator (Rl) is selected from table 3
𝐶𝑖𝑗 : Column Vector
prepared by Saaty according to the number of criteria.
𝑑𝑖𝑗 : Consistency Vector
Table 3. Saaty’s Randomness Indicator
n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
RI 0.00 0.00 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49 1.51 1.48 1.56 1.57 1.59

𝑤𝑛 𝑤𝑛 𝑤𝑛+1 𝑤𝑛
With the randomness indicator (Rl) and λ (basic and 𝜑𝑛⁄(𝑛+1) = are × = must
𝑤𝑛+1 𝑤𝑛+1 𝑤𝑛+2 𝑤𝑛+2
value), the consistency ratio (CR) is calculated (Equation
satisfy the mathematical equation. If we combine the two
5).
𝜆−𝑛 equations, we get equation 9.
𝐶𝑅 = (5) 𝑤𝑛 (9)
(𝑛 − 1)𝑥𝑅𝐼 𝜑𝑛/𝑛+1) 𝑥 𝜑𝑛+1/𝑛+2) =
𝑤𝑛+2

2.3. Full consistency method (FUCOM) Full consistency is achieved if the conditions in
equation 8 and Equation 9 are met for criterion
The Full Consistency Method (FUCOM) method is weighting. Full consistency, consistency deviation (min
one of the criteria weighting methods developed by (DFC(X))) is expected to be minimal. Maximum
Pamučar, Stevic and Sremac in 2018. The FUCOM consistency is achieved if the deviation from full
algorithm is based on the pairwise comparisons of consistency is zero (0).
criteria, where only the n − 1 comparison in the model is Finally, using the expressions in equation 10 to find
necessary. The model implies the implementation of a the criterion weights, linear programming model and
simple algorithm with the ability to validate the model by solutions can be made with programs such as Excel
determining the deviation from full consistency (DFC) of Solver or MATLAB with simple codes.
the comparison. The consistency of the model is defined 𝑀𝑖𝑛 𝑋
on the basis of the satisfaction of mathematical
𝑤𝑗(𝑛)
transitivity conditions. One of the characteristics of the | − 𝜑𝑛/𝑛+1) | ≤ 𝑋, 𝐴𝑗
developed new method is the lowering of decision- 𝑤𝑗(𝑛+1)
makers’ subjectivity, which leads to consistency or 𝑤𝑗(𝑛)
symmetry in the weight values of the criteria (Pamučar | − 𝜑𝑛/𝑛+1) 𝑥𝜑(𝑛+1)/𝑛+2) | ≤ 𝑋, 𝐴𝑗 (10)
𝑤𝑗(𝑛+2)
et al., 2018).
The FUCOM method takes place in three stages. At 𝑤𝑗 > 0, 𝐴𝑗
the first stage, decision makers are asked to rank n ∑ 𝑤𝑗 = 1
criteria from the most important to the less important
𝑗
criteria (Equation 6).
𝐶𝑗(1) > 𝐶𝑗(2) = 𝐶𝑗(3) > ⋯ > 𝐶𝑗(𝑛) (6)
2.4. Determination of criteria and criterion maps
In the second stage, the comparative priorities of the
criteria ranked by the decision makers in order of Determining of the criteria is one of the most
importance (φ n/ (n+1) the comparative priority vector important phase of producing risk map. The criteria are
with n-1 elements is obtained (Equation 7). the decision components used in the evaluation of
(7) alternatives to reach the goals. It should be known that
𝜑 = {𝜑1/2 , 𝜑2/3 … 𝜑𝑛/𝑛+1) }
each criterion included in the decision problem is
effective in the decision process, as well as the criteria
In the FUCOM method, the decision maker(s) can not addressed in the problem have an indirect effect on
use integers, decimals or values of certain scales for the decision output (Yildirim, 2019).
comparisons of criteria. This provides flexibility to Slope, lithology, land use potential or vegetation,
decision makers in the evaluation of criteria. (Ayçin & slope direction, distance to main faults, drainage and
Aşan, 2021) relative height are the parameters often used in the risk
In the last stage, the following two conditions must studies (Gökçeoğlu & Ercanoğlu, 2001).
be met in order to calculate the criteria weights. In the current study; slope shape, slope, elevation,
Condition 1: The ratio of the weights of the two aspect, lithology, precipitation, proximity to the river,
criteria to each other should be equal to the priority value proximity to the road, NDVI (Normalized Difference
in the pairwise comparison (Equation 8). Vegetation Index), land use, soil type, fault line were used
𝑤𝑛 as the criteria which affect to the landslide. The raster
= 𝜑𝑛/𝑛+1) (8) data of each criterion were prepared by performing
𝑤𝑛+1
various spatial analyzes with the ArcGIS software.
a) Elevation: It has been reported that the height
Condition 2: The final values of the weight
conditions of the topography are also an effective factor
coefficients must satisfy the mathematical transitivity
in the formation of landslides. Because the determination
condition. Since φ n/ (n+1) × φ (n+1)/ (n+2) = φ n/ (n+2)
of the heights of the landslides occurring in any region
16 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21
can be accepted as a data that can only give a preliminary
idea. On the other hand, it has been suggested that
landslides tend to occur more in high altitude areas.
(Özşahin, 2015) The highest value of the region is 1956,
and the lowest value is 170. A total of five classes were
created in these value ranges.
b) Slope: The general tendency among researchers
is that as the slope increases, the sensitivity to landslides
will also increase (Gökçeoğlu & Ercanoğlu, 2001). The
slope in the region varies between 0-62°.
c) Slope shape: In the studies, the effect of the shape
of the slope on the landslide susceptibility was examined,
but some researchers said that more landslides occurred
on concave slopes, while some researchers suggested
that more landslides occurred on convex slopes.
In addition, statistical evaluation of this parameter
is quite difficult. Because during a landslide, the initial
appearance of the slope is often distorted and this may Figure 4. Aspect map
lead to erroneous assessments during data collection.
(Gökçeoğlu & Ercanoğlu, 2001). This study was carried
out by accepting the statement more landslides occur on
concave slopes.
d) Aspect: The slope direction (aspect) indicates the
direction of the land surface and is expressed by the
direction of the tangent plane at any point on the surface.
Slope direction is an important parameter that is
frequently used in studies related to the preparation of
landslide susceptibility maps (Dağ, 2007).
The map of these four criteria was obtained using
Digital Elevation Model (DEM) data in the '3D ANALYST
TOOLS' analysis. (Figure 2 -5).

Figure 5. Slope shape map

e) Proximity to the fault line: Proximity to the fault

line increases the risk of landslides. The landslide
analysis was carried out by considering the faults
remaining in the study area in the fault line map
published by MTA.
f) Proximity to the stream: Since being close to the
stream will increase the water saturation of the soil, the
risk of landslide increases as you get closer to the stream.
g) Proximity to the road: The roads opened on the
Figure 2. Elevation map
slopes cause a load reduction in both the topography and
the slope toe. The change in topography and the decrease
in load cause stress increases behind the slope and this
causes the development of stress cracks (Yalçın, 2007).
The maps of these three criteria were obtained by
using the multiple ring buffer analysis of the proximity
tool (Figure 6-8).

Figure 3. Slope map

17 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21
i) Land use: Although the land use situation
includes a part of the NDVI analysis such as forest,
meadow, swamp, residential area, agricultural area,
pasture, etc. It was used as a separate criterion as it
would affect the landslide in certain situations.
j) Soil type: The type of soil the ground has is also
very important for landslides. The soil mass covering the
ground of the topography also causes the formation of
landslides. In fact, soils affect landslide formation
according to grain size, arrangement and types (Özşahin,
2015).
The lithology map was taken from the earth sciences
site of MTA and the soil types map was taken from the
agriculture portal site, coordinated and digitized. The
land use map was obtained from the Copernicus page by
classifying the CORINE 2018 vector data. (Figure 9-11)

Figure 6. Distance to fault line

Figure 9. Lithology map

Figure 7. Distance to rivers

Figure 10. Soil type map

Figure 8. Distance to roads
h) Lithology: Lithology is one of the important
parameters affecting landslide formation and plays an
important role in landslide susceptibility studies. Using
the earth sciences website published by MTA, it was
determined that there are five different lithologies in the
region.

18 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21

Figure 13. Precipitation map

Figure 11. Land use map
3. Results and Discussion
k) NDVI: Landslide risk increases in areas with low
vegetation density. Therefore, the NDVI map was
The criteria weighting steps above were carried out
produced and the places with low vegetation were
sequentially and the criteria weights were calculated for
determined.
both methods. The criteria weights calculated using AHP
l) Precipitation: Annual average precipitation is
method and FUCOM method are shown in table 4.
considered as an important factor for landslide
According to the results, the most important weight
susceptibility analysis. Because, as a result of
calculated using AHP was lithology, and the least
precipitation, the ground becomes saturated with water,
important criterion was NDVI. The consistency
the groundwater level rises and the leakage forces reach
calculated in the AHP was found 0.02 and since it was less
their maximum value (Özşahin, 2015). The annual
than 0.1, the measurements were considered consistent.
precipitation of Taşova district is 967mm.
The most important criterion calculated using
The last two criteria maps were made as follows:
FUCOM was lithology, and the least important criterion
NDVI data was calculated with the help of band4 and
was NDVI. Since the FUCOM method is based on full
band5 in the Landsat satellite image (Equation 11).
consistency, the consistency deviation (DFC(X)) was
𝐵𝑎𝑛𝑑5 − 𝐵𝑎𝑛𝑑4 / 𝐵𝑎𝑛𝑑5 + 𝐵𝑎𝑛𝑑4 (11) found to be 0 as a result of the calculations and full
consistency was obtained in the measurements.
The precipitation map is produced at the end of the
The closer the consistency rate is to zero in the AHP
calculations made with the help of climate data. (Figure
method, the more consistent the measures are, while in
12, 13).
FUCOM, full consistency is essential. This shows that the
measures of FUCOM are more consistent.
In the current study, a total of 144 comparisons
were made with AHP, and 11 comparisons were made
with FUCOM. It was observed that the FUCOM method
differs from the AHP method with less pairwise
comparison. With fewer comparisons, the effect of expert
opinion is reduced.
Table 4. Criterion weights with AHP and FUCOM
Criteria AHP FUCOM
Lithology 0.204 0.2473
Slope 0.162 0.1236
Slope Shape 0.150 0.1236
Precipitation 0.125 0.0824
Aspect 0.093 0.0618
Prox. to Fault Line 0.072 0.0618
Prox. to the Stream 0.061 0.0618
Distance to Road 0.043 0.0618
Figure 12. NDVI map Land Use 0.032 0.0495
Soil Type 0.025 0.0495
Elevation 0.019 0.0495
NDVI 0.013 0.0275

19 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21
The raster data produced for each criterion and risky areas were different results in the two
separately were overlapped using 'Weighed Sum' methods.
analysis based on the weights calculated from two
Table 5. Map Classes Rates calculated with AHP and
different methods and two landslide susceptibility maps
FUCOM
were obtained
Classes AHP (%) FUCOM (%)
Figure 14 shows the map produced using the AHP Risk-Free Area 7.25 7.89
method and Figure 15 shows the map produced using the Low Risk Area 10.75 8.08
FUCOM method. The Landslide Susceptibility maps are Medium Risk Area 50.60 35.97
divided into five classes; Risk-free areas are dark green, Risk Area 27.95 42.98
low-risk areas are light green, medium-risk areas are High Risk Area 3.45 5.08
white, risky areas are pink, high-risk areas are red.
As a result of the calculations, according to AHP the
risky area was determined as 27.95%, and the high-risk
area was 3.45%. According to the FUCOM method, the
risky area was determined as 42.98% and the high-risk
area was 5.08%. The percentage of risky areas in the map
produced based on FUCOM method was higher than the
AHP method.
In general, it is seen that high-risk areas are in the
same places in both maps. These high-risk areas are seen
as areas where the slope is high and the vegetation is low.

4. Conclusion

In the current study, two different landslide

susceptibility maps were produced using AHP and
FUCOM criterion weighting for Taşova district.
Twelve criteria were used for both landslide
susceptibility maps. These criteria are slope shape, slope,
elevation, aspect, lithology, precipitation, proximity to
Figure 14. Landslide susceptibility map with AHP
stream, proximity to road, NDVI, land use, soil type, fault
line. While the most important criterion among the AHP
criteria weights was lithology with 0.204, the least
important criterion was NDVI with 0.013. The most
important criterion among the FUCOM criterion weights
was lithology with 0.247, while the least important
criterion was NDVI with 0.027. The consistency of the
criteria weights was calculated for both methods. The
consistency ratio with AHP was found to be 0.02, and it
was seen that the measurements were consistent.
Maps of each criterion were obtained according to
the studied area. Criterion maps were converted to raster
data according to weights calculated by AHP and FUCOM
methods. Landslide susceptibility maps were obtained
by combining the weighted criterion maps in both
methods.
The landslide susceptibility maps were divided into
five classes and risk classes were determined. The risk of
landslide increases from green to red. The ratios of the
classes were calculated by pixel measurements from the
Figure 15. Landslide susceptibility map with FUCOM maps. It was observed that risky areas in FUCOM were
15.03% higher than in AHP. It was observed that the
When we interpret by looking at the two maps in medium-risk areas in FUCOM were 14.63% less than in
Figure 14 and Figure 15, it is seen that the red colored the AHP. Other classes gave similar results in both
areas are more in FUCOM. methods.
The area of Landslide Susceptibility classes were Both methods are made by taking expert opinion,
calculated with the help of the pixels of the classes from but FUCOM minimizes expert opinion with less pairwise
the maps obtained. By making area calculations from comparison and provides full consistency. As a result, the
pixels according to colors, ratio calculations were made FUCOM method, which is the version developed in 2018
over the total area. of the AHP method, which is frequently used in the
Table 5 shows that; risk-free, low-risk and high-risk literature, can also be preferred and used in map
areas results were almost similar, but medium-risk areas production studies.

20 Advanced GIS
 Advanced GIS – 2023; 3(1); 14-21
Finally, landslide susceptibility maps can be landslide susceptibility maps. Earth Sciences,
prepared with various methods and criteria data. The 22(23), 189-206.
aim of this study is to compare the results which obtained Gökkaya, M. A. (2014). Coğrafi Bilgi Sistemleri (CBS) ve
AHP and FUCOM methods. Analitik Hiyerarşi Yöntemi (AHY) ile üretilen deprem
tehlike haritalarının duyarlılık analizi (Publication
Author Contributions No. 356077) [Master Thesis, Istanbul Technical
University]. YÖK National Thesis Center.
The contributions of the authors of this article is equal. Ishizaka, A., & Labib, A. (2009). Analytic hierarchy
process and expert choice: benefits and limitations.
Statement of Conflicts of Interest OR Insight, 22(4), 201-220.
Jian-Zhong, X., Li-Jing, W., & Jun, L. (2008). A Study of
There is no conflict of interest between the authors. AHP-Fuzzy comprehensive evaluation on the
development of eco-enterprise. 15th International
Statement of Research and Publication Ethics Conference On Management Science & Engineering,
Long Beach, USA. 219-224.
This study was improved from the paper presented in the https://doi.org/10.1109/ICMSE.2008.4668919
5th Intercontinental Geoinformation Days (IGD). Kabak, M., Erbaş, M., Çetinkaya, C., & Özceylan, E. (2018).
Research and publication ethics were complied with in A GIS-based MCDM approach for the evaluation of
the study. bike-share stations. Journal of Cleaner Production,
201, 49–60.
References https://doi.org/10.1016/j.jclepro.2018.08.033
Özşahin, E. (2015). Landslide Susceptibility Analysis by
Acar, E. (2019) Production of landslide susceptibility maps Geographical Information Systems: The Case of
by using AHP Method and GIS analyses (Publication Ganos Mount (Tekirdağ). Electronic Journal of Map
No. 588102 ) [Master Thesis, Hacettepe University]. Technologies 7(1), 47-63.
YÖK National Thesis Center. https://doi.org/10.15659/hartek.15.04.68
Arslankaya, S., & Göraltay, K. (2019) Çok kriterli karar Pamučar, D., Stević, Ž., & Sremac, S. (2018). A new model
verme yöntemlerinde güncel yaklaşımlar. Iksad for determining weight coefficients of criteria in
Publishing House. MCDM Models: full consistency method (FUCOM).
Ayçin, E., & Aşan, H. (2021). Determination of Symmetry, 10(9), 393.
the weight coefficients of https://doi.org/10.3390/sym10090393
criteria in the selection of business Saaty, T. L. (1994). How to make a decision: the analytic
intelligence applications with FUCO. hierarchy process. Interfaces, 24(6), 19-43.
KOCATEPE İİBFD, 23(2), 195-208. https://doi.org/10.1287/inte.24.6.19
https://doi.org/10.33707/akuiibfd.903563 Yalçın, A. (2007). The use of analytical hierarchy process
Boyacı, A. Ç. (2020) Selection of eco-friendly cities in and GIS in production of landslide susceptibility
Turkey via a hybrid hesitant fuzzy decision making maps. Selcuk University Journal of Engineering,
approach Applied Soft Computing, 89,106090. Science and Technology, 22(3), 1-14.
https://doi.org/10.1016/j.asoc.2020.106090 Yılmaz, M. (2010). The analytic hierarchy process (AHP)
Cankaya, Z. C., Suzen, M. L., Yalciner, A. C., Kolat, C., and an application: the selection of a library director
Zaytsev, A., & Aytore, B. (2016). A new GIS-based as a leader. Turkish Librarianship, 24(2), 206-234.
tsunami risk evaluation: MeTHuVA (METU tsunami Yildirim, B. (2019). Multi-Criteria Decision Making
human vulnerability assessment) at Yenikapı, Methods. Retrieved March 10, 2023, from
Istanbul. Earth, Planets and Space, 68, 1-22. https://bahadirfyildirim.com/kutuphane/kitap/co
https://doi.org/10.1186/s40623-016-0507-0 k-kriterli-karar-verme-yontemleri/
Dağ, S. (2007) Landslide susceptibility analysis of Çayeli Zahedi, F. (1986). The analytic hierarchy process: a
region (Rize) by statistical methods (Publication No. survey of the method and its applications. Interface
212109) [Doctoral Thesis, KTU]. YÖK National 16(4), 96-108.
Thesis Center. https://doi.org/10.1287/inte.16.4.96
Ergünay, O. (2007) Türkiye’nin Afet Profili, TMMOB Afet
Sempozyumu (in Turkish), Ankara, Turkey.
Gökçeoğlu, C., & Ercanoğlu, M. (2001). Uncertainties on
the parameters employed in preparation of

© Author(s) 2023.
This work is distributed under https://creativecommons.org/licenses/by-sa/4.0/

21 Advanced GIS