A fuzzy multi-criteria decision method for ecotourism development locating

[Research]

A fuzzy multi-criteria decision method for ecotourism development locating

A. Mahdavi¹*, M. Niknejad², O. Karami³

1. Dept. of Forest Science, Faculty of Agriculture and Natural Resources, Ilam University, Ilam, Iran

2. Dept. of Forest Science, Faculty of Agriculture and Natural Resources, Ilam University, Ilam, Iran.

3. Dept. of Forestry, Faculty of Natural Resources, University of Agriculture and Natural Resources of Sari, Sari, Iran.

* Corresponding author’s E-mail: a_amoli646@yahoo.com

ABSTRACT

The County of Khorram-Abad enjoys a high potential for ecotourism because of its mountains, forests, natural mineral springs, natural waterfalls and diversity in folks and cultures. But, un-planned and uncontrolled ecotourism can have negative effects on environment, economy, culture and even the security of eco-tourists. The main purpose of this study is to present a fuzzy multi-criteria decision making (FMCDM) method for ecotourism development location selection. In this study we created 5 main criteria and 14 sub-criteria for locating the suitable areas for ecotourism development from literature reviews and experts opinions. Delphi method was used to obtain the significant criteria and sub-criteria for ecotourism development by interviewing the foregoing experts and related managers. Then, the methods of fuzzy set theory, linguistic value, hierarchical structure analysis, and fuzzy analytic hierarchy process (FAHP) were applied to find the relative weights or importance degree of each criterion and rank the overall criteria as the measurable indices for ecotourism development. Different layers were prepared and were combined using weighted linear combination (WLC) method in GIS environment. The results showed that 6.57 and 38.65 percentages of the area have an excellent and good potential for the ecotourism development, respectively. In addition, the study confirm that the combination of FAHP with GIS could be a powerful combination to apply for different land use planning.

INTRODUCTION

Nature-based tourism in general is one of the fastest growing sectors within the global tourism industry (Buckley, 2000; Ryan et al., 2000). Reasons for this growth include demographic changes in source countries (such as older populations and, in turn, the growing number of more experienced travelers) and increasing environmental awareness on the part of the general public (Ayala, 1996). Ecotourism is viewed as a means of protecting natural areas through the generation of revenues, environmental education and the involvement of local people (in both decisions regarding appropriate developments and associated benefits) (Ross & Wall, 1999). Therefore, sustainable ecotourism is a type of tourism that produces economic advantages, in addition to maintaining environmental diversity and quality thus ‘combining conservation with economic development’ (Wild, 1994). Unsustainable ecotourism is the result of inappropriate developments taking place in sensitive locations. The environmental effects caused by overcrowding, overdevelopment, unregulated recreation, pollution, wildlife disturbances and vehicle uses are more serious for ecotourism than mass tourism (McNeely, 1989). Without appropriate regulations and planning, problems of overexploitation, and in particular ecological degradation, may be intensified with the development of ecotourism (Issacs, 2000; Kamauro, 1996; Mieczkowski, 1995). Thus in reality there is a need for suitable planning strategies to be formulated and implemented to ensure that the future expansion of ecotourism takes place in accordance with the principles of sustainable development (Wearing & Neil, 2009). Undoubtedly, only some areas suitable for ecotourism should be developed to maximize the positives impacts and minimize negative impacts on all aspects of ecotourism. In this respect, site suitability evaluation for ecotourism should be regarded as an important tool and a prerequisite for sustainable development of ecotourism. Site suitability evaluation can be judged with the help of criteria and indicators approach, which is basically a concept of sustainable ecotourism management developed in a set of principles, criteria and indicators (Bunruamkaew & Murayama, 2011).

Ecotourism should satisfy several criteria such as conservation of biological and cultural diversities through ecosystem protection and promotion of sustainable use of biodiversity with minimal impact on the environment being a primary concern (Bunruamkaew & Murayama, 2011). Anderson (1987) surveyed different methods for land capability/suitability analysis such as pass/fail screening, graduated screening, weighted factors, composite rating, and weighted composite rating and so on. A complex site selection process involves a measure of trade-offs among the criterion factors (Banai-kashani, 1989). The weighted factor method provides a procedure where each suitability factor is assigned a score, which is multiplied by the weight of that factor. The results of the multiplications are added, and thus a site composite score is determined. The composite score is compared with a predetermined standard, which is used to select or reject a site. This approach to site assessment is operational when standards are known. But for which no standards have been established or intangible criteria are used to assess alternatives the weighted-factors method is of limited use (Banai-kashani, 1989). A common feature of different the suitability methods (for more information referred to Anderson, 1987) is their reliance upon expert judgment. But, various sources of uncertainty, such as about the planning environment, about value judgments, and about the decisions of other participants, contribute to errors in decision making and forecasting by experts (uncertainties of the economic, demographic, and political environment) (Hall, 1980). Analytical Hierarchy Process (AHP) is an alternative to the methods used in suitability studies which can help the expert faced with decision making under uncertainty (Saaty & Vargas, 1987). However, AHP has been shown to be effective in evaluation problems involving multiple and diverse criteria and flexibility in dealing with both the qualitative (intangible) and quantitative (tangible) factors but has some shortcomings in the performance. In the conventional AHP, the pair wise comparisons for each level with respect to the goal of the best alternative selection are conducted using a nine-point scale. AHP is criticized for using lopsided judgmental scales and its inability to properly consider the inherent uncertainty and carelessness of pair comparisons (Shaverdi et al., 2013).

In order to overcome this kind of uncertainty in human preference, fuzzy sets theory could be incorporated with the pair-wise comparison as an extension of AHP. A variant of AHP, called Fuzzy AHP, comes into implementation in order to overcome the compensatory approach and the inability of the AHP in handling linguistic variables.

The fuzzy AHP approach allows a more accurate description of the decision making process (Vahidnia et al., 2009).

There are enormous challenges toward proper management of ecotourism in Khorram-Abad province.

The challenges reveal the importance of taking appropriate strategies to manage ecotourism in a sustainable manner in this region. We believe that sustainable ecotourism development efforts can be improved if priority areas for ecotourism and sustainable land uses are modified based on a comprehensive land suitability evaluation. In this regard, the study will use from the integration of GIS technology and fuzzy AHP method in locating the suitable sites for ecotourism development in the county of khorram-Abad.

MATERIAL AND METHODS

STUDY AREA

The county of Khorram-Abad as the capital for Khorram-Abad province is located in west of Iran. Its area is about 500000 hectares and is located between east longitude from 48°, 2´, 56" to 49°, 0´, 4" and north latitude from 33°, 53´, 42" to 33°, 53´, 27" (Fig. 1).

There are some important characteristics that make the area suitable for a successful ecotourism development program. For example, the county has an attractive mountainous forest landscapes, a rich vegetation cover and considerable wildlife, traditional indigenous people groups and folks and so on.

Such attributes suit the selection of the area for a case study to demonstrate the application of the methodology.

Fig 1. The location of study area in Khorram-Abad province and Iran.

DATA SOURCES

Data used in the study were assembled from a variety of sources. Firstly, the primary data from the field survey were collected through interviews and questionnaires answered by experts in the related fields for identifying factors and criteria that are important for ecotourism in Khorram-Abad Province along with statistics data, Global Positioning System (GPS) field survey data and other GIS datasets and maps.

METHODS

Basic concept of Fuzzy Analytical Hierarchy Process

The concept of fuzzy theory is introduced and addressed by Zadeh in 1965 for the first time

Fuzzy theory is composed of three key factors, which are fuzzy set, membership function, and fuzzy number to change vague data into useful data efficiently. The merit and strength of using fuzzy approach is to express the relative importance of the alternatives and the criteria with fuzzy numbers instead of using simple crisp numbers as most of the decision making problems in the real world takes place in a situation where the pertinent data and the sequences of possible actions are not precisely known. In this study the modified synthetic extent FAHP is utilized, which was originally introduced in Chang (1996).  A brief explosion of triangular fuzzy numbers and the FAHP method are given next.

Triangular fuzzy numbers (TENs)

Triangular fuzzy numbers are the most utilized in FAHP studies (Tang & Beynon, 2005). We define a fuzzy number M by a triplet (l, m, u) and membership function can be defined by Equation (1) (Chang, 1996):

  (1)

Two important operations used in this paper are illustrated. Define two TFNs M1 and M2 by the triplets

M1 = (l₁, m₁, u₁) and M₂ = (l₂, m₂, u₂).

Then: (1) Addition:

M₁ (+) M₂ = (l₁, m₁, u₁) (+) (l₂, m₂, u₂) = (l₁ + l₂, m₁ + m₂, u₁ + u₂),

(2) Multiplication:

M₁ * M₂ = (l₁, m₁, u₁) * (l₂, m₂, u₂) = (l₁ l₂, m₁m₂, u₁ u₂),

SET UP FUZZY PAIRED COMPARISON MATRICES

The central to the FAHP method is a series of pair-wise comparisons that indicating the relative preferences between pairs of criteria in the same hierarchy.

Using triangular fuzzy numbers with the pair-wise comparisons made, the fuzzy comparison matrix X = (x_ij)_n*m is constructed. The pair-wise comparisons are described by values taken from a pre-defined set of ratio scale values as presented in Table 1 and Fig. 2.

The ratio comparison between the relative preference of elements indexed i and j on a criterion can be modeled through a fuzzy scale value associated with a degree of fuzziness. Then an element of X, x_ij (i.e., a comparison of the ith decision alternative (DA) with the jth DA) is a fuzzy number defined as xij (l_ij, m_ij, u_ij) where, mij, lij, and uij are the modal, lower bound, and upper bound values for xij  respectively.

Let C = {C₁, C_2… C_n} be a criteria set, where n is the number of criteria and A ={A₁, A₂… Am} is a DA set with m the number of DAs. Let,,… be values of extent analysis of the ith criteria for m DAs. Here i = 1, 2… n and all the  (j = 1, 2… m) are triangular fuzzy numbers (TENs). The value of fuzzy synthetic extent _si with respect to the _ith criteria is defined as:

     (2)

Table 1. Linguistic variables describing weights of criteria and values of ratings.

Fig 2. Linguistic Variables for the Importance Weight of Each Criterion.

Where superscript -1 represents the fuzzy inverse. For more information about the concepts of synthetic extent, refer to Chang (1996).

CALCULATION OF THE SETS OF WEIGHT VALUES OF THE FAHP

To obtain the estimates for the sets of weight values under each criterion, it is necessary

to consider a principle of comparison for fuzzy numbers (Chang, 1996). For example, for

two fuzzy numbers M₁ and M₂, the degree of possibility of M₁ ≥ M₂ is defined as:

V(M₁ ≥ M₂) = sup _{x ≥ y} [min (μ M ₁ (x) , μ M ₂ (y)],                                                                     (3)

where sup represents supremum (i.e., the least upper bound of a set) and when a pair (x, y) exists such that x ≥ y and (μ_{M 1} (x) = μ_{M 2} (y) =1, it follows that V(M₁ ≥ M₂) =1 and V(M₂ ≥ M₁) =0. Since M₁ and M₂ are convex fuzzy numbers defined by the TFNs (l₁, m₁, u₁) and (l₂, m₂, u₂) respectively, it follows that:

V(M₁ ≥ M₂) = 1 iff m₁ ≥ m₂;

V(M₂ ≥ M₁) = hgt (M₁ ∩ M₂) = μ M₁  (_xd ),    (4)

where iff represents “if and only if” and d is the ordinate of the highest intersection point between the μ _M1 and μ _M2 TFNs (see Fig. 3) nd xd is the point on the domain of μ _M1 and μ _M2 where the ordinate d is found. The term hgt is the height of fuzzy numbers on the intersection of M₁ and M₂. For M₁ = (l₁, m₁, u₁) and M₂ = (l₂, m₂, u₂), the possible ordinate of their intersection is given by Equation (4). The degree of possibility for a convex fuzzy number can be obtained from the use of Equation (5)

  (5)

The degree of possibility for a convex fuzzy number M to be greater than the number of k convex fuzzy numbers M_i (i = 1, 2,…, k) can be given by the use of the operations max and min (Dubois and Prade, 1980) and can be defined by:

V (M ≥ M₁, M₂,……M_k) = V[(M ≥ M₁) and (M ≥ M₂) and…… (M ≥ M_k)]

Assume that d′(Ai) = min V(S_i ≥ S_k), where k = 1, 2, …, n, k ≠ i, and n is the number of criteria as described previously. Then a weight vector is given by:

W'= (d' (A₁), d' (A₂),…, d' (A_m)),

where A_i (i = 1, 2, …, m) are the m DAs. Hence

each d′(A_i) value represents the relative preference of each DA. To allow the values in the vector to be analogous to weights defined from the AHP type methods, the vector W′ is normalized and denoted:

W= (d (A₁), d (A₂),…, d (A_m)).

Fig 3. The comparison of two fuzzy number M₁ and M₂

CONSISTENCY TEST

The important thing about the pair -wise comparison matrixes is their incompatibility. According to consideration Professor Saaty (1980) for stability arbitrations is necessary that rate of their incompatibility matrixes be less or equal to 0.1. Otherwise, the respective expert is required to repeat itself adjudication as a stable matrixes (Amiri et al., 2008). The Consistency index (CI) is performed as follows:

Where λmax is the maximum eigenvalue, and n is the dimension of matrix. The consistency ratio (CR) was introduced to aid the decision on revising the matrix or not. It is defined as the ratio of the CI to the so-called random index (RI), which is a CI of randomly generated matrices:

Determination and weighting of effective criteria and sub-criteria for ecotourism development using FAHP

Identifying of criteria and sub-criteria

This study selected 5 main criteria and 14 sub-criteria in the form of GIS-based layers in determining what areas are best suited for ecotourism development. In order to identify the effective criteria and sub-criteria for ecotourism development in the study area, firstly based on literature review and previous

studies (Bunruamkaew & Murayama, 2011; Lawal et al., 2011; Anane et al., 2012), special conditions of the region and expert’s opinions, 5 main criteria and 14 sub criteria were selected. The selected criteria and sub criteria have been shown in Table 2.

Delphi method and estimating the relative weights of criteria and sub-criteria

Delphi method mostly aims at easy common understanding of group decisions through twice provision of questionnaires (Hsu et al., 2010). This study also conducted a Delphi method based on FAHP questionnaire survey with 10 expert scholars specializing in the field ecotourism and government tourism offices for weighting of criteria and sub-criteria. We sent 15 provided questionnaires to the experts that 10 from them were acceptable. In addition, for some cases that were requested for more information, we conducted the face to face interview with experts based on provided questionnaires.

Weighting to criteria and sub criteria were performed based on pair-wise comparison technique and fuzzy values taken from a pre-defined set of ratio scale values as presented in table 1 and Fig. 2. Questionnaires properly evaluated and the criteria weighted in the Matlab 2009 software. After normalized weight of each criterion, the aggregation of ten experts’ opinions for the five main criteria and 14 sub-criteria was performed using the geometric mean approach (Kabir & Sumi, 2013).

Table 2. Hierarchical structure, Criteria and sub-criteria in land suitability analysis for ecotourism.

PROVIDING THE MAPS

For mapping the suitable areas for ecotourism development in the study area, firstly, the respective layers to selected criteria should be prepared. For this regard, some maps (topography, soil, geology and vegetation) as hardcopy were provided from related offices. All these maps were digitized and classified using Arc GIS 9.3 software in GIS environment. After providing a digital elevation model (DEM) from topography map, different layers such as slop, aspect and elevation were extracted. The layers for other used criteria in this study like distances from recreational tourist attractions, negative factors, roods, water sources and settlements were created in GIS environment after providing some maps and field visiting and recording their location with GPS. To create Isohyetal map and Isotherms map for the study area, after providing related meteorological information, we used Inverse Distance Weighted interpolation method in GIS environment. In the next step, to be

comparable all the created map layers in terms of units and scales, the standardization of maps were performed. For this regard, the pixel values of all sub-criteria raster layers were transformed on a scale suitability ranging from 0 (least suitable) to 255 (most suitable) using fuzzy membership functions

extension in IDRISI software. However each sub-criteria value is processed differently depending on their continuous or discrete form or the defined suitability classes in Table 2.

Extracting the final composition map of potential area for ecotourism

After creating different layers and determination of their final weights by FAHP, the layers were integrated with their assigned weights using Weighted Linear Combination technique in GIS environment (Sante-Riveira et al., 2008). This technique can be done with by calculating the composite decision value (Rij) for each pixel (ij) as follows:

Rij= ∑ w_k r_ijk Where, W_k is the assigned weight for sub-criteria k and r_ijk is the standardized value of pixel (i,j) in the map of sub-criterion k. r_ijk varies between 0 and 255 where 0 is the least suitable value and 255 is the most suitable value. (Anane et al., 2012).

RESULTS AND DISCUSSION

The Fuzzy Analytical Hierarchy Process (FAHP)

The results of the weighting criteria based on FAHP method and analysis was performed using MATLAB software is shown Table 3. These weights are obtained based on Delphi method and mathematical relations in FAHP. Inconsistency ratio (CR) calculated less than 0.1 that is indicating an acceptable level of pair wise comparisons in the FAHP matrix. According to this method in the study area as it shows in Table 3, distance from water resources (with final weight of 0.205), distance from the access roods (with final weight of 0.117), and vegetation type and density (with final weight of 0.114) are the most effective criteria in evaluation capability of ecotourism in the Khorram-Abad county, respectively.

Criteria layers creation and their classification

The related criteria and sub-criteria as seen in Table 2 were created and kept as GIS layers (Fig. 4 to 17).

The layers were classified based on Table 2 and fuzzy concept theory, as the biggest fuzzy number value was assigned for the most suitable class. For instance, between slop classes, the class that has the least slop the biggest value was assigned.

Extract the most suitable areas based on their composite decision value

From the suitability map for ecotourism as seen in Fig.18, it was found that the total area of excellent and good suitable areas (C₁ and C₂) for ecotourism development is about 45.22% and these are located mostly in the eastern part of the county.

The area of moderately suitable (C₃) is about 48.44% and these are in the central, northern and southern parts of the county. Only a few percentages (4.54% and 1.8%) of the area were classified as weak and not suitable (C₄, C₅) respectively (Table 4).

Table 3. Criteria, sub-criteria and their final layer weight.

Fig. 4. Temperature map.                                      Fig. 5. Precipitation map.

Fig. 6. Aspects classes map.                               Fig. 7. Slop classes map.

Fig. 8. Soil types map.                            Fig. 9. Elevation classes map.

                       Fig. 10. Erosion intensity classes map.                Fig. 11. Petrology map.

Fig. 12. Distance from water sources map.              Fig. 13. Vegetation classes map.

               Fig. 14. Distance from settlements map                 Fig. 15. Distance from roods map   

       Fig. 16. Distance from recreational attractions        Fig. 17. Distance from negative factors map   

Extract the most suitable areas based on their composite decision value

From the suitability map for ecotourism as seen in Fig.18, it was found that the total area  of excellent and good suitable areas (C₁ and C₂) for ecotourism development is about 45.22% and these are located mostly in the

eastern part of the county. The area of moderately suitable (C₃) is about 48.44% and these are in the central, northern and southern parts of the county. Only a few percentages (4.54% and 1.8%) of the area were classified as weak and not suitable (C₄, C₅) respectively (Table 4).

   Fig. 18. Final map of suitable areas for ecotourism development in the region.

Table 4. The area and percentages of different suitable classes for ecotourism development.