Optimizing Tourist Destination Selection Using AHP and Fuzzy AHP Based on Individual Preferences for Personalized Tourism

Abstract

Tourism is a dynamic industry that significantly contributes to the global economy, driven by the increasingly diverse preferences of tourists. Addressing these preferences requires sophisticated decision-making models capable of handling the uncertainty and subjectivity of human judgments. This study proposes sustainable models for effectively capturing and evaluating individual tourist preferences using the Analytic Hierarchy Process (AHP) and the Fuzzy Analytic Hierarchy Process (Fuzzy AHP). These models leverage the strengths of the AHP to construct a flexible decision-making framework that adapts to diverse tourist preferences, offering personalized recommendations. In this study, three main criteria are considered: types of tourism, tourism facilities, and tourism areas. Tourists are encouraged to provide their preferences for these criteria and sub-criteria, enabling the AHP and Fuzzy AHP to recommend suitable destinations. An analysis was conducted with 30 respondents providing pairwise comparisons of the tourism criteria, which were then used to generate tourist attraction recommendations using both the AHP and Fuzzy AHP. The study assessed respondents’ satisfaction with the recommendations, finding that both methods were effective, with a slight preference for the Fuzzy AHP due to its ability to better capture individual preferences. The results underscore the potential of these models in sustainably enhancing decision support systems in the tourism industry, offering tailored recommendations that align more closely with tourist expectations.

1. Introduction

Tourism is an industry that plays a crucial role in the global economy. The increasing diversity of tourist preferences and the complexity of tourism destinations have necessitated the development of sophisticated decision-making models to enhance the quality of services offered [1]. To address the diverse needs and preferences of tourists, it is essential to employ robust methodologies that can handle the inherent uncertainty and subjectivity associated with human expectations.

The AHP is a well-established MCDM technique that helps in structuring complex decision problems into a hierarchical model, allowing decision-makers to systematically evaluate multiple criteria [2]. However, the traditional AHP may fall short in addressing the ambiguity and vagueness present in human judgments. To overcome this limitation, the Fuzzy AHP extends the conventional AHP by incorporating fuzzy logic principles, which are adept at handling imprecise and subjective information. The Fuzzy AHP enhances the decision-making process by enabling a more accurate representation of the decision makers’ preferences, thus it can provide a more reliable and comprehensive evaluation.

In the context of tourism, individual preferences vary widely based on numerous factors such as cultural background, personal interests, and past experiences [3]. In the past, tourists may have failed to select the right destinations due to the overwhelming number of options available and the lack of a structured approach to evaluate these choices. This diversity of preferences often leads to confusion and dissatisfaction when planning trips, as travelers might struggle to identify destinations that truly align with their unique desires and values. To address this challenge, implementing advanced decision-making frameworks can help tourists effectively navigate their options and select destinations that best meet their specific preferences and expectations. Therefore, in this study, we propose multi-criteria decision-making (MCDM) models, the Analytic Hierarchy Process (AHP) and Fuzzy Analytic Hierarchy Process (Fuzzy AHP), to effectively capture and evaluate individual tourists’ preferences.

2. Literature Reviews

The selection of attractive tourist destinations is a complex decision-making process influenced by individual preferences, which can vary widely due to factors such as cultural background, personal interests, and past experiences. The AHP has been widely used for destination selection; however, these methods often assume a level of certainty and consistency in judgments that may not reflect the subjective and uncertain nature of individual preferences in tourism. To address this challenge, this paper explores the application of both the AHP and Fuzzy AHP in designing personalized tourist trips. By incorporating fuzzy logic into the AHP framework, we aim to better capture the inherent uncertainty and subjectivity in tourists’ preferences, leading to a more accurate and tailored selection of attractive places. This approach seeks to enhance the decision-making process by providing a more nuanced understanding of individual preferences, ultimately improving the overall tourist experience.

The AHP is a decision-making technique that can be effectively applied in selecting attractive places in tourist trip design. It is a structured decision-making technique developed by Thomas L. Saaty in the 1980s [4] and widely used in various fields, including business, engineering, and management, to help individuals and groups make complex decisions involving multiple criteria and alternatives [1,5,6].

Canco et al. (2021) discusses the AHP as an effective method for making quality business decisions [7]. The AHP method aids in the identification of decision-making criteria based on the perceptions of managers and consumers and is structured into three levels, allowing for the assessment and prioritization of goals. Yang and Lee (1997) presents an AHP model that helps organizations in making facility location decisions. This decision-making framework allows managers to evaluate potential sites by considering both qualitative and quantitative factors, facilitating the incorporation of managerial experience and judgment [8]. The AHP model is used to match decision-makers’ preferences with the characteristics of various site alternatives, based on structured, hierarchical analysis and quantification of subjective judgments. The model is illustrated with an example problem, demonstrating its practical application.

Tam and Tummala (2001) discuss the importance of vendor selection in the telecommunications industry, particularly for telecom companies that view these vendors as long-term investments [9]. The selection process is complex, involving multiple decision-makers and criteria, including technical specifications, cost, reliability, and vendor reputation. The paper emphasizes the need for a systematic decision-making process and proposes a model based on the AHP to improve group decisions in vendor selection. This model aims to balance various objectives and criteria, resulting in a more efficient and effective vendor choice that meets customer specifications and strategic goals.

The AHP is also used in road selection. Han et al. (2020) discusses a method for selecting roads using the AHP for road selection, which comprises the four following steps [10]: using points of interest to build contextual characteristic indicators for roads; forming an AHP model with topological, geometrical, and contextual indicators to determine each road’s importance; selecting roads based on their importance and adaptive thresholds for their density partitions; and ensuring the global connectivity of the road network is maintained. The article claims that this method preserves the original network structure better than other methods and aligns well with manual selection results. Road selection is a crucial operation in the generalization of geographic information in cartography, and it aims to simplify road networks while preserving essential patterns and connectivity.

The AHP has been utilized in various tourism contexts to address complex decision-making challenges. Susano et al. (2019) introduced a decision-making system designed to identify the best tourism destination [11]. This decision support system utilized the AHP method to assist the community in selecting tourist attractions. The study concluded that the system produces a ranking of tourism destinations, derived from the calculation of priority weights and the evaluation of each destination’s attributes. Blešic et al. (2021) explore the application of the AHP in understanding the factors influencing travel behavior, destination selection, and tourist expectations [12]. The study highlights how the AHP can support tourism decision-making by analyzing various factors, such as destination choice, travel motives, hotel location preferences, and tourist indicators, while evaluating the significance of each factor in the decision-making process. Božić et al. (2018) propose strategies for diversifying tourism products on Phuket Island, Thailand using the AHP method [13]. They ranked the attractiveness of six cultural heritage sites to recommend those best suited for inclusion in cultural tourism development. The study also employed a quantitative–qualitative evaluation framework with weighted criteria based on input from local experts. The findings highlight which cultural sites hold the greatest appeal for tourists and demonstrate the effectiveness of the AHP method, combined with quantitative–qualitative evaluation, in supporting decision-making for tourism destination development.

The Fuzzy AHP is an extension of the traditional AHP. The AHP relies on pairwise comparisons and a hierarchical structure to assess the relative importance of various criteria in decision-making. However, the AHP’s use of crisp values can sometimes be limiting, as it does not account for the uncertainty and vagueness inherent in human judgment [14]. The Fuzzy AHP addresses this limitation by integrating fuzzy logic, which allows for more nuanced and flexible evaluations. The incorporation of fuzzy logic makes the Fuzzy AHP particularly suitable for complex decision-making problems where linguistic variables and imprecise data are prevalent [8,15,16].

The core concept of the Fuzzy AHP involves using fuzzy numbers, often triangular or trapezoidal, instead of precise numerical values for pairwise comparisons. This approach captures the uncertainty and ambiguity in the decision-making process, providing a more realistic representation of the decision-maker’s preferences [17]. The fuzzy pairwise comparison matrices generated in the Fuzzy AHP can be defuzzified using various techniques to obtain a crisp value, which is then used to calculate the weights of the criteria and alternatives [18]. This process allows decision-makers to express their judgments using linguistic terms such as “slightly more important”, “moderately more important”, or “significantly more important”, which are then translated into fuzzy numbers [19].

The Fuzzy AHP has been widespread across various fields, including supply chain management [15,20,21], environmental management [22,23], project selection [24], and risk assessment [25,26]. In supply chain management, for instance, the Fuzzy AHP has been used to evaluate and select suppliers based on criteria such as cost, quality, delivery time, and flexibility. In environmental management, it has been applied to assess the sustainability of different energy sources, considering factors like environmental impact, economic feasibility, and social acceptability. The method’s ability to handle the complexity and vagueness of real-world problems makes it a valuable tool in these domains.

The Fuzzy AHP has been increasingly applied in the tourism industry to address complex decision-making problems that involve uncertainty and subjective judgments [27]. By incorporating fuzzy logic into the AHP framework, the Fuzzy AHP allows decision makers to use linguistic terms (e.g., “very important”, “moderately important”) instead of precise numerical values, providing a more flexible and accurate reflection of their preferences [28]. This is particularly useful in tourism, where decisions often depend on qualitative factors such as customer satisfaction, cultural significance, and environmental impact.

Göksu and Kaya (2014) employed the AHP and TOPSIS methods in combination with fuzzy logic to address the inherent ambiguity in tourism and tourist decision-making [29]. The collected data were evaluated using the Fuzzy Analytic Hierarchy Process (Fuzzy AHP). Recognizing that each tourist has a unique perspective on selecting a destination, the study considered various factors, including convenient transportation, cost, historical and cultural beliefs and doctrines, natural beauty, and entertainment options. Bire et al. (2021) aimed to develop a decision support system for selecting tourist attractions in Kupang City using the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) method [30]. The system allows users to input the priority scale for nine human need attributes and then provides recommendations for tourist attractions based on these inputs. The study also compares the Fuzzy AHP method with the traditional AHP calculations, finding that while both methods are effective for multicriteria decision-making, the Fuzzy AHP method offers a more optimal solution in scenarios where there is uncertainty in the comparisons between elements.

Liao et al. (2023) recently analyzed the applications of fuzzy multi-criteria decision-making (MCDM) methods in the hospitality and tourism industries, while also exploring potential future research directions [31]. Their analysis of bibliometric data, methodologies, and applications revealed that the AHP and TOPSIS methods are the most used MCDM techniques. The study also identified tourism evaluation, hotel evaluation and selection, and tourism destination evaluation and selection as the most prominent research topics within the hospitality and tourism industries.

Despite the widespread application of the AHP and its extension, the Fuzzy AHP, in various decision-making contexts, there is a notable research gap regarding their use specifically for capturing individual tourist preferences. Most existing literature has primarily focused on AHP’s general decision-making frameworks or site selection criteria, overlooking the unique nuances of tourist preferences influenced by factors such as personal interests and past experiences. Furthermore, while the Fuzzy AHP has been recognized for its capability to address uncertainty and subjectivity in judgments, empirical research demonstrating its application in the tourism sector remains limited. This gap is crucial, as understanding tourists’ preferences is essential for tailoring travel experiences and enhancing satisfaction. Therefore, the lack of studies that utilize the AHP and Fuzzy AHP to systematically evaluate and analyze tourist preferences presents an opportunity for the research to provide a more nuanced understanding of how various factors influence tourist decision-making, ultimately leading to more personalized travel recommendations.

The choice to utilize the AHP and Fuzzy AHP for personalized tourism recommendation systems is justified by their structured, transparent, and adaptable nature. The AHP provides a systematic approach for breaking down complex decision-making processes into hierarchical levels, allowing stakeholders to evaluate various criteria and sub-criteria effectively. This is particularly important in tourism, where individual preferences can vary widely based on personal factors. The Fuzzy AHP further enhances this process by incorporating fuzzy logic, which accommodates the inherent uncertainty and ambiguity often present in human judgment. In the context of tourism, where preferences are subjective and can fluctuate based on context, the Fuzzy AHP allows for a more flexible evaluation of criteria, enabling a better capture of nuances in tourist preferences. This adaptability leads to more tailored recommendations that resonate with individual travelers, ultimately enhancing their experiences. In comparison to the TOPSIS approach, the AHP and Fuzzy AHP offer distinct advantages. While TOPSIS provides a straightforward ranking mechanism based on proximity to an ideal solution, it may not fully account for the complexity of individual judgments and the interdependence of criteria. Furthermore, TOPSIS typically relies on crisp data, which may overlook the vagueness and subjectivity inherent in preferences.

3. Methodology

3.1. AHP Model

The AHP is a structured technique used for organizing and analyzing complex decisions. Developed by Thomas L. Saaty in the 1970s. The AHP helps decision-makers to prioritize alternatives and make the best decision based on multiple criteria.

Figure 1 illustrates the AHP key steps. The process begins by defining the problem and structuring it into a hierarchy, which typically includes the overall goal at the top, followed by criteria and sub-criteria, and finally, the alternatives at the bottom (Figure 2). Once the hierarchy is established, the next step involves performing pairwise comparisons of the elements at each level of the hierarchy, based on their relative importance or preference in relation to an element at the higher level. This is conducted using a scale of 1 to 9, shown in

Table 1. Saaty’s scale for pairwise comparisons in the AHP.

Judgment Term (A over B)	Numerical Term
Absolute preference	9
Very strong preference	7
Strong preference	5
Weak preference	3
Indifference of A and B	1
Intermediate values	2, 4, 6, 8
Weak preference	1/3
Strong preference	1/5
Very strong preference	1/7
Absolute preference	1/9

, called Saaty’s scale for pairwise comparisons in the AHP.

After completing the pairwise comparisons, the data are used to construct comparison matrices for each level of the hierarchy. These matrices are then normalized, and the priority weights of the criteria and alternatives are calculated. This involves computing the eigenvectors of the matrices, which represent the relative weights of the elements. A consistency check is conducted to ensure that the judgments made during the pairwise comparisons are consistent. The consistency ratio (CR) is calculated, and a CR of 0.1 or less is generally considered acceptable.

To determine the weights of n elements, a pairwise comparison is conducted between elements i and j using Saaty’s scale for pairwise comparisons, as shown in Table 1. The results of these comparisons are then organized into a matrix A, where each element a_ij represents the relative importance of element i compared to element j.

$$A=\begin{array}{cccc}{a}_{11}& a_{12}& ...& a_{1n}\\ a_{21}& a_{21}& ...& a_{2n}\\ ..& ..& ..& ..\\ a_{n1}& a_{n2}& ..& a_{nn}\end{array}$$

After pairwise comparison matrices are formed, the weights are computed in two steps. First, the pairwise comparison matrix, A = [a_ij]_nxn, is normalized by Equation (1), and then the weights, w_i = [w₁, w₂, …, w_n], are computed by Equation (2).

$$\begin{array}{cc}{a}_{ij}^{\mathrm{*}}=\frac{a_{ij}}{{\sum }_{i=1}^n{a}_{ij}}& \hspace{1em}\hspace{1em}for all j=1, 2, \dots , n\end{array}$$

(1)

$$\begin{array}{cc}{w}_i=\frac{{\sum }_{j=1}^n{a}_{ij}^{\mathrm{*}}}{n}& \hspace{1em}\hspace{1em}for all i=1, 2, \dots , n\end{array}$$

(2)

$${\lambda }_{max}=\frac{1}{n}{\sum }_{i=1}^n\left\{\frac{{\sum }_{j=1}^n{a}_{ij·w_j}}{w_i}\right\}$$

(3)

Equation (3) determines λ_max, a reference index to assess the consistency of the pairwise comparisons. λ_max is the maximum eigenvalue of the matrix of comparison. The closer λ_max is to the number n, the smaller the inconsistency will be. At the end the consistency ratio (CR) can be calculated from the ratio of the consistency index (CI) and the random index (RI), where RI refers to the number in

Table 2. Saaty’s ratio index for different values of n.

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

.

$$CI=\frac{{\lambda }_{max}}{n-1}$$

(4)

$$CR=\frac{CI}{RI}$$

(5)

If the CR is below 0.10, the results are considered sufficiently accurate. However, if the CR exceeds 0.10, the results should be re-evaluated to identify the sources of inconsistency. This may include adjusting by partially repeating the pairwise comparisons. If multiple rounds of this process do not decrease the consistency to an acceptable level of 0.10, all results should be disregarded, and the entire procedure must be started anew [32].

The final step of the AHP is to synthesize the results by combining the priority weights of the criteria and alternatives to determine the overall ranking of the alternatives. This synthesis provides a comprehensive evaluation, allowing decision-makers to select the best alternative based on a structured and quantitative analysis of the criteria and sub-criteria. The AHP helps in breaking down complex decisions into manageable parts, providing a clear rationale for each decision and ensuring a balanced consideration of both qualitative and quantitative factors.

3.2. Fuzzy AHP Model

Fuzzy incorporates fuzzy logic to handle the uncertainty and vagueness often present in the decision-making process. The traditional AHP uses precise numerical values for pairwise comparisons, while the Fuzzy AHP uses fuzzy numbers, typically triangular fuzzy numbers, to represent the uncertainty in the judgments.

In Figure 3, the step of the Fuzzy AHP begins with defining the problem and structuring the hierarchy the same as the AHP process. Decision-makers then use fuzzy numbers to make pairwise comparisons between elements, capturing their subjective judgments more accurately. These fuzzy comparison matrices are normalized and used to compute fuzzy synthetic extents, which are subsequently defuzzied to obtain crisp priority weights. Finally, the priority weights are aggregated to rank the alternatives [33].

Table 3. Definition and membership function of a fuzzy number.

Judgment Term (Linguistic Term) (Element A over Element B)	Scrip Number	Fuzzy Number	Membership Function
Absolute preference	9	$\tilde{9}$	(8, 9, 10)
Very strong preference	7	$\tilde{7}$	(6, 7, 8)
Strong preference	5	$\tilde{5}$	(4, 5, 6)
Weak preference	3	$\tilde{3}$	(2, 3, 4)
Indifference of A and B	1	$\tilde{1}$	(1, 1, 2)

displays the triangular fuzzy number, $\tilde{1}$ to $\tilde{9}$. This fuzzy number represents subjective pairwise comparisons of selection process in order to capture the vagueness [34]. A fuzzy number is special fuzzy set $F=\left\{\right(x,{\mu }_F\left(x\right)),x\in R\}$, where $x$ takes it values on the real number, $R:-\infty <x<+\infty$ and ${\mu }_F\left(x\right)$ is a continue mapping from $R$ to the close interval [0, 1]. Let $M\in F\left(R\right)$ be called a fuzzy number $M$. The triangular membership should satisfy Definition 1.

Definition 1.

A fuzzy number is defined to be a triangular fuzzy number if its membership function is equal to:

$${\mu }_M(x)=\left\{\begin{array}{c}\frac{x}{m-l}-\frac{l}{m-l},x\in [l,m],\\ \frac{x}{m-u}-\frac{u}{m-u},x\in [m,u],\\ 0, otherwise\end{array}\right\}$$

(6)

where $l\le m\le u$, l and u stand for the lower and upper value of the support of M, respectively, and m for the modal value. The triangular fuzzy number can be denoted by (l, m, u). The support of M is the set of elements $\{x\in R|l<m<u\}$, when $l=m=u$ is a nonfuzzy number by convention.

The triangular fuzzy number, $\tilde{1}$ to $\tilde{9}$, are employed to enhance the traditional nine-point scaling method. To take the impression of human qualitative assessments into consideration, the five triangular fuzzy numbers are defined with the corresponding membership function as shown in Figure 4.

After defining the problem and structuring the hierarchy in the same way as the AHP process in Step 1, the analysis process is then continued to construct pairwise comparison matrices in Step 2. It is to decide on the relative importance of each pair of factors in the same hierarchy. By using triangular fuzzy numbers, via pairwise comparison, the fuzzy evaluation matrix $A=(a_{i,j}{)}_{n\times m}$ is constructed. $a_{i,j}$ is the relative importance of element i over element j under a certain criterion, then $a_{i,j}=(l,m,u)$, where l and u represent a fuzzy degree of judgment. If the relative importance of element j over element i holds, then the pairwise comparison scale can be represented by the fuzzy number $a_{i,j}^{-1}=(\frac{1}{u},\frac{1}{m},\frac{1}{l})$.

In the Fuzzy AHP, Step 3 considers the value of fuzzy synthetic extent. Let $X=\{x_1,x_2,...,x_n\}$ be an object set, and $U=\{u_1,u_2,...,u_n\}$ be a goal set. According to the method of extent analysis, we now use the fuzzy geometric mean [17] to aggregate these comparisons to derive priority weights. This process involves calculating the fuzzy geometric mean for each criterion or alternative in the pairwise comparison matrices. The aggregated fuzzy values are then normalized and defuzzified to determine the final priority weights.

Definition 2.

Consider two triangular fuzzy numbers $M_1$ and $M_2$, $M_1=(l_1,m_1,u_1)$ and $M_2=(l_2,m_2,u_2)$. Their operational laws are as follows:

$$(l_1,m_1,u_1)\oplus (l_2,m_2,u_2)=(l_1+l_2,m_1+m_2,u_1+u_2)$$

(7)

$$(l_1,m_1,u_1)\odot (l_2,m_2,u_2)\approx (l_1\cdot l_2,m_1\cdot m_2,u_1\cdot u_2)$$

(8)

$$(\lambda ,\lambda ,\lambda )\odot (l_1,m_1,u_1)\approx (\lambda l_1,\lambda m_1,\lambda u_1), \hspace{1em}\lambda >0,\lambda \in R$$

(9)

$$(l_1,m_1,u_1{)}^{-1}\approx (\frac{1}{u_1},\frac{1}{m_1},\frac{1}{l_1})$$

(10)

Definition 3.

The fuzzy geometric mean and fuzzy weights in a pairwise comparison matrix of alternatives, $A$, are calculated by the following formula:

$${\tilde{r}}_i={\prod }_{j=1}^n(a_{i,j}{)}^{\frac{1}{n}}$$

(11)

where ${\tilde{r}}_i$ is the fuzzy geometric mean value of alternative i, $i=\mathrm{1,2},...,n$, $n$ is the number of alternatives. Then, the fuzzy weight ${\tilde{w}}_i$ is calculated as follows:

$${\tilde{w}}_i={\tilde{r}}_i\odot {\left[{\sum }_{i=1}^n{\tilde{r}}_i\right]}^{-1}$$

(12)

The next step of the Fuzzy AHP is defuzzification. The center of area (COA) method, also known as the centroid method, is commonly used for this purpose in the Fuzzy AHP. The COA method helps convert the fuzzy numbers into crisp values, which represent the final priority weights for the criteria or alternatives.

$$w_i=\frac{{\tilde{w}}_l+{\tilde{w}}_m+{\tilde{w}}_u}{3}$$

(13)

After applying the COA method to defuzzify the fuzzy geometric means, the crisp values obtained represent the final priority weights. These weights can then be used to rank the alternatives or criteria in order of importance, leading to a decision or recommendation based on the aggregated opinions and preferences.

4. Case Study

In this paper, the AHP and Fuzzy AHP are applied for the selection of attractive places for individual tourist trip design. Individual preferences in the tourism case study are captured based on three main criteria: types of tourism, tourism facilities, and tourism areas. Each of these criteria includes specific sub-criteria for consideration. The detailed explanations are as follows.

4.1. Types of Tourism (C1): Different Types of Tourism Have Varying Appeals to Different Tourists Based on Their Interests and Motivations

• Cultural Tourism (C1.1) focuses on learning about and experiencing local cultures, including visiting historical sites, palaces, museums, and attending local art and cultural events.
• Nature Tourism (C1.2) is for those who love the outdoors and natural beauty, such as national parks, forests, mountains, waterfalls, and other natural areas.
• Adventure Tourism (C1.3) involves thrilling and challenging activities such as mountain climbing, white-water rafting, diving, rock climbing, and trekking.
• Religious Tourism (C1.4) involves pilgrimages or visits to religious sites, such as temples, churches, mosques, and sacred places.
• Entertainment Tourism (C1.5) focuses on relaxation and fun, including trips to amusement parks, zoos, shopping centers, and entertainment venues.

4.2. Tourism Facilities (C2): The Availability and Quality of Facilities Play a Significant Role in the Attractiveness of a Destination

• Accommodation Facilities (C2.1) is the quality, variety, and availability of lodging options, including hotels, motels, resorts, and vacation rentals.
• Food and Beverage Facilities (C2.2) is the availability of restaurants and local cuisine.
• Transportation Facilities (C2.3) is the accessibility and convenience of travel options.
• Parks and Gardens (C2.4) provide open spaces for relaxation and recreation.
• Safety and Security Facilities (C2.5) refers to the police stations or security services to ensure safety.

4.3. Tourism Areas (C3): The Regions or Districts Within a Destination That May Offer Unique Attractions, Cultural Experiences, or Local Specialties

In this paper, the districts in Tak Province are used as a Case Study. Tak Province in Thailand is divided into nine districts, each with its own unique attractions and characteristics.

• Mueang Tak District (C3.1) is the capital district of Tak Province, it serves as the administrative and commercial center.
• Mae Sot District (C3.2) is located on the border with Myanmar. This district is a hub for trade and cross-border activities.
• Ban Tak District (C3.3) is known for its historical significance and scenic landscapes.
• Sam Ngao District (C3.4) features beautiful natural landscapes and is home to the Bhumibol Dam.
• Mae Ramat District (C3.5) is a rural district with lush forests and agricultural activities.
• Phop Phra District (C3.6) is known for its agricultural land and cross-border trade with Myanmar.
• Tha Song Yang District (C3.7) is rich in natural beauty and cultural heritage, especially of the Karen ethnic group.
• Wang Chao District (C3.8) is primarily an agricultural area with a peaceful rural environment.
• Umphang District (C3.9) is known for its remote and rugged terrain. It is a favorite for adventure and nature lovers.

According to a case study of Tak province, 65 attractive locations are considered as options for the tourist. Figure 5 displays the hierarchy structure of the case study. Each tourist is convened to decide and compare each pair of the main criteria and sub-criteria based on their opinions. This pairwise comparison allows for the subjective preferences of the tourists to be quantitatively analyzed and incorporated into the decision-making process. By considering these criteria and sub-criteria, the model aims to facilitate informed decision-making for an individual tourist to decide where to go during their trip. The calculation of the AHP and Fuzzy AHP are explained in the next sub-section.

5. Results and Discussion

5.1. AHP Method

The experiment starts with taking sample answers to compare each pair of the main criteria and sub-criteria based on their opinions. This process provides the necessary data for the AHP model to analyze and prioritize the factors that influence tourists’ decisions.

Table 4. Comparison matrix of three main criteria for the AHP model.

Main Criteria	C1	C2	C3
C1	1	3	5
C2	1/3	1	1
C3	1/5	1	1

,

Table 5. Comparison matrix of sub-criteria in C1 for the AHP model.

C1	C1.1	C1.2	C1.3	C1.4	C1.5
C1.1	1	3	3	3	5
C1.2	1/3	1	1	1	1
C1.3	1/3	1	1	1	1
C1.4	1/3	1	1	1	1
C1.5	1/5	1	1	1	1

,

Table 6. Comparison matrix of sub-criteria in C2 for the AHP model.

C2	C2.1	C2.2	C2.3	C2.4	C2.5
C2.1	1	1/3	1	1	1
C2.2	3	1	3	3	5
C2.3	1	1/3	1	1	1/3
C2.4	1	1/3	1	1	1
C2.5	1	1/5	3	1	1

,

Table 7. Comparison matrix of sub-criteria in C3 for the AHP model.

C3	C3.1	C3.2	C3.3	C3.4	C3.5	C3.6	C3.7	C3.8	C3.9
C3.1	1	1/3	3	3	3	3	3	3	3
C3.2	3	1	3	3	3	3	3	3	3
C3.3	1/3	1/3	1	1	1	1	1	1	1
C3.4	1/3	1/3	1	1	1	1	1	1	1
C3.5	1/3	1/3	1	1	1	1	1	1	1
C3.6	1/3	1/3	1	1	1	1	1	1	1
C3.7	1/3	1/3	1	1	1	1	1	1	1
C3.8	1/3	1/3	1	1	1	1	1	1	1
C3.9	1/3	1/3	1	1	1	1	1	1	1

and

Table 8. Comparison matrix of 65 locations respect to Cultural Tourism (C1.1).

C1.1	L1	L2	L3	L4	L5	L6	L..	L..	L65
L1	1	1	1	1	1/3	1	…	…	1
L2	1	1	1	1	1/3	1	…	…	1
L3	1	1	1	1	1/3	1	…	…	1
L4	1	1	1	1	1/3	1	…	…	1
L5	3	3	3	3	1	3	…	…	3
L6	1	1	1	1	1/3	1	…	…	1
L..	…	…	…	…	…	…	1	…	…
L..	…	…	…	…	…	…	…	1	…
L65	1	1	1	1	1/3	1	…	…	1

are examples of tourists’ opinion on each aspect. Also, each location is compared based on both criteria and sub-criteria. The comparison matrix of 65 locations with respect to the sub-criteria Cultural Tourism (C1.1) is shown in Table 8. This step is repeated for all sub-criteria, C1.2 to C3.9.

According to the comparison matrix in Table 4, it is evident that C1 is compared to itself with a preference score of 1, indicating its equal importance when measured against itself. When comparing C1 to C2 and C3, the preference scores of 3 and 5 suggest that tourists display a weak preference for C1 over C2 and a strong preference for C1 over C3. This interpretation indicates that tourists may feel indifferent regarding the type of tourism itself; however, they exhibit varying degrees of preference when assessing the type of tourism in relation to tourism facilities and tourism areas. Specifically, tourists appear to prioritize the characteristics of tourism facilities and areas more significantly than the tourism type, suggesting that these factors play a more crucial role in their decision-making processes.

Table 5 is the pairwise comparison matrix for determining the relative weights of the sub-criteria within the main criterion C1. It is indicated that C1.1 is compared to C1.2, C1.3, and C1.4 with a score of 3, meaning that C1.1 is preferred more strongly over those three sub-criteria. Furthermore, the preference score of 5 when comparing C1.1 to C1.5 indicates a very strong preference, emphasizing its critical role in the overall evaluation. In contrast, C1.2, C1.3, and C1.4 exhibit equal preference when compared to one another, with a score of 1 indicates that tourists view them as equally important.

The interpretations for Table 6 and Table 7 can be explained in a manner similar to Table 4 and Table 5, highlighting the composition of preferences and the relative importance of sub-criteria within each main criterion C2 and C3.

To evaluate each sub-criterion for 65 locations, we require 19 comparison matrices. An example of the comparison matrix of 65 locations in relation to the sub-criterion C1.1 (Cultural Tourism) is shown in Table 8. This matrix facilitates a systematic evaluation of the cultural attributes and offerings of each location. For instance, when comparing L5 to L1 with a score of 3, it indicates that L5 is perceived to have a stronger emphasis on Cultural Tourism than L1. This scoring reflects the preference for L5 based on its cultural significance and offerings, highlighting its appeal as a cultural destination when compared to L1. Each score in the matrix allows for the ranking of locations, ultimately assisting tourists and decision-makers in identifying the most culturally rich destinations. This systematic interpretation is applied across all 19 matrices.

Weights and consistency ratios (CR) for each main criterion and sub-criterion of the example tourist are illustrated in

Table 9. Weights and CR of the example tourist.

Main Criteria	Weights	CR	Sub-Criteria	Weights	CR
C1: Types of Tourism	0.66	0.027	C1.1: Cultural Tourism	0.46	0.006
			C1.2: Nature Tourism	0.14
			C1.3: Adventure Tourism	0.14
			C1.4: Religious Tourism	0.14
			C1.5: Entertainment Tourism	0.13
C2: Tourism Facilities	0.19		C2.1: Accommodation Facilities	0.13	0.058
			C2.2: Food and Beverage Facilities	0.45
			C2.3: Transportation Facilities	0.12
			C2.4: Parks and Gardens	0.13
			C2.5: Safety and Security Facilities	0.17
C3: Tourism Areas	0.16		C3.1: Mueang Tak District	0.21	0.009
			C3.2: Mae Sot District	0.26
			C3.3: Ban Tak District	0.08
			C3.4: Sam Ngao District	0.08
			C3.5: Mae Ramat District	0.08
			C3.6: Phop Phra District	0.08
			C3.7: Tha Song Yang District	0.08
			C3.8: Wang Chao District	0.08
			C3.9: Umphang District	0.08

. These weights provide a quantitative measure of the relative importance of each criterion in the decision-making process, while the CR indicates the consistency of the pairwise comparisons, ensuring that the assessments made are reliable and valid.

Table 10. The 10 selected tourist destinations from the AHP.

Rank	Tourist Attraction	Types of Tourism	Tourism Facilities	Tourism Areas
1	Chinese house alley	C1.1	C2.1, C2.2, C2.4, C2.5	C3.1
2	Ban Pak Rong Huai Chi Learning Center	C1.1	C2.1, C2.2, C2.4	C3.3
3	Shrine of King Taksin	C1.2	C2.1, C2.2, C2.4, C2.5	C3.1
4	Bhumibol Dam	C1.2	C2.1, C2.2, C2.3, C2.4, C2.5	C3.4
5	Taksin Maharat National Park	C1.2	C2.1, C2.2, C2.4, C2.5	C3.1
6	2nd Thai–Myanmar Friendship Bridge	C1.5	C2.2, C2.4	C3.2
7	Rimmoei Market	C1.5	C2.2, C2.4	C3.2
8	Thararak Waterfall	C1.3	C2.2, C2.4	C3.2
9	Wat Phothikhun	C1.4	C2.2, C2.4, C2.5	C3.2
10	Wat Mani Praison	C1.4	C2.2, C2.4, C2.5	C3.2

presents the recommendations for the tourist based on the calculated weights and evaluations. These recommendations are drawn from the analysis and ranking of the locations, allowing the tourist to select the most suitable options that align with their preferences and needs.

5.2. Fuzzy AHP Method

In the Fuzzy AHP, tourists’ opinions on main and sub-criteria can be aggregated to account for the uncertainty and subjectivity in their preferences.

Table 11. Fuzzy comparison matrix of three main criteria.

Main Criteria	C1	C2	C3
C1	1	$\tilde{3}$	$\tilde{5}$
C2	${\tilde{3}}^{-1}$	1	$\tilde{1}$
C3	${\tilde{5}}^{-1}$	${\tilde{1}}^{-1}$	1

,

Table 12. Fuzzy comparison matrix of sub-criteria in C1.

C1	C1.1	C1.2	C1.3	C1.4	C1.5
C1.1	1	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{5}$
C1.2	${\tilde{3}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C1.3	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$
C1.4	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$
C1.5	${\tilde{5}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1

,

Table 13. Fuzzy comparison matrix of sub-criteria in C2.

C2	C2.1	C2.2	C2.3	C2.4	C2.5
C2.1	1	${\tilde{3}}^{-1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C2.2	$\tilde{3}$	1	$\tilde{3}$	$\tilde{3}$	$\tilde{5}$
C2.3	${\tilde{1}}^{-1}$	${\tilde{3}}^{-1}$	1	$\tilde{1}$	${\tilde{3}}^{-1}$
C2.4	${\tilde{1}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$
C2.5	${\tilde{1}}^{-1}$	${\tilde{5}}^{-1}$	$\tilde{3}$	${\tilde{1}}^{-1}$	1

,

Table 14. Fuzzy comparison matrix of sub-criteria in C3.

C3	C3.1	C3.2	C3.3	C3.4	C3.5	C3.6	C3.7	C3.8	C3.9
C3.1	1	1/3	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$
C3.2	$\tilde{3}$	1	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$
C3.3	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C3.4	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C3.5	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C3.6	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$
C3.7	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$
C3.8	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$
C3.9	${\tilde{3}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1

and

Table 15. Fuzzy comparison matrix of 65 locations respect to Cultural Tourism (C1.1).

C1.1	L1	L2	L3	L4	L5	L6	L..	L..	L65
L1	1	$\tilde{1}$	$\tilde{1}$	$\tilde{1}$	${\tilde{3}}^{-1}$	$\tilde{1}$	…	…	$\tilde{1}$
L2	${\tilde{1}}^{-1}$	1	$\tilde{1}$	$\tilde{1}$	${\tilde{3}}^{-1}$	$\tilde{1}$	…	…	$\tilde{1}$
L3	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	$\tilde{1}$	${\tilde{3}}^{-1}$	$\tilde{1}$	…	…	$\tilde{1}$
L4	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	1	${\tilde{3}}^{-1}$	$\tilde{1}$	…	…	$\tilde{1}$
L5	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	$\tilde{3}$	1	$\tilde{3}$	…	…	$\tilde{3}$
L6	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{3}}^{-1}$	1	…	…	$\tilde{1}$
L..	…	…	…	…	…	…	1	…	…
L..	…	…	…	…	…	…	…	1	…
L65	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{1}}^{-1}$	${\tilde{3}}^{-1}$	${\tilde{1}}^{-1}$	…	…	1

present the fuzzy comparison matrices for each criterion based on an example of tourists’ opinions. The Fuzzy AHP method described in Section 3.2 is employed to generate a personalized list of tourist attractions for individual tourists based on their preferences. The process involves constructing fuzzy comparison matrices, calculating fuzzy geometric means, normalizing these means, and defuzzifying them to obtain crisp weights. These weights can then be used to generate a personalized list of tourist attractions based on individual preferences.

The weight results in

Table 16. Weights of the example tourist from the Fuzzy AHP.

Main Criteria	Normalize Weights	Sub-Criteria	Normalize Weights
C1: Types of Tourism	0.65	C1.1: Cultural Tourism	0.45
		C1.2: Nature Tourism	0.16
		C1.3: Adventure Tourism	0.15
		C1.4: Religious Tourism	0.13
		C1.5: Entertainment Tourism	0.11
C2: Tourism Facilities	0.20	C2.1: Accommodation Facilities	0.12
		C2.2: Food and Beverage Facilities	0.43
		C2.3: Transportation Facilities	0.12
		C2.4: Parks and Gardens	0.18
		C2.5: Safety and Security Facilities	0.15
C3: Tourism Areas	0.15	C3.1: Mueang Tak District	0.20
		C3.2: Mae Sot District	0.25
		C3.3: Ban Tak District	0.09
		C3.4: Sam Ngao District	0.09
		C3.5: Mae Ramat District	0.08
		C3.6: Phop Phra District	0.08
		C3.7: Tha Song Yang District	0.07
		C3.8: Wang Chao District	0.07
		C3.9: Umphang District	0.07

are illustrated in the same manner as those obtained from the AHP model in Table 9. The suggested attractive locations from both the AHP and Fuzzy AHP are slightly similar (

Table 17. The 10 selected tourist destinations from the Fuzzy AHP.

Rank	Tourist Attraction	Types of Tourism	Tourism Facilities	Tourism Areas
1	Chinese house alley	C1.1	C2.1, C2.2, C2.4, C2.5	C3.1
2	Ban Pak Rong Huai Chi Learning Center	C1.1	C2.1, C2.2, C2.4	C3.3
3	Shrine of King Taksin	C1.2	C2.1, C2.2, C2.4, C2.5	C3.1
4	Bhumibol Dam	C1.2	C2.1, C2.2, C2.3, C2.4, C2.5	C3.4
5	Taksin Maharat National Park	C1.2	C2.1, C2.2, C2.4, C2.5	C3.1
6	Rattanakosin 200 Years Celebration Hanging Bridge	C1.2	C2.2, C2.4, C2.5	C3.1
7	Thi Lo Su Waterfall	C1.3	C2.2, C2.4	C3.9
8	Thararak Waterfall	C1.3	C2.2, C2.4	C3.2
9	Wat Phothikhun	C1.4	C2.2, C2.4, C2.5	C3.2
10	Wat Mani Praison	C1.4	C2.2, C2.4, C2.5	C3.2

).

The similarity between the results from the AHP and Fuzzy AHP can be attributed to the fact that both methods are designed to prioritize alternatives based on pairwise comparisons of criteria. However, they handle uncertainty and subjectivity in different ways, which can lead to slightly different outcomes.

5.3. AHP and Fuzzy AHP Models Testing

To evaluate the AHP and Fuzzy AHP models for selecting tourist attractions, data were collected from 30 respondents. The steps are as follows.

• Step 1: Data Collection.

The data were collected from 30 respondents (

Table 18. General information of 30 respondents.

General Information
Age:	Number of Respondents	Gender:	Number of Respondents
18–25	3	Male	15
26–35	15	Female	15
36–45	5	Prefer not to say	-
46–55	4
56 and above	3
How often do you travel for leisure:
Once a year	5
2–3 times a year	20
4–5 times a year	4
More than 5 times a year	1

). Each respondent was asked to provide pairwise comparisons of elements at each level of the hierarchy based on their individual preferences. These pairwise comparisons captured the relative importance of different criteria and alternatives, providing a foundation for the AHP and Fuzzy AHP methods.

• Step 2: Model Application.

The collected responses were used as input for both the AHP and Fuzzy AHP methods. For the AHP, the pairwise comparison matrices were processed to compute the priority weights of each criterion and alternative. For the Fuzzy AHP, fuzzy pairwise comparison matrices were constructed based on the respondents’ inputs, and fuzzy logic was applied to derive the priority weights. Both methods generated recommendations for tourist attractions based on the input data.

• Step 3: Results Distribution.

The recommendations derived from the AHP and Fuzzy AHP models were returned to each respondent. This step ensured that each respondent could review and evaluate the recommendations generated by both methods.

• Step 4: Satisfaction Assessment.

To assess the effectiveness and accuracy of the AHP and Fuzzy AHP models, respondents were asked to complete a questionnaire. The questionnaire focused on evaluating how well the recommendations aligned with respondents’ decision-making criteria and measured the respondents’ perceptions of the accuracy and relevance of the recommendations provided by both methods.

In the testing process, we provide participants with a clear explanation of the study’s purpose, specifically outlining how their responses will be utilized in developing the recommendation system. We utilize a consent form that explicitly states that the data collected will be used solely for testing the model. Importantly, we do not collect any information that could negatively affect the participants or compromise their privacy. This approach ensures that participants feel safe and informed about their involvement in the research.

Table 19. Experience with the AHP and Fuzzy AHP results.

Questions	Very Dissatisfied	Dissatisfied	Neutral	Satisfied	Very Satisfied
(1) How satisfied are you with the tourist destination recommended by the AHP method?	-	-	26.7%	73.3%	-
(2) How satisfied are you with the tourist destination recommended by the Fuzzy AHP method?	-	-	23.3%	70.0%	6.7%
(3) Did you find the recommendations from the AHP method to be aligned with your personal preferences?	-	-	33.3%	63.3%	3.3%
(4) Did the Fuzzy AHP method better capture your preferences and expectations compared to the AHP method?	-	-	40.0%	60.0%	-

summarizes respondents’ satisfaction with tourist destination recommendations generated by the AHP and Fuzzy AHP methods. Regarding the AHP method, 73.3% of respondents reported being satisfied, while 26.7% remained neutral, with no respondents expressing dissatisfaction. Similarly, for the Fuzzy AHP method, 70.0% were satisfied, 23.3% were neutral, and 6.7% were very satisfied, with no dissatisfaction reported. When it comes to the alignment of the AHP recommendations with personal preferences, 63.3% of respondents felt that the AHP method aligned with their preferences, 33.3% were neutral, and 3.3% were very satisfied. Lastly, 60.0% of respondents believed that the Fuzzy AHP method better captured their preferences and expectations compared to the AHP method, while 40.0% remained neutral, and none expressed dissatisfaction.

6. Discussion and Conclusions

This study explored the application of the AHP and Fuzzy AHP models in the context of tourism, focusing on their effectiveness in capturing and evaluating individual tourist preferences. The findings of this study contribute significantly to the existing body of literature on tourism decision-making by reinforcing the relevance of the AHP and Fuzzy AHP in capturing and analyzing individual tourist preferences. Previous studies, such as those by [11,13], have underscored the effectiveness of the AHP in various contexts, including tourism destination selection and cultural heritage site evaluation. By extending this research to include the Fuzzy AHP, our study not only builds upon the existing framework but also addresses the inherent uncertainties and subjectivities that the traditional AHP may overlook, thereby enhancing the granularity of tourist preference modeling.

The proposed models, the AHP and Fuzzy AHP, are tested through the empirical analysis involving 30 respondents and it was demonstrated that both the AHP and Fuzzy AHP are effective tools for generating recommendations for tourist attractions. The satisfaction assessment revealed a high level of satisfaction with both methods, with the Fuzzy AHP showing a slight edge in better capturing and aligning with respondents’ preferences. This finding suggests that while the AHP is a robust method for decision-making, the integration of fuzzy logic in the Fuzzy AHP provides a more nuanced approach, particularly in dealing with the subjectivity of human judgments. Consequently, the study concludes that the Fuzzy AHP can be a more effective model in tourism decision-making, offering a valuable tool for tailoring tourism services to meet the unique needs of individual tourists.

The AHP and Fuzzy AHP approaches have demonstrated their effectiveness in providing suitable recommendations, as evidenced by the percentage of satisfaction among respondents. However, there is always the possibility that some results may not fully align with tourists’ perspectives and expectations. If this occurs, it is crucial to have a mechanism in place for obtaining further feedback from the tourists. These insights contribute to the advancement of decision support systems in the tourism industry, potentially leading to improved service quality and enhanced tourist satisfaction.