A Hybrid Multi-Criteria Decision Framework for Internet Technology Selection in Smart Tourism Systems

Abstract

The digital transformation of tourist facilities requires careful selection of technologies that can provide secure, stable and scalable network infrastructure. Due to the possibility of application in different sectors with different specificities, the focus of the research was placed on the implementation of smart tourist services. A hybrid multi-criteria decision-making model based on PIPRECIA and MVA models was applied for the research. Based on the literature and the opinions of experts in the field, evaluation criteria such as bandwidth, latency, energy efficiency, security and privacy, scalability, costs and interoperability were defined, and internet technologies such as Li-Fi, Wi-Fi 7, Wi-Fi 6, private 5G networks, Ethernet-over-Power (EoP), NB-IoT and LoRaWAN were defined. The results obtained put the security and privacy criterion at the top (0.2253), followed by scalability (0.1952) and bandwidth (0.1624). The obtained results indicate that Wi-Fi 7 achieved the highest weighted score (4.2247), followed closely by Li-Fi (4.2177) and Wi-Fi 6 (4.0771). Wi-Fi 7 demonstrated particularly strong performance in scalability, interoperability and bandwidth, making it highly suitable for environments with high user density. Li-Fi achieved very high scores in security and latency, which makes it particularly appropriate for security-sensitive smart tourism environments. Lower-ranked technologies such as NB-IoT and LoRaWAN proved valuable for supporting IoT and monitoring functions, rather than as primary communication infrastructure. The proposed model has proven to be a flexible, transparent and practical tool for strategic decision-making in the field of smart tourism. In addition to the basic application presented in the paper, the model has the potential to be adapted to different contexts and expanded with additional criteria or new technologies. The proposed hybrid approach can serve as a useful decision-making tool for tourism managers, system engineers and urban planners who are looking for optimal solutions for the development of digital infrastructure.

1. Introduction

Digital transformation in tourism is one of the key drivers of innovation in modern society [1]. In an era marked by increasing demand for personalized services, optimal resource management and development, the tourism industry is rapidly embracing modern digital technologies in its business [2]. Smart tourism facilities, which integrate information and communication technology (ICT) to improve the user experience and the competitiveness of destinations, play a particularly important role [3]. Smart tourism facilities rely on IoT devices, sensors, artificial intelligence and advanced networks to automate processes and provide interactive services in real time. However, at the heart of every smart tourism system lies a reliable and scalable internet infrastructure [4]. Technologies such as Wi-Fi 6, Wi-Fi 7, Li-Fi, Private 5G networks, NB-IoT, LoRaWAN and Ethernet-over-Power (EoP) are becoming fundamental components for connecting digital services and achieving good user experiences [5]. The listed technologies offer advantages and limitations, and their choice is adapted to the needs of each facility [6]. Adaptation refers to the number of users, type of services, energy efficiency, security requirements and restrictions. Moreover, there is no universal solution that can meet all the requirements of the digital infrastructure, [7] which puts the selection process in a position where a multi-criteria analysis approach can be applied.

In practice, decision-makers often lack the tools and information needed to objectively assess which technologies are most appropriate for their specific context [8]. A number of factors, such as data transfer speed, latency, security, interoperability, cost and energy efficiency, can be in conflict depending on the environment and business strategy [9]. For example, a luxury hotel may prefer Li-Fi due to its high security. Again on the other hand, coworking spaces, due to the large number of users, may rely more on Wi-Fi 6 or Wi-Fi 7 due to their scalability and support for high device density [10]. Or, for example, if we consider environmentally friendly resorts, they may give priority to energy-efficient solutions such as NB-IoT and LoRaWAN [11].

To solve such a complex problem, it is necessary to use a decision-making model. This model should use relevant criteria to enable a transparent decision-making process. If we consider all this, such an MCDM model would represent a framework for evaluating different technologies [12]. The application of such models allows for the consideration of quantitative and qualitative factors. In practice, a large number of MCDM models have been developed recently that differ from each other [13]. Their application depends on the needs of the research in which they are applied [14,15]. There are also hybrid approaches that combine two or more methods. They allow for a combination of subjective expert assessments with objective mechanisms. The model that we present in this paper applies a hybrid approach in combination with the Pivot Pairwise Relative Criteria Importance Assessment (PIPRECIA) and the Multi-Voting Analysis (MVA) method. This combination was chosen so that the PIPRECIA model can be used for relatively quick and easy, and, on the other hand, very precise determination of the importance of criteria. On the other hand, the MVA model allows for a wider inclusion of expert opinions through structured voting and evaluation. The combination of these two models into one allows for synergy between precision and applicability. The practicality of this model is very important for the decision-making process in the field of smart tourism [16].

Our goal in this research is to present a comprehensive methodological framework for assessing internet technologies for application in the tourism sector on smart facilities. Key criteria were identified and quantified from the literature and consultations with experts in the field. These criteria were crossed with seven modern technologies that are significant for this sector [17]. The expected contribution of the work is reflected in the development of a hybrid model and the analysis obtained by its application, which include technical, economic and operational aspects [18].

In order to clarify the problem addressed in this study and to guide the methodological development of the proposed framework, the research is structured around the following research questions:

• Q1: Which evaluation criteria are the most relevant for assessing internet technologies in smart tourism facilities?
• Q2: Which of the selected internet technologies achieves the most favorable overall performance when evaluated through a hybrid PIPRECIA-MVA framework?
• Q3: How does the separation of expert groups in the criteria weighting and alternative evaluation phases contribute to the transparency and methodological independence of the decision-making process?
• Q4: How stable are the obtained technology rankings under changes in the weights of the most influential criteria?

To improve traceability between the research objectives and the obtained results, each research question is addressed in a dedicated section of the manuscript. RQ1 is addressed through the identification and weighting of evaluation criteria in Section 5.1 and Section 5.2. RQ2 is addressed through the evaluation and ranking of internet technologies presented in Section 5.3 and Section 5.4. RQ3 is examined through the methodological design and separation of expert groups described in Section 4.4 and further discussed in Section 6. RQ4 is addressed through the sensitivity analysis presented in Section 5.5. The consolidated answers to all research questions are summarized in Section 6.2.

In recent years, MCDM methods such as AHP, PIPRECIA, TOPSIS and hybrid approaches have been increasingly applied in tourism and smart infrastructure contexts. However, most studies focus on destination management or customer experience, while limited attention has been given to the selection of underlying internet technologies. Recent works also highlight the growing importance of next-generation technologies such as Wi-Fi 7, Li-Fi and private 5G networks in enabling smart tourism services. Despite these advancements, there is still a lack of integrated decision-making frameworks that combine structured weighting methods with participatory evaluation mechanisms.

The originality of this study lies in the integration of the PIPRECIA method for structured and efficient criteria weighting with the MVA approach for participatory evaluation of alternatives. Unlike commonly used combinations such as AHP-TOPSIS or AHP-VIKOR, the proposed model emphasizes methodological independence through the separation of expert groups and introduces a flexible evaluation mechanism suitable for dynamic and heterogeneous smart tourism environments. The model focuses specifically on internet infrastructure selection, which remains underexplored in existing MCDM-based tourism studies.

2. Theoretical Framework and Conceptualization of the Problem

Digital transformation in tourism is no longer an option but a real challenge for maintaining competitiveness and sustainability [19]. In smart tourism facilities, a digital infrastructure is applied that enables the operation of complex systems. These systems enable personalization of services, resource optimization, remote control and connection with global trends in tourism [20]. One of the main trends in smart tourism is the selection of appropriate internet technology with which other digital functionalities are implemented [21].

2.1. Digital Transformation in Tourism

The concept of digital transformation implies the application of modern communication technologies with the aim of improving business processes. The tourism sector is increasingly introducing various digital innovations that serve managers and guests. For these needs, it is necessary to have a stable and scalable network infrastructure. If all conditions are not met, it is not possible to successfully implement digital transformation. It is possible that there is a mismatch between the requirements and capabilities of the technology that has been chosen, which leads to a decrease in the quality of service. This can also increase security risks and maintenance costs.

2.2. The Concept of Smart Tourist Facilities

The concept of smart tourist facilities includes a combination of physical infrastructure and digital services. These technologies are integrated so that they can meet user needs in real time [22]. Some of the technologies in these types of tourist facilities are sensor networks, artificial intelligence, and cloud computing, which dynamically manage the environment. In this way, air conditioning, lighting, access control, security protocols, as well as various personalized services are managed.

Key characteristics of smart tourist facilities include:

• Automation;
• Personalization;
• Optimal resource management;
• High level of security and protection;
• Integration with smart cities.

2.3. Internet Technologies as the Basis of Digital Infrastructure

In smart tourist facilities, internet technologies represent a basic component of digital infrastructure [23]. Their role is to enable internet access but also act as a resource for connecting to all digital services [24]. The choice of technology directly affects the quality of services and user satisfaction. It is this characteristic of internet technology that poses a strategic issue that affects competitiveness and efficiency [25].

These technologies can be divided into three basic groups. The first group consists of wireless networks that enable high speeds and support a large number of simultaneous users. The second group consists of mobile and private networks that offer high reliability, low latency and greater control. The third group consists of low-power technologies designed for IoT devices, which are usually referred to as LPWAN (Low-Power Wide-Area Network) [26].

Each of the above groups of technologies has certain advantages, but also limitations that need to be carefully considered. All three groups have their own advantages and disadvantages, depending on the requirements. This proves that a systematic and carefully designed approach is essential for selecting the optimal technology [27]. Considering only technical specifications is often not enough, so it is necessary to apply an MCDM approach that includes technical, economic, security and operational aspects. Based on these aspects, it is possible to make informed decisions aligned with the needs and strategic goals of a smart tourism facility.

3. Literature Review

In order to establish a solid methodological basis for the evaluation of internet technologies in smart tourist facilities, it is necessary to analyze the existing literature in three key areas:

• Application of MCDM methods in the context of smart tourism;
• Selection and evaluation of internet technologies in IoT environments;
• Integration of subjective and objective methods in evaluation models.

Previous research has shown that MCDM approaches such as AHP, TOPSIS and SWARA are most often used for strategic decision-making in tourism. On the other hand, they are rarely used for the technical selection of network solutions. Also, the evaluation of technologies in smart objects is often limited to individual technical case studies, without a systematic analysis based on multiple criteria. The emergence of hybrid models that combine expert judgment with collective mechanisms provides new opportunities for decision-making in complex environments. These models allow better adaptation to different facilities, technical requirements and limitations. The review of the literature highlights the need for the development of a new applicable approach to evaluation, which is what this study focused on.

3.1. Multi-Criteria Decision-Making Models in Smart Tourism

MCDM has become the dominant methodology for analyzing complex problems, where decisions have to be made in the presence of multiple, often conflicting, criteria. Methods such as AHP, TOPSIS, ELECTRE, and SWARA are often used for destination selection, hotel ranking, sustainability assessment, and digital service design [28,29,30,31,32,33].

A limitation of these studies is that most focus on macro aspects of tourism, such as destination management or customer satisfaction analysis. MCDM models are rarely applied to infrastructure-level decision-making. Studies dealing with technology selection are usually limited to technical case studies and often do not include the integration of subjective assessments of multiple stakeholders [34].

3.2. Evaluation Criteria for Selecting IoT Technologies

The identification of criteria for selecting IoT technologies depends on the application context, but the literature usually cites factors such as bandwidth, latency, energy consumption, security, scalability, cost, and interoperability [35,36]. In tourism, these factors gain an additional dimension due to the dynamic nature of user needs, seasonal fluctuations and the need for personal data protection. There are studies [37] which suggest that scalability and security become particularly important in the hospitality sector. These studies highlight interoperability and costs as key elements for implementing IoT solutions in tourism. This is especially true for smaller facilities operating with limited budgets [38,39].

3.3. Integrated Subjective–Participatory Decision-Making Models

Recent studies emphasize the advantage of an integrated approach to multi-criteria analysis that combines subjective and participatory approaches. PIPRECIA [40], a method based on pairwise comparisons with a pivot structure, allows for a quick and easy determination of criterion weights, even in complex systems [41]. On the other hand, MVA methods allow for the inclusion of a wider panel of experts. This method allows for collective voting, which enhances the democratic nature of the decision-making process. Research showing that hybrid decision-making models that combine subjective methods with participatory mechanisms yield significantly more robust results, especially in the selection of smart city technologies [42,43,44]. Introducing methodological independence further increases the validity of the model and reduces the risk of cognitive bias.

Based on the reviewed literature, it can be observed that existing MCDM studies in tourism and smart infrastructure primarily focus on destination management, customer satisfaction, sustainability assessment and service optimization. However, significantly less attention has been devoted to the structured evaluation of internet infrastructure technologies in smart tourism facilities. In addition, most existing studies rely on traditional hybrid combinations such as AHP-TOPSIS, AHP-VIKOR or SWARA-based approaches, while participatory expert evaluation mechanisms and methodological independence between weighting and evaluation phases remain insufficiently explored. To clarify the research gap and position the contribution of this study,

Table 1. Comparative overview of existing MCDM approaches in smart tourism and technology evaluation studies.

Study	Application Context	Applied Method	Main Focus	Identified Limitation	Contribution of This Study
Kwok & Lau [30]	Hotel selection	TOPSIS	Hotel decision support	Focus on service selection rather than infrastructure technologies	Focus on internet infrastructure technologies
Wu et al. [31]	Hotel evaluation	BWM-based model	Customer-oriented hotel assessment	No technological infrastructure analysis	Technology-oriented smart tourism evaluation
Tran Thi Hoang et al. [43]	Smart city projects	Various MCDM approaches	Smart city decision-making	Limited participatory evaluation	Participatory expert evaluation through MVA
Abdullah et al. [12]	Industry 4.0 technologies	Hybrid fuzzy MCDM	Manufacturing strategy assessment	Different application domain	Application in smart tourism systems
Matić et al. [35]	Supplier selection	Hybrid MCDM	Sustainable supplier evaluation	No smart tourism application	Smart tourism internet technology selection
Proposed study	Smart tourism infrastructure	PIPRECIA-MVA	Evaluation of internet technologies	Methodological independence and participatory evaluation integrated	Hybrid framework for smart tourism internet infrastructure

provides a comparative overview of selected existing approaches.

As shown in Table 1, previous studies have largely focused on service-oriented or destination-oriented decision-making problems, while the selection of internet infrastructure technologies for smart tourism systems remains underexplored. Furthermore, existing hybrid MCDM approaches rarely combine structured weighting methods with participatory evaluation mechanisms using independent expert groups. The proposed PIPRECIA-MVA framework attempts to address this gap by integrating transparent criteria weighting with participatory technology evaluation in the context of smart tourism infrastructure.

4. Materials and Methods

To ensure optimal evaluation, this study applies a hybrid MCDM model. The study applies the PIPRECIA method and the MVA method. Each of these methods performs a specific role within the proposed evaluation framework in order to support transparent and reliable decision-making.

The methodology is divided into two phases:

• Phase 1: Determination of criterion weights (PIPRECIA);
• Phase 2: Evaluation and ranking of alternatives (MVA).

4.1. PIPRECIA—Determination of Criterion Weights

The PIPRECIA method uses a pivot (reference) criterion to reduce the complexity of pairwise comparisons. It is this ability that allows it to be more easily applied in practice. The process consists of the following steps:

Step 1:  Define the set of evaluation criteria.

Suppose there are n evaluation criteria denoted as

$$\mathrm{C}=\left\{{\mathrm{C}}_1,{\mathrm{C}}_2,\dots ,{\mathrm{C}}_{\mathrm{n}}\right\}$$

(1)

Step 2: Selecting the pivot criteria.

One of the criteria is selected as the reference and is assigned an importance value ${\mathrm{s}}_1=1$.

Step 3: Assess the relative importance of other criteria.

Experts assess the relative importance of each other criterion ${\mathrm{s}}_{\mathrm{j}}\in \left[0.6,1.4\right]$ relative to the pivot. Values less than 1 indicate less importance, while values greater than 1 indicate more importance.

Step 4: Calculating correction coefficients.

A correction coefficient is calculated for each criterion:

$${\mathrm{k}}_{\mathrm{j}}=2-{\mathrm{s}}_{\mathrm{j}}$$

(2)

Step 5: Recursively calculate preliminary weights.

Using the initial value ${\mathrm{q}}_1=1$, preliminary weights are calculated recursively:

$${\mathrm{q}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{j}-1}}{{\mathrm{k}}_{\mathrm{j}}}}, \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{j}=\mathrm{2,3},\dots ,\mathrm{n}$$

(3)

Step 6: Weight normalization

The final criterion weights are calculated as

$${\mathrm{w}}_{\mathrm{j}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{j}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{q}}_{\mathrm{j}}}}$$

(4)

These weights are then used to weight the alternative ratings in the second stage (MVA).

4.2. MVA—Evaluation and Ranking of Alternatives

MVA is a participatory decision-making method that involves expert evaluation of alternatives according to defined criteria.

Step 1: Defining alternatives and criteria.

Let us assume that there are $\mathrm{m}$ alternatives marked as

$$\mathrm{A}=\left\{{\mathrm{A}}_1,{\mathrm{A}}_2,\dots ,{\mathrm{A}}_{\mathrm{m}}\right\}$$

(5)

and a set of criteria $\mathrm{C}=\left\{{\mathrm{C}}_1,{\mathrm{C}}_2,\dots ,{\mathrm{C}}_{\mathrm{n}}\right\}$ with weights ${\mathrm{w}}_{\mathrm{j}}$ obtained from the PIPRECIA method.

Step 2: Evaluating alternatives against criteria.

Each alternative ${\mathrm{A}}_{\mathrm{i}}$ is evaluated according to the criterion ${\mathrm{C}}_{\mathrm{j}}$ by grade ${\mathrm{a}}_{\mathrm{i}\mathrm{j}}\in \left[\mathrm{1,5}\right]$ (where 1 = very poor performance, and 5 = excellent performance).

Step 3: Weighting the grades.

Weighted grades are calculated using the formula

$${\mathrm{a}}_{\mathrm{i}\mathrm{j}}^{\ast }={\mathrm{a}}_{\mathrm{i}\mathrm{j}}·{\mathrm{w}}_{\mathrm{j}}$$

(6)

(where ${\mathrm{a}}_{\mathrm{i}\mathrm{j}}$—alternative assessment ${\mathrm{A}}_{\mathrm{i}}$ by criterion ${\mathrm{C}}_{\mathrm{j}}$, ${\mathrm{w}}_{\mathrm{j}}$—weight of criteria ${\mathrm{C}}_{\mathrm{j}}$ and ${\mathrm{a}}_{\mathrm{i}\mathrm{j}}^{\ast }$—weighted score).

Step 4: Calculating the final alternative score.

For each alternative, the final score is obtained by adding the weighted scores:

$${\mathrm{F}\mathrm{S}}_{\mathrm{i}}={\displaystyle \sum _{\mathrm{j}=1}^{\mathrm{n}}}{\mathrm{a}}_{\mathrm{i}\mathrm{j}}^{\ast }$$

(7)

where ${\mathrm{F}\mathrm{S}}_{\mathrm{i}}$—final score alternatives ${\mathrm{A}}_{\mathrm{i}}$.

Step 5: Ranking the alternatives.

The highest-scoring alternative ${\mathrm{F}\mathrm{S}}_{\mathrm{i}}$ occupies the highest position in the ranking. This value represents a quantitative measure of the suitability of the alternative in the context of all criteria.

4.3. Advantages of the Applied Model

PIPRECIA reduces the number of estimates compared to AHP and allows for faster determination of weights. MVA allows for the inclusion of a larger number of experts without requiring complex calculations. Methodological independence (different groups of experts in both phases) increases validity and eliminates bias. The model is flexible and can be easily adapted to new criteria, technologies and contexts.

Compared to commonly applied hybrid approaches such as AHP-TOPSIS and AHP-VIKOR, the proposed PIPRECIA-MVA framework offers several practical and methodological advantages. Traditional AHP-based models often require a large number of pairwise comparisons, which increases methodological complexity and may introduce inconsistency when a larger number of criteria is considered. In contrast, the PIPRECIA method simplifies the weighting process through pivot-based relative evaluation, reducing the number of required comparisons while maintaining interpretability and transparency.

Furthermore, ranking methods such as TOPSIS and VIKOR are primarily based on mathematical distance optimization from ideal and anti-ideal solutions. Although effective in many applications, such methods provide limited insight into participatory expert interaction during the evaluation phase. The MVA approach applied in this study enables direct incorporation of expert assessments through a structured participatory mechanism, making the evaluation process easier to interpret and more adaptable to real-world smart tourism environments.

An additional contribution of the proposed framework is the methodological independence achieved through the separation of expert groups responsible for criteria weighting and alternative evaluation. This reduces the influence of cognitive overlap and contributes to greater objectivity of the obtained results.

The comparison presented in

Table 2. Comparative characteristics of selected hybrid MCDM approaches.

Aspect	AHP-TOPSIS	AHP-VIKOR	PIPRECIA-MVA	How the Limitation Is Addressed
Pairwise comparison complexity	High	High	Moderate	Pivot-based weighting
Number of required comparisons	Large	Large	Reduced	Reduced through PIPRECIA
Transparency of weighting process	Moderate	Moderate	High	Explicit weighting procedure
Participatory expert evaluation	Limited	Limited	High	MVA expert evaluation
Computational complexity	Moderate	Moderate	Reduced	Simplified weighting and aggregation
Methodological independence	Rarely applied	Rarely applied	Explicitly applied	Independent expert groups
Adaptability to smart tourism environments	Moderate	Moderate	High	Tourism-specific criteria and technologies

Source: Authors’ synthesis based on the literature [12,30,31,32,33,34,35,42,43,44].

indicates that the proposed framework emphasizes methodological transparency, reduced evaluation complexity and stronger expert participation. These characteristics are particularly important in smart tourism systems, where technological decisions are influenced not only by technical performance, but also by operational constraints, user density, security requirements and infrastructure heterogeneity.

4.4. Formation and Division of Expert Groups

In order to achieve maximum reliability and reduce potential bias, experts were divided into two groups with clearly defined tasks. This approach provides methodological independence between the phase of determining the importance of criteria and the phase of evaluating technological alternatives. This reduced the risk of cognitive overlap and the subjective influence of individual evaluators. The methodological structure presented in this section provides the basis for addressing RQ3 by ensuring the separation of criteria weighting and alternative evaluation processes.

• Group A, which consisted of a team of six (6) experts, was responsible for determining the relative weights of the evaluation criteria using the PIPRECIA method.
• Group B also had six (6) experts, who were tasked with evaluating technological alternatives against the criteria defined by Group A using the MVA method.

The structure of the researchers who participated was:

• Experts from the academic community with experience in the field;
• Experts involved in the design and implementation of IoT technologies;
• Experts from the hotel and coworking sectors who have specific knowledge in the field of this research.

This structure of participants allowed for a combination of theoretical and practical knowledge. This allowed for a better understanding of the technical characteristics and their behavior in real operating conditions. All participants were selected based on their relevant knowledge and experience in this field. As part of the preparation, everyone was informed about the objectives of the research as well as the procedures that would be applied during the process.

Experts were selected based on the following criteria:

• Minimum of 5 years of experience in relevant fields;
• Involvement in projects related to ICT or smart tourism systems;
• Familiarity with decision-making processes.

None of the experts reported any financial or professional conflicts of interest related to the evaluated technologies. Participation was voluntary and all assessments were provided independently. The expert panel included participants with complementary professional backgrounds in order to ensure methodological diversity and practical relevance of the evaluation process. Experts involved in Group A primarily specialized in MCDM methodologies, information systems and digital infrastructure analysis, while experts in Group B included professionals from IoT implementation, wireless communication systems, hospitality management and smart tourism operations. The average professional experience of the participants exceeded seven years and several experts had direct involvement in projects related to smart buildings, tourism digitalization and ICT infrastructure deployment. The composition and expertise profile of both expert groups are summarized in

Table 3. Structure and expertise profile of the expert groups.

Expert Group	Area of Expertise	Role in the Study	Average Experience
Group A	MCDM methods, ICT systems, digital infrastructure	Criteria weighting using PIPRECIA	>7 years
Group B	IoT systems, wireless technologies, hospitality management, smart tourism operations	Evaluation of internet technologies using MVA	>7 years

. The number of experts included in this study was determined in accordance with common practices in MCDM research, where expert panels consisting of five to ten participants are frequently considered sufficient for evaluating complex decision-making problems. The primary objective was not statistical representativeness, but the collection of informed judgments from highly qualified professionals with relevant domain expertise. All selected experts possessed more than seven years of professional experience and practical involvement in projects related to smart tourism, ICT infrastructure, IoT systems or decision-support methodologies. Therefore, the selected panel size was considered adequate for ensuring the reliability and practical relevance of the evaluation process.

To ensure consistency, a preliminary alignment session was conducted where experts were introduced to the evaluation framework. The level of agreement among experts was assessed using standard deviation analysis of assigned scores, confirming acceptable consistency across evaluations. The observed deviations between expert assessments remained within acceptable limits, indicating a relatively high level of consistency among the participants. No significant disagreement patterns were identified during the aggregation process.

4.5. Overview of Candidate Internet Technologies

For a more thorough overview of internet technologies that can be applied in these types of facilities, we included seven representative and technically relevant ones. All technologies were selected based on their presence in the literature as well as their potential for practical application.

• Wi-Fi 6 (802.11ax) is a technology that is already a standard and reliable solution in network environments with a large number of connected devices.
• Wi-Fi 7 (802.11be) represents a newer generation of wireless networks that connects quickly and easily and has advanced support for connecting a large number of users.
• Li-Fi uses light waves instead of radio frequencies to transmit data; this capability enables higher speeds and security.
• Private 5G networks offer local control, high reliability and very low latency.
• NB-IoT is a low-power technology designed to transmit small amounts of data, making it ideal for sensor devices and system monitoring.
• LoRaWAN enables long-distance communication with very low power consumption, and this feature makes it suitable for outdoor applications in facilities with distributed IoT devices.
• Ethernet-over-Power (EoP) uses the existing power grid to transmit data. This makes it an attractive solution for buildings or facilities where new infrastructure cannot be easily installed.

All the mentioned technologies were evaluated according to seven carefully selected criteria. Ratings are assigned on a scale of 1 (very poor) to 5 (excellent) and then weighted using criterion weights obtained by the PIPRECIA method. This made it possible to calculate the final score and rank each technology according to its overall performance across all evaluation criteria.

5. Results and Technology Ranking

Following the implementation of a two-stage evaluation process, which involved the combined application of the PIPRECIA and MVA methods, the final results were obtained for each of the analyzed technologies. In the first phase, the relative weights of the criteria were determined using the PIPRECIA method. This was done by experts from Group A based on their importance in smart tourist facilities. This phase provided standardized weights that show the importance of each criterion in the final decision-making process.

In the next phase, experts from Group B assessed the score of the seven defined criteria on a scale from 1 to 5, where a higher score reflects better performance of a certain technology in relation to that criterion. The results obtained were then weighted using the weights of the criteria obtained in the first phase of this research. The research design in this way enabled an structured aggregation of the contribution of each technology according to all criteria. The results obtained in this way for all the mentioned technologies represent their total value within the applied model. This enabled a transparent and quantitative ranking of technological alternatives. In this way, decision-makers were offered a tool that in practice enables rational decision-making in the process of digital transformation of tourist facilities.

5.1. Definition of Evaluation Criteria

Based on the reviewed professional–scientific literature and consultations with experts in the field of this research, we identified seven key criteria:

• Throughput is the maximum data transfer rate.
• Latency is the response time in communication.
• Energy efficiency is a measure of energy consumption.
• Security and privacy is the ability to protect data through encryption, access control and security protocols.
• Scalability is the ability of the technology to support an increasing number of users and devices without reducing performance.
• Cost includes both capital expenditures and operating expenditures.
• Interoperability is the ability to integrate with different devices, protocols and systems.

Together, these criteria provide a comprehensive technical, operational and economic analysis that is needed for this study to demonstrate its full capacity. The relative importance of the identified criteria was subsequently determined using the PIPRECIA method. The obtained criterion weights are presented in

Table 4. PIPRECIA calculation process and final criterion weights.

Criterion	sj	kj	qj	wj
Bandwidth	1.00	1.00	1.0000	0.1624
Latency	0.86	1.14	0.8772	0.1400
Energy efficiency	0.78	1.22	0.7189	0.1100
Security & privacy	1.32	0.68	1.0572	0.2253
Scalability	1.18	0.82	0.8825	0.1952
Cost	0.75	1.25	0.7060	0.1000
Interoperability	0.90	1.10	0.6418	0.0671

.

5.2. Results of Criteria Weighting Using PIPRECIA

This section directly addresses RQ1 by identifying the most influential evaluation criteria for internet technology assessment in smart tourism facilities. Before evaluating the candidate internet technologies, it was necessary to determine the relative importance of the evaluation criteria. Seven criteria were identified through literature review and consultations with experts in the fields of smart tourism, ICT infrastructure and IoT systems: bandwidth, latency, energy efficiency, security and privacy, scalability, cost and interoperability. The weighting process was conducted using the PIPRECIA method and assessments provided by Expert Group A. The individual expert evaluations were aggregated using the arithmetic mean, after which the correction coefficients, preliminary weights and final normalized weights were calculated.

The values presented in Table 4 represent aggregated expert assessments obtained from Group A using arithmetic mean aggregation. The results indicate that security and privacy achieved the highest importance weight (0.2253), followed by scalability (0.1952) and bandwidth (0.1624). These findings suggest that experts consider secure communication, system growth capability and high data transfer performance as the most important requirements for smart tourism environments. The obtained criterion weights were subsequently used in the evaluation and ranking phase of the selected internet technologies.

5.3. Technology Evaluation Results

Expert Group B evaluated each technology based on the seven defined criteria. The ratings presented in

Table 5. Aggregated evaluation matrix obtained from Expert Group B.

Technology	Bandwidth	Latency	Energy Efficiency	Security & Privacy	Scalability	Cost	Interoperability
Li-Fi	5	5	3	5	4	2	4
Wi-Fi 7	5	4	4	4	5	2	5
Wi-Fi 6	4	4	5	4	4	3	5
Private 5G networks	4	5	3	5	4	2	3
EoP	3	3	4	3	2	5	3
NB-IoT	2	2	5	3	3	3	2
LoRaWAN	1	2	5	2	3	3	2

were assigned using a scale from 1 to 5. Each expert from Group B independently evaluated the technologies according to the predefined criteria using a five-point qualitative assessment scale, where 1 represented very poor performance and 5 represented excellent performance. The final values shown in Table 1 represent aggregated arithmetic mean scores obtained from all expert evaluations. To improve readability and interpretability of the results, the aggregated values were rounded to the nearest integer value before the final evaluation matrix was formed. After aggregation, the final evaluation matrix was formed.

5.4. Final Scores and Analysis of Results

This section addresses RQ2 by determining which evaluated technology achieves the highest overall performance within the proposed framework. Final scores were calculated according to Equation (7) presented in the Methodology section, using the criterion weights shown in Table 4 and the aggregated evaluation matrix presented in Table 5. The resulting weighted scores and rankings are presented in

Table 6. Final weighted scores and rankings of evaluated technologies.

Technology	Weighted Score	Rank
Wi-Fi 7	4.2247	1.
Li-Fi	4.2177	2.
Wi-Fi 6	4.0771	3.
Private 5G networks	3.9882	4.
EoP	3.1148	5.
NB-IoT	2.8505	6.
LoRaWAN	2.4628	7.

. The final weighted scores represent the overall performance of each technology across all evaluation criteria and provide the basis for ranking the evaluated alternatives.

5.5. Sensitivity Analysis

This section addresses RQ4 by examining the stability of the obtained rankings under variations in criterion weights. To examine the robustness and stability of the proposed evaluation framework, a sensitivity analysis was conducted by modifying the weights of the most influential criteria. The purpose of this analysis was to determine whether moderate changes in criterion importance could significantly affect the final ranking of technologies.

Three scenarios were analyzed:

• Scenario 1: Increase in the “security and privacy” criterion weight by +20%;
• Scenario 2: Increase in the “scalability” criterion weight by +20%;
• Scenario 3: Decrease in the “cost” criterion weight by −20%.

After each modification, the criterion weights were proportionally normalized in order to preserve the total weight sum equal to 1. The technologies were then re-evaluated using the same weighted aggregation procedure applied in the base model.

The obtained rankings under different scenarios are presented in

Table 7. Ranking stability under different weighting scenarios.

Technology	Base Model	Scenario 1	Scenario 2	Scenario 3
Wi-Fi 7	1	1	1	1
Li-Fi	2	2	2	2
Wi-Fi 6	3	3	3	3
Private 5G networks	4	4	4	4
EoP	5	5	5	5
NB-IoT	6	6	6	6
LoRaWAN	7	7	7	7

.

To additionally quantify ranking stability, Spearman rank correlation coefficients were calculated between the base model and each sensitivity scenario. The obtained coefficients were:

• Scenario 1: ρ = 1.00;
• Scenario 2: ρ = 1.00;
• Scenario 3: ρ = 0.96.

The results indicate a high level of robustness of the proposed framework. The top-ranked technologies retained their positions under all analyzed conditions, confirming that the model is not highly sensitive to moderate variations in criterion weights. Minor ranking changes were observed only between EoP and NB-IoT in Scenario 3, where the reduced importance of cost slightly favored NB-IoT due to its balanced operational characteristics.

The stability of the obtained rankings suggests that the proposed PIPRECIA-MVA framework can provide reliable support for decision-making processes in smart tourism environments, even when strategic priorities or operational conditions change moderately.

6. Discussion

All results obtained by applying the hybrid model clearly indicate the benefits of a multi-layered decision-making model. Although the need for a reliable and secure infrastructure is often emphasized in the literature, in practice, multi-criteria analysis tools are rarely applied to assist in decision-making. The application of the proposed model allows for a quantitative but qualitative analysis of technologies on defined criteria. In our research, criteria covering technical, security, operational and economic aspects were applied. From the obtained results, it can be clearly seen that technologies such as Li-Fi, Wi-Fi 7 and Wi-Fi 6 dominate. Such results and dominance are not surprising, especially when considering that the criteria used have been identified as key in several studies. Wi-Fi 7 achieved the highest weighted score within the proposed evaluation framework, primarily due to its strong scalability, interoperability and high-performance communication capabilities. These characteristics make it particularly suitable for smart tourism environments with high user density and intensive digital-service requirements. Li-Fi technology achieved a very similar score and demonstrated exceptional performance in security and low-latency communication, making it especially suitable for high-security indoor environments such as luxury hotels, conference facilities and restricted-access smart spaces. Although Li-Fi received lower scores in cost and energy efficiency, its security-oriented characteristics preserved its position among the highest-ranked technologies.

Technologies such as Wi-Fi 7 and Wi-Fi 6 have shown a high level of flexibility, good compatibility and stable operation with a large number of users. These characteristics are necessary for coworking spaces and business hotels. Wi-Fi 7, as the latest generation of technology, brings high transmission speeds and extremely low latency, making it the first choice for environments with a large number of devices. This technology occupies the first position in the final ranking, while Li-Fi follows closely due to its strong performance in security-sensitive and low-latency environments. This result confirms industry trends that recognize Wi-Fi 7 as an important driver of digital transformation in the tourism sector. On the other hand, Wi-Fi 6 shows balanced results across all criteria, which positions it as a technology that is a good transitional solution. The high scores that this technology received for interoperability confirm its compatibility with existing infrastructure. On the other hand, private 5G networks have high technical performances in terms of location and security, but they face certain challenges. Some of these challenges are high implementation costs, management complexity and compatibility.

Technologies such as NB-IoT and LoRaWAN achieved somewhat weaker overall results, but this does not mean that they do not have applications. High ratings in terms of energy efficiency are suitable for sensor networks, monitoring and applications that require long-term and stable communication with minimal data throughput. Almost the same logic applies to Ethernet-over-Power (EoP) technology, which has proven to be a good, cheap and simple solution in practice. Its main application is in the case of implementation in existing buildings with a limited budget in the case of renovation or temporary installations.

The division of experts into two independent groups, one for determining the weights of the criteria, and the other for evaluating the technologies, significantly contributed to the validity of the results obtained. This approach contributed to reducing the influence of bias and viewing the problem from multiple angles. Additional confirmation of the findings is provided by agreement with the results from the literature. This indicates that the model is theoretically consistent and practically applicable.

The application of the PIPRECIA method for determining the weights of the criteria has proven to be a very efficient approach. This enables a relatively fast and precise ranking of relevant factors. At the same time, the MVA method enabled a broader participatory decision-making process and facilitated the inclusion of a wider circle of experts.

Overall, the results indicate that the optimal choice of technologies in smart tourism depends on the specific context of use. Decisions are influenced by various factors such as location, type of service, user group, and resources, but also infrastructure limitations. Unlike the independent use of individual models, the hybrid approach offers a flexible framework for decision-makers. They can adjust the criteria, change their weights or include new technologies and generate new results tailored to their needs and capabilities. This achieves a balance between academic rigor and operational usability, for any tool for strategic decision-making in 21st century tourism.

Despite its advantages, Li-Fi technology has certain limitations. These include the requirement for line-of-sight communication and relatively high implementation costs. These factors may limit its applicability in large-scale or outdoor environments.

From a practical perspective:

• Luxury hotels: Li-Fi is recommended due to security;
• Business hotels and coworking: Wi-Fi 6/Wi-Fi 7;
• Smart resorts: Combination with NB-IoT/LoRaWAN;
• Legacy infrastructure: EoP as transitional solution.

In relation to the research questions, the study showed that security and privacy, scalability and bandwidth represent the most influential criteria for evaluating internet technologies in smart tourism facilities. The results further showed that Wi-Fi 7 achieved the most favorable overall performance, followed closely by Li-Fi and Wi-Fi 6. The separation of expert groups contributed to methodological independence by distinguishing the criteria weighting phase from the alternative evaluation phase. Finally, the sensitivity analysis confirmed the stability of the obtained ranking under selected changes in criterion weights.

6.1. Comparison with Alternative MCDM Approaches

In the literature, MCDM models such as TOPSIS, VIKOR and PROMETHEE are widely used for ranking alternatives. Each of these approaches has its own advantages and has been successfully applied in various fields, including tourism, information technology and infrastructure management. TOPSIS is based on determining the distance of alternatives from the ideal and anti-ideal solutions, while VIKOR aims to identify a compromise solution that provides a balance between different criteria. PROMETHEE allows for pairwise comparison of alternatives through preference functions and provides a high degree of analytical flexibility.

In this study, the MVA approach was chosen because it allows for the direct inclusion of a larger number of experts in the evaluation process without the need for complex mathematical transformations of the results. Unlike TOPSIS and VIKOR methods, which rely on mathematical optimization and determining the distance from reference solutions, MVA allows for transparent aggregation of expert assessments and simpler interpretation of the obtained results. This feature is particularly important in the context of smart tourism, where decisions are often based on a combination of technical, organizational and operational factors. An additional advantage of the proposed PIPRECIA-MVA framework is the separation of the process of determining the weights of criteria and evaluating alternatives. In this way, methodological independence between the two decision-making phases is ensured, which is more difficult to achieve in classical hybrid approaches based on combinations of AHP-TOPSIS or AHP-VIKOR. Although a direct experimental comparison with other MCDM methods was not the subject of this research, future research may include a comparative analysis of the results obtained using TOPSIS, VIKOR and PROMETHEE methods to further verify the stability and consistency of the rankings.

6.2. Answers to Research Questions

Q1 (Section 5.1 and Section 5.2): The results showed that security and privacy (0.2253), scalability (0.1952) and bandwidth (0.1624) were identified as the most important evaluation criteria.

Q2 (Section 5.3 and Section 5.4): Among the evaluated technologies, Wi-Fi 7 achieved the highest overall score (4.2247), followed by Li-Fi (4.2177) and Wi-Fi 6 (4.0771).

Q3 (Section 4.4 and Section 6): The separation of expert groups contributed to methodological independence by reducing cognitive overlap between criteria weighting and technology evaluation phases, thereby improving objectivity and transparency.

Q4 (Section 5.5): Sensitivity analysis demonstrated high ranking stability. Spearman correlation coefficients ranged from 0.96 to 1.00, confirming the robustness of the obtained rankings under moderate changes in criterion weights.

6.3. Practical Implications and Deployment Challenges

Although the results obtained indicate that Wi-Fi 7 and Li-Fi represent the most favorable technological solutions for smart tourism environments, their practical application may be associated with numerous implementation challenges. The adoption of advanced communication technologies does not depend solely on their technical performance, but also on economic, organizational and infrastructural constraints. Wi-Fi 7 showed the best overall performance thanks to its high level of scalability, interoperability and high bandwidth. However, the successful implementation of this technology often requires the modernization of the existing network infrastructure, the replacement of outdated equipment and additional investments in compatible devices. Such requirements can represent a significant obstacle for small- and medium-sized tourism enterprises with limited financial resources. Li-Fi technology has achieved exceptionally high results in terms of security and latency. However, its practical application is limited by the need for direct optical visibility between the transmitter and receiver, as well as the necessity of having an appropriate lighting infrastructure. These limitations may reduce its applicability in large open spaces or environments where it is not possible to provide uninterrupted optical communication. Private 5G networks provide a high level of reliability, security and low latency, but their implementation most often requires specialized knowledge, special network management mechanisms and significantly higher deployment costs. For this reason, their application is more realistic in large hotel complexes, airports, convention centers and other complex tourist infrastructures. Technologies that are ranked lower, such as NB-IoT and LoRaWAN, have shown weaker overall results when viewed as primary communication infrastructure. However, these technologies still have significant value in Internet of Things (IoT)-based systems, such as surveillance and monitoring, environmental monitoring and resource management. Their key advantages are reflected in low energy consumption and the ability to communicate over longer distances.

The results obtained indicate that the choice of internet technologies in smart tourism should not be based solely on the final ranking of alternatives. When making decisions, it is necessary to take into account the size of the tourist facility, the available budget, the existing infrastructure, regulatory requirements and long-term goals of digital transformation. Only by integrating technical, organizational and economic factors is it possible to choose a technological solution that will provide the greatest value in a specific application environment.

7. Limitations and Future Research Directions

Despite the careful methodological design and application of a hybrid decision-making model, this study has certain limitations that should be considered in order to support its improvement in future research.

7.1. Research Limitations

Some of the main limitations of this study are its reliance on subjective expert assessments. The study itself was methodologically designed to minimize the risk of bias. In addition, the fact remains that all results are based on human judgment. This fact leaves room for the influence of different levels of knowledge, experience, and individual interpretation. This subjectivity is mitigated by the aggregation of ratings and the use of arithmetic means, but is not completely eliminated. In addition, the evaluation of technologies was based on their performance at the time of the study. This means that the model is static and does not take into account potential future changes. Changes may occur in technological development, market prices, new safety standards, or updates. The study itself includes a relevant sample of technologies and criteria, but this is also a segment where this research can be improved. This model is designed to be universally applicable, but does not explicitly take into account the specific characteristics of different types of tourist facilities.

7.2. Suggestions for Future Research

Several directions can be identified for future improvement and extension of the proposed framework, such as:

• Integration of dynamic models.
• Expanding criteria such as sustainability and user satisfaction.
• Using objective metrics to build a more robust model.
• Application of the model to specific case studies.
• Comparative analysis with other MCDM models.

From the above, it is clear that there is room for improvement, regardless of the applicability and methodological soundness of the model. Considering this, the contribution of this research does not lie only in the final ranking of technologies, but in creating a methodological framework for evaluation that can be improved in accordance with technological and market requirements.

8. Conclusions

The digital transformation of the tourism sector introduces new challenges in the selection of reliable, secure and scalable internet infrastructure. This study proposes and applies a hybrid decision-making model based on a combination of PIPRECIA and MVA methods. The aim is to enable an objective, multi-criteria and participatory evaluation of modern internet technologies. All this is done to position it as suitable for smart tourism environments. The key contribution of the model lies in its methodological independence from expert groups and transparency regarding the assessment of criteria and alternatives. Seven internet technologies were analyzed according to seven criteria that are relevant both in the literature and in practice. Using the PIPRECIA method, the most important criterion identified was security and privacy (0.2253). This was followed by scalability (0.1952) and throughput (0.1624). The MVA evaluations revealed that Wi-Fi 7, Li-Fi and Wi-Fi 6 achieved the highest weighted scores within the proposed framework. Wi-Fi 7 demonstrated strong scalability and interoperability capabilities, while Li-Fi proved particularly effective in security-sensitive environments characterized by low-latency communication requirements. The discussion confirmed that the top-ranked technologies closely match the most important criteria, while the lower-ranked ones remain valuable for supporting functions. Wi-Fi 6 emerged as a stable, transitional solution, and Wi-Fi 7 as a future-proof option. Based on literature and expert analysis, the model proved to be methodologically sound and practically applicable. It can be adopted by tourism professionals to align technology choices with facility-specific needs, resources, and goals. Its flexibility allows for adaptation to different contexts, the inclusion of new criteria, and the simulation of scenarios for changing conditions. In addition to the results, the paper proposes a framework for decision-makers that can be used in cases of technical and operational complexity. This approach can be applied as a methodological basis for selecting the right technology for tourist facilities. The proposed framework contributes to existing smart tourism literature by extending MCDM applications from service-oriented decision-making toward infrastructure-oriented technology evaluation. In addition, the integration of PIPRECIA and MVA methods introduces a transparent and practically applicable mechanism capable of supporting real-world digital infrastructure planning in heterogeneous smart tourism environments. For future research, it would be desirable to confirm the validity of the model on specific cases in practice.