Bridging the Accessibility Gap in Green Tourism: A Framework for Sustainable Integration of Specialised Off-Road Wheelchair Services with Public Transport Networks

Abstract

Reducing social exclusion through technology is a key challenge for sustainable development, particularly within the context of accessible tourism. This study, as part of the “MOUNTAINS WITHOUT BARRIERS” project, addresses this issue by aiming to identify optimal locations for specialized all-terrain wheelchair rental stations in mountainous regions. The primary purpose is to ensure these locations are seamlessly integrated with existing local transport systems, fostering genuine accessibility. A dedicated methodology was developed to analyze the spatial integration of the accessible trail network with the transport system in the Beskid Agglomeration. The analysis, conducted using Geographic Information System (GIS) tools, considers access via both individual transport and public transport, with a clear emphasis on prioritizing the latter to promote sustainable mobility patterns. Applying this approach, the study identified potential station locations that are not only conveniently situated at trailheads but are also highly accessible via public transport. The main finding indicates that strategic placement can significantly minimize the necessity for private car usage. Integrating tourism infrastructure with public transport is crucial for increasing the real-world accessibility of mountain areas for people with disabilities. Furthermore, the results and methodology provide valuable recommendations that can serve as a practical input for Sustainable Urban Mobility Plans (SUMP).

1. Introduction

The freedom of movement is a fundamental human need, enabling individuals to participate in social, professional, and recreational activities [1,2,3,4,5]. Travel, in its many forms, enriches human experience, and modern society has a growing responsibility to ensure that these opportunities are available to everyone [6,7,8]. Fulfilling the diverse needs of all individuals without exclusion, regardless of their physical abilities, is a cornerstone of an inclusive and sustainable community [9,10,11]. For individuals with mobility impairments, the wheelchair has been a transformative invention, evolving from a basic mobility aid to a technologically advanced device. Contemporary technology, including lightweight materials, advanced batteries, and digital navigation systems, now offers unprecedented possibilities to reduce social exclusion by opening previously inaccessible environments, such as challenging natural landscapes [12,13,14,15,16,17].

This paradigm of enhanced accessibility is the foundation of the “MOUNTAINS WITHOUT BARRIERS” project, initiated by the Silesian University of Technology and funded by the National center for Research and Development under the ‘Things for people’ competition. The project aligns with the goals of the Polish government’s “Accessibility Plus” program (2018–2025), which aims to improve the accessibility of public services and spaces. The core objective of “MOUNTAINS WITHOUT BARRIERS” is to increase the availability of the tourist offer in non-urbanized areas for people with physical functioning difficulties. The project encompasses three main components: the development of a specialized off-road vehicle; an integrated digital platform for route planning, navigation, and real-time support, including safety alerts from services like GOPR (Mountain Volunteer Search and Rescue); and a methodology for preparing and verifying the accessibility of tourist infrastructure. The developed vehicle is shown in Figure 1.

Figure 1. The appearance of the concept model of the developed specialized off-road vehicle.

The indicated wheelchair has the following parameters:

• total length (including footrest) 200 cm;
• front wheelbase 100 cm;
• rear wheelbase 80 cm;
• height 142 cm (up to the anti-rollbar frame), 146 cm (including the emergency brake mechanism);
• clearance of 20 cm (for the currently used wheels).

In addition, the developed wheelchair is characterized by modular construction, i.e., it is possible to install other wheels (e.g., adapted to different challenges in the area) and fully adapt the control to the required user and their disability.

While the development of the vehicle and planning system is foundational, true accessibility depends on a critical logistical challenge: ensuring users can conveniently and sustainably reach the starting points of accessible trails where they can rent the specialized vehicles. Therefore, the purpose of this paper is to extend the project’s analysis to the integration of accessible mountain trails with both public and individual transport systems.

For users with disabilities, public transport must meet their specific needs and expectations. While general amenities such as air conditioning, smartphone chargers, or free Wi-Fi improve comfort for all passengers, the core issue for people with disabilities is different. For this group, accessibility is not simply another critical aspect it is the fundamental prerequisite for travel [18,19].

True accessibility means removing fundamental barriers and includes [20,21,22,23]:

• Physical access through a modern fleet with low-floor vehicles, functional ramps, and adequate space for wheelchairs or mobility aids.
• Information access through clear audio and visual announcements of route information, which is essential for passengers with sensory impairments.

Only when these non-negotiable requirements are met can other features contribute to a genuinely inclusive and high-quality travel experience.

This study focuses on a spatial analysis aimed at promoting sustainable access to trailheads, which are the prospective locations for the specialized off-road wheelchair rental stations.

Although the literature contains various spatial analysis methods for optimizing the location of infrastructure points, such as public transport stops or electric vehicle charging stations, this research presents a novel contribution [24,25]. While accessibility is widely discussed in the literature, it lacks a single, universal definition [26,27,28,29]. It has been described as the ability to use transport services to reach destinations or, more specifically, the ease of accessing public transport facilities and completing a journey. From the perspective of disability studies, these definitions must be interpreted through a practical lens. “Ease of access” is not about convenience but about the complete absence of physical, sensory, and informational barriers throughout the entire travel chain.

To fully grasp this concept, accessibility is often broken down into four interrelated components: transport, land use, time, and the individual [30,31,32].

• The transport and land use components concern the ease of travel between points and the attractiveness of destinations. For a person with a disability, this means the entire path from their front door to the bus stop, onto the vehicle, and to the final destination must be a continuous, barrier-free chain [33,34,35].
• The temporal component refers to service availability. This is not just about the frequency of service, but the frequency of accessible service, as not all vehicles or times may be suitable [36,37,38].
• The individual component, focusing on the unique needs and characteristics of the passenger, is paramount. For people with disabilities, this component acts as the central filter through which all other aspects of accessibility are experienced and judged.

Numerous methods exist to measure transport accessibility. Outlines five primary approaches: infrastructure equipment, distance, cumulative (isochronous), potential, and personified. When applied to disability research, these methods require a specific focus. For instance, the potential model, which is popular in general analyses, must be adapted. The “attractiveness” of a destination is effectively zero if it is physically inaccessible, regardless of its size or theoretical proximity.

Because of this, the personified measurement method, which directly incorporates the individual’s unique characteristics and abilities, is the most relevant and powerful tool for this study. It enables an analysis that transcends abstract metrics to reflect the lived reality of navigating the transportation system with a disability. Recent literature highlights a trend towards increasingly sophisticated, data-driven models for accessibility and pedestrian dynamics analysis. For instance, advanced approaches now leverage experimental data and complex metrics like lane entropy or angular momentum to capture spatiotemporal dynamics [39,40,41] within pedestrian flows, particularly under specific constraints such as public health scenarios [42]. These methods provide high-resolution insights into behavioral self-organization and system efficiency. However, while these highly quantitative models are powerful for analyzing crowd movement, they often operate at a level of abstraction that may not capture the specific, qualitative barriers faced by individuals with disabilities. Factors like the absence of a single ramp, the width of a door, or the lack of a low-floor vehicle are binary, “go/no-go” conditions that are not easily represented in complex behavioral metrics.

In contrast to these existing models, this study introduces a method that uniquely integrates the qualitative aspects of tourism, the specific requirements of people with special needs, transport demands, and the technical capabilities of the specialized vehicle, addressing a complex gap in current accessibility research. It operationalizes the ‘personified measurement’ by focusing on a hierarchy of practical, non-negotiable needs, thus bridging the gap between theoretical accessibility modeling and real-world, inclusive service planning.

In contrast to existing models, this study introduces a method that uniquely integrates the qualitative aspects of tourism, the specific requirements of people with special needs, transport demands, and the technical capabilities of the specialized vehicle, addressing a complex gap in current accessibility research.

This article addresses the challenge of finding optimal locations for specialized wheelchair rental points by integrating public and individual transport access with high tourist attractiveness, while acknowledging the economic constraint of a limited number of feasible sites. The proposed method allows for the ranking of these potential locations. The article also applies the authors’ previous research [43,44] on analyzing mountain trails to reduce the number of field tests performed.

A case study was conducted in the Beskid Agglomeration (Silesian Voivodeship, Poland) to validate the proposed methodology, successfully identifying several potential locations for rental points. This versatile method can be adapted for other tourist regions and can be used to site services for different modes of transport in remote or hard-to-reach areas.

2. Materials and Methods

In order to identify optimal locations for specialized off-road wheelchair rental points, a methodological framework was developed. The proposed approach relies on spatial analysis supported by open-source GIS software such as QGIS 3.40 [45]. Input data may originate from open data platforms (e.g., OpenStreetMap [46]) as well as from national transport institutions or police databases and should be stored in a georeferenced format to enable effective visualization and processing. This structure ensures replicability of the procedure in different regions and facilitates integration with existing mobility datasets. A schematic representation of the method is presented in Figure 2.

Figure 2. Scheme of the proposed method.

The developed method consists of four main components. The first component is input data collection, which specifies the type of data required and their potential sources.

• identification of public transport stops—$PT$—Identification of all potential public transport stops, including transfer hubs, within the study area. Potential data sources: OpenStreetMap, General Transit Feed Specification (GTFS);
• identifcation of mountain trails—$MT$—All mountain trails, walking paths, and linear information on tourist routes present within the study area. Potential data sources: OpenStreetMap and local repositories (in Poland, for example, the Polish Tourist and Sightseeing Society, PTTK). Incorporating these datasets allows for the assessment of potential synergies between transport accessibility and recreational opportunities, which is particularly relevant in mountain regions;
• identification of parking lots—$P$—Identification of all available parking facilities. Potential data sources: OpenStreetMap and local spatial development plans. This stage is essential for evaluating the integration of private vehicle use with public and recreational transport options, enabling a more comprehensive understanding of intermodal accessibility within the study area. A key element of this step was an initial audit of the inventoried parking facilities. The primary objective of this GIS-based audit was to verify public accessibility and eliminate private or functionally inaccessible objects (e.g., those located behind entry barriers without public access). This verification was performed using OpenStreetMap data attributes and cross-referenced with satellite imagery. Therefore, the phrase “positive evaluation” signifies that a facility was confirmed to be publicly accessible based on available data. A more detailed assessment, including crucial criteria such as the presence of designated accessible parking spaces, surface quality, and fee structures, would be necessary during a subsequent field verification stage. To formalize this process, a recommended checklist is provided in Appendix A;
• identification of mountain shelters—$MS$—Location of mountain shelters or rest facilities along tourist routes. In Poland, these facilities operate within a legal framework, which ensures their functionality and usability. Potential data sources: OpenStreetMap and local repositories (in Poland, for example, the Polish Tourist and Sightseeing Society, PTTK). Including these elements in the analysis provides a reliable basis for evaluating service availability and user comfort, which are critical factors in accessibility studies of mountain areas.

In the second stage of the method, an analysis of tourist attractiveness in relation to transport accessibility was conducted, with primary emphasis placed on public transport in order to promote sustainable development. Individual transport is considered in subsequent steps of the method. A key assumption is the limited number of planned rental stations, which should be located in areas with high tourist potential to maximize their use by prospective clients. This approach reflects the principle of resource efficiency, ensuring that infrastructure investments are concentrated where demand is most likely to emerge.

The analytical procedure begins with the delineation of 50 m buffers around each identified mountain trail $MT$:

$$MT=\left\{{mt}_1,{mt}_2,\dots ,\left.{mt}_n\right\}\right.$$

(1)

The 50 m threshold was not chosen arbitrarily. It was established based on consultations with an expert panel composed of potential end-users within the “MOUNTAINS WITHOUT BARRIERS” project. The experts emphasized that the path from a public transport stop to a trailhead in mountainous terrain often lacks basic infrastructure, such as paved sidewalks, and can present significant micro-barriers (e.g., curbs, steep gradients, uneven ground). Therefore, a minimal distance is a critical prerequisite, ensuring that the rental station is practically at the trailhead itself, eliminating a challenging “last few meters” journey for users. Subsequently, public transport stops are identified within these buffer zones $PT:$

$$PT=\left\{{pt}_1,{pt}_2,\dots ,\left.{pt}_n\right\}\right.$$

(2)

The purpose of this analysis is to identify points that provide access to one, two, and ultimately three or more different trails. Such an approach makes it possible to highlight the most strategic locations from which tourists can begin their trips across a diversified network of routes. Identification of public transport stops in the vicinity of each mountain trail is therefore carried out ${mt}_i\in MT$ a buffer with a radius of 50 m is created and designated as ${B(mt}_i, 50 \mathrm{m})$. The set of public transport stops ${Pt}_i$ located within the buffer zone of the trail ${mt}_i$ is defined as:

$${PT}_i=\left\{{pt}_j \in PT | {pt}_j\cap B\left({mt}_i, 50 \mathrm{m}\right)\ne \left.0\right\}\right.$$

(3)

The set of public transport stops providing access to at least three different trails $({PT}_{{MT}_3}$):

$${PT}_{{MT}_3}=\left\{pt \in PT |\exists i,j,k:i\ne j\ne k\ne i,pt\in {PT}_i\wedge pt\in {PT}_j\wedge pt\in \left.{PT}_k\right\} \right.$$

(4)

To eliminate data redundancy resulting from the presence of directional stops at the same location, a reduction procedure was implemented. This procedure assumes the removal of stops located less than 20 m from each other, retaining a single representative point for a given area. The set of ${PT}_{{MT}_3}$ the most tourist-attractive set is filtered to remove stops located too close to one another, in order to obtain a refined dataset ${PT}_{{FMT}_3}$:

$${PT}_{{FMT}_3}=\left\{pt \in {PT}_{{MT}_3} |\forall {pt}_j\in {PT}_{{MT}_3}, i\ne j ⟹d({pt}_i,{pt}_j)\left.\ge 20 \mathrm{m}\right\} \right.$$

(5)

where

$d({pt}_i,{pt}_j)$—this refers to the Euclidean distance between stops.

The final element of this stage is the verification of accessibility to mountain shelters $MS:$

$$MS=\left\{{ms}_1,{ms}_2,\dots ,\left.{ms}_n\right\}\right.$$

(6)

from the selected potential rental station locations, the distance along the road and trail network to the nearest mountain shelter is assessed. A maximum distance of 5 km was adopted, corresponding to approximately 1.5 h of travel. This value was determined in collaboration with the project’s expert panel. It reflects a crucial psychological and safety factor, especially given the novelty of the specialized vehicle technology. A 5 km (approx. 1.5 h) journey was identified as a distance that inspires user confidence, mitigates “range anxiety” related to battery life, and ensures that a safe haven (a mountain shelter) is always reachable within a comfortable timeframe should technical issues or sudden weather changes occur. The resulting set of locations is recommended for further analysis in the third component of the method:

$$L=\left\{pt \in {PT}_{{FMT}_3} |{min}_{ms\in MS}{d}_{net}\left(pt,ms\right)\left.\le 5 \mathrm{k}\mathrm{m}\right\} \right.$$

(7)

where

$d_{net}\left(pt,ms\right)$—denotes the network distance from potential rental station locations to mountain shelters.

The third component of the method focuses on verifying the accessibility of individual transport for the locations recommended thus far. For each point identified in the previous stages, a spatial analysis is carried out to locate publicly accessible parking facilities $P$:

$$P=\left\{p_1,p_2,\dots ,\left.p_n\right\}\right.$$

(8)

within a radius of 500 m:

$$L_F=\left\{l \in L |\exists p\in P:d(l,p)\left.\le 500 \mathrm{m}\right\}\right.$$

(9)

where

$d(l,p)$—denotes the Euclidean distance between location l and parking facility p.

The 500 m radius is based on established transport planning principles for maximum acceptable walking distance, adapted for the specific user group through expert panel consultations. This distance was considered a reasonable maximum for individuals who might arrive by car but still face mobility challenges when moving from the parking facility to the trailhead rental point. A key element of this step is the audit of the inventoried parking facilities $L_F$, which allows for the elimination of private or inaccessible objects (e.g., those located behind entry barriers without public access). In this way, only parking facilities that can realistically be used by potential tourists are taken into account. This filtering process strengthens the validity of the analysis by ensuring that accessibility assessments are based exclusively on functionally available infrastructure.

For the obtained locations, the final analysis—Qualitative Assessment of the Remaining Stops—is carried out using a multi-criteria method. Five evaluation criteria for the stops were proposed, taking into account both the objectives of the project and the needs of people with special requirements. This stage ensures that the final recommendations are not only spatially and functionally justified but also aligned with principles of inclusivity and user-oriented design. Criteria:

• K1: The number of services operated at a given stop within the specified time frame 06:00–20:00;
• K2: The occurrence of services on both weekdays and weekends. This criterion captures the temporal availability of public transport, ensuring that accessibility is maintained not only during peak commuting periods but also throughout leisure-oriented travel times;
• K3: Coverage (whether a connection with the city center exists)—this is advantageous from the perspective of traveling from the municipal center to the starting point of a given mountain trail. The presence of such connections enhances the strategic importance of the stop, as it links core urban areas with key recreational destinations;
• K4: The percentage of services operated with low-floor vehicles—this is the most important element, as it determines the technical feasibility of access. Stops not served by low-floor vehicles are excluded at this stage. This criterion directly addresses the needs of people with reduced mobility and ensures that the proposed solutions comply with accessibility standards. This criterion acts as a non-negotiable prerequisite; therefore, any location with a score of 0% for K4 is automatically assigned a final result of 0 and disqualified from the ranking, as it is considered fundamentally inaccessible;
• K5: The presence of a bus shelter—this is an additional factor that improves passenger comfort in the event of deteriorating weather conditions. Although not essential for basic accessibility, this element enhances the overall quality of service and can influence user satisfaction and willingness to use public transport.

The criteria weights were established through a structured consensus-building process with an expert panel, which included project researchers and, crucially, representatives of the end-user community (people with disabilities). This participatory approach ensures that the weights reflect the real-world priorities and non-negotiable needs of the target group. The panel identified a clear hierarchy of needs:

• Fundamental Prerequisites (Weight 0.3): The highest weight was assigned to K4 (percentage of low-floor vehicles) and K1 (number of services). Criterion K4 is a binary accessibility factor; without low-floor vehicles, the service is unusable. Criterion K1 reflects the fundamental availability and flexibility of travel options.
• Key Service Usability (Weight 0.2): K2 (weekend service) was rated as the next most important, as mountain tourism is predominantly a leisure activity, making weekend availability essential.
• Comfort and Convenience (Weight 0.1): The lowest weight was given to K3 (city center connection) and K5 (bus shelter). While valuable, these were deemed secondary factors enhancing comfort rather than enabling basic access.

Weights were proposed for the criteria, as presented in Table 1.

Table 1. Weights for the multi-criteria analysis.

Criteria	Weights
K1: The number of services operated at a given stop within the specified time frame	0.3
K2: The occurrence of services on both weekdays and weekends	0.2
K3: Coverage	0.1
K4: The percentage of services operated with low-floor vehicles	0.3
K5: The presence of a bus shelter	0.1

The following min-max scaling formula was used for normalization:

$$K_{norm}={\displaystyle \frac{K_i-K_{min}}{K_{max}-K_{min}}}$$

(10)

The outcome of the proposed method is a set of recommended locations for rental stations of specialized off-road wheelchairs for people with special needs. These recommendations represent the synthesis of spatial, functional, and qualitative analyses, ensuring that the final locations are both strategically positioned and socially inclusive.

3. Results

A case study of the developed method was conducted in the southern part of the Silesian Voivodeship in Poland, within the area known as the Beskid region, and more specifically under the governance of the Aglomeracja Beskidzka (Beskid Agglomeration). The region is located within the Carpathian Mountain range, which endows it with significant natural, scenic, and recreational values. The Beskidy are among the most visited mountain areas in Poland, offering a wide range of hiking trails, winter sports infrastructure, protected landscape parks, as well as opportunities for leisure and ecotourism.

The study area spans approximately 2354 square kilometers, encompassing terrain in two distinct ecoregions the Carpathian Mountains and the Eastern Plain and includes mountain ranges such as Beskid Śląski, Beskid Żywiecki, and Beskid Mały. These ranges vary in altitude and character, offering diverse landscapes. The highest peak in Beskid Śląski is Skrzyczne (1257 m a.s.l.), while Beskid Mały is dominated by Czupel (931 m a.s.l.). Most notably, the Beskid Żywiecki range, featuring Babia Góra (1725 m a.s.l.), is the second-highest mountain range in Poland after the Tatra Mountains, making it a key destination for mountain enthusiasts. The selection of municipalities was aligned with the statutory boundary of 38 municipalities making up the subregion designated by Aglomeracja Beskidzka.

The tourism importance of this region is underlined by its high accessibility from urban centers (e.g., Bielsko-Biała), rich natural heritage (landscape parks, scenic vistas, panoramic peaks), and a variety of recreational amenities. These factors make the area particularly relevant for assessing how specialized off-road wheelchair rental stations can enhance inclusive access to natural and outdoor services. The article’s contribution lies in proposing locations that are not only spatially and functionally optimal but also socially equitable, by ensuring that people with special needs can realistically reach and enjoy the mountain environment while relying on accessible transport options.

To ensure the reproducibility of this study, it is important to specify the data sources and timeframes used. All spatial data from OpenStreetMap (OSM) and the Polish Tourist and Sightseeing Society (PTTK) were extracted in July 2024. The analysis was conducted using QGIS version 3.28 “Firenze”. The public transport service analysis, including service frequency (K1), weekend availability (K2), and the percentage of low-floor vehicles (K4), was based on General Transit Feed Specification (GTFS) data and official timetables valid for July 2025. This period was chosen to represent a typical, non-holiday autumn service schedule, thus avoiding anomalies from summer or winter holiday periods. For the purpose of this study, “weekdays” were defined as Monday through Friday, while “weekends” included Saturday and Sunday. Public holidays were treated as Sundays, following standard transport modeling practice.

In the first step, datasets for the study area were collected. Public transport data were obtained from OpenStreetMap and supplemented with information provided by local transport authorities. The PT dataset comprises 2227 unique stops, including transfer hubs.

Data on mountain trails were retrieved from OpenStreetMap and further complemented with records published by the Polish Tourist and Sightseeing Society (PTTK). The MT dataset includes 219 unique trails within 32 municipalities of the southern subregion of the Silesian Voivodeship.

The dataset of mountain shelters was assembled from OpenStreetMap and enriched with data made available by the Polish Tourist and Sightseeing Society (PTTK). The MS dataset contains 32 mountain shelters located across 14 municipalities.

The final identified dataset concerns parking facilities, extracted and filtered from OpenStreetMap. The P dataset comprises 1147 unique elements distributed across all municipalities of the analyzed area. Together, these four datasets provide the empirical foundation for the subsequent stages of the method, ensuring comprehensive coverage of transport, recreational, and supporting infrastructure.

To identify bus stops located within 50 m of mountain trails, an iterative spatial analysis was performed, considering in sequence stops with access to one trail, two trails, and three trails. The result of the iterative removal of stops is presented in Figure 3. This procedure made it possible to distinguish locations of increasing strategic importance, where accessibility is reinforced by the convergence of multiple trails. The set of bus stops was as follows:

Figure 3. Results of the analysis of tourism attractiveness (a) for stops located in the vicinity of a single mountain trail, (b) for stops located in the vicinity of two mountain trails, (c) for stops located in the vicinity of three mountain trails.

• for stops located along a single mountain trail ${PT}_{{MT}_1}=476$;
• for stops located along two mountain trails ${PT}_{{MT}_2 }$ = 106;
• for stops located along three mountain trails ${PT}_{{MT}_3}=37$.

This categorization enabled the comparison of stop distribution by level of trail connectivity, highlighting nodes of greater strategic significance for accessibility planning.

In the next step, bus stops located less than 20 m apart were removed, as their proximity distorts the overall picture of the situation the analysis seeks to identify suitable locations for establishing rental stations. This refinement ensures that the selected points represent distinct, functionally meaningful sites rather than redundant clusters of stops. A dataset was obtained ${PT}_{{FMT}_3}$ = 25.

The final element in assessing the tourist attractiveness of the locations is a network analysis verifying which of the 25 potential sites provide access to a mountain shelter located within 5 km. Four locations did not allow for reaching the designated facility. The resulting dataset comprises $L=15$. This filtering step ensures that only those sites offering realistic and user-friendly connections to mountain shelters are retained, aligning the recommendations with both functional accessibility and visitor safety considerations.

For each of the 15 bus stops, a buffer analysis was performed and the number of parking facilities within its range was verified. On this basis, one location was removed, as no parking facilities were identified within the specified range. For the remaining locations, the number of available parking facilities varied between 1 and 11. The resulting dataset comprises $L_F$ = 14 This step ensured that only stops supported by nearby parking infrastructure were retained, strengthening the functional relevance of the proposed rental station locations. The audit of the identified parking facilities resulted in their positive evaluation. This confirmation indicates that the facilities meet the functional and accessibility requirements necessary for their inclusion in the analysis.

Figure 4 presents the results, with mountain trails and mountain shelters highlighted. The visualization illustrates the spatial relationships between transport stops, trail networks, and tourist facilities, providing an integrated overview of the analyzed area.

Figure 4. The results obtained after completing the second step of the method left 14 potential locations.

To establish the ranking of the remaining locations, input data for the multi-criteria analysis were collected and are presented in Table 2.

Table 2. Data collected for the remaining 14 sites across the five criteria.

Number of Location	Criteria
	K1	K2	K3	K4	K5
1	30	1	1	100	1
2	120	3	1	0	0
3	120	2	1	0	0
4	72	3	1	15	1
5	30	3	1	30	1
6	60	3	1	30	1
7	120	2	0	0	1
8	65	3	1	30	1
9	20	3	0	30	0
10	30	3	0	40	0
11	90	2	0	25	1
12	60	3	1	20	1
13	30	2	1	100	1
14	20	3	1	100	1

Before calculating the final scores, a min-max scaling procedure was applied to the raw data from Table 2 to normalize the values of each criterion to a common scale of [0, 1]. This step is essential to ensure that a criterion with a wide range of values (e.g., K1: number of services) does not disproportionately influence the final result compared to criteria with smaller ranges (e.g., K3: binary).

The ranges used for normalization were:

• K1: min = 20, max = 120;
• K2: min = 1, max = 3;
• K3: min = 0, max = 1;
• K4: min = 0, max = 100;
• K5: min = 0, max = 1;

Second, a crucial accessibility rule was applied: if a location’s raw score for criterion K4 was 0, its final calculated score was manually overwritten to 0.00. This ensures that any location failing the fundamental prerequisite of having accessible low-floor vehicle services is unequivocally disqualified from the final ranking.

After data normalization (min-max scaling) and the application of the weights presented in Section 2, results for individual locations were obtained and are presented in Table 3. This step integrates the multi-criteria framework into a final comparative assessment, enabling the ranking of potential rental station sites.

Table 3. Results of the multi-criteria analysis.

Number of Location	Results
1	0.77
2	0.00
3	0.00
4	0.59
5	0.76
6	0.67
7	0.00
8	0.66
9	0.59
10	0.59
11	0.47
12	0.64
13	0.87
14	1.00

The spatial distribution of the proposed locations is presented in Figure 5, while Figure 6 shows the reduced set of locations, limited to a maximum of one site per municipality, based on the evaluation results. This step ensures territorial balance and prevents the overconcentration of rental stations within individual administrative units.

Figure 5. Spatial distribution of the multi-criteria analysis results.

Figure 6. Recommendation of rental station locations for specialized off-road wheelchairs, assuming one location per municipality.

Ultimately, the method made it possible to recommend nine locations across nine municipalities. The most favorable sites were identified in the municipalities of Szczyrk, Wilkowice, and the city of Bielsko-Biała, which is associated with their good accessibility by public transport and high tourist attractiveness. These findings underscore the dual importance of transport integration and recreational potential in determining optimal locations for specialized wheelchair rental stations.

4. Discussion

The results of this study clearly confirm that the strategic integration of specialized off-road wheelchair rental stations with the public transport network can significantly increase the accessibility of mountain tourism for people with disabilities. This will also contribute to sustainable development in the field of transport. The indicated sites may also improve both demand and supply related to the number of public transport services. The developed methodology constitutes a replicable tool for identifying optimal locations, filling a gap in the literature, where accessibility analyses most often focus on urban transport systems or general recreation planning, overlooking the specific needs of people with reduced mobility. The identification of nine priority locations in the Beskid Agglomeration, with the highest-rated sites in Szczyrk, Wilkowice, and Bielsko-Biała, validates the working hypothesis that the strategic placement of rental stations near transport hubs and tourist trails maximizes functional accessibility. These municipalities achieved the highest scores not only due to their tourist attractiveness but also because of their role as regional transport hubs, which translates into higher service frequency and a greater likelihood of being served by modern low-floor vehicles a key factor in the conducted analysis.

The approach presented in the article is unique due to the synthesis of four dimensions: the attractiveness of tourist resources, the specific transport requirements of people with disabilities, the demand for multimodal transport, and the technical parameters of vehicles. The stability of the final ranking was also considered. While quantitative uncertainty analyses, such as bootstrapping weight vectors or data perturbations, are common in multi-criteria studies, their application is less suitable for our needs-based framework. Our weights do not represent statistical probabilities but a deliberate, qualitative hierarchy where certain criteria (notably K4) function as non-negotiable prerequisites. A random perturbation that assigns a low weight to K4 would create a scenario that is logically inconsistent with the core principles of accessibility for wheelchair users.

Instead, we performed a qualitative robustness check by examining the performance profiles of the top-ranked locations. This inspection reveals that the ranking is highly robust. The top-ranked locations (e.g., #14, #13, and #1) dominate the list not because of a fragile combination of weights, but because they achieve perfect or near-perfect scores on the most heavily weighted criteria, especially the critical K4 (100% low-floor vehicles) and K1 (high service frequency). Conversely, locations ranked at the bottom are primarily those that fail the essential K4 prerequisite. This indicates that the final ranking is not a sensitive artifact of minor weight variations but is driven by clear and significant differences in fundamental accessibility performance, confirming the stability of our recommendations.

By adopting a person-centered measurement perspective, the methodology treats accessibility not as an amenity but as a fundamental condition. The exclusion of stops not served by low-floor vehicles (criterion K4) acts as an absolute filter, reflecting the real experience of a wheelchair user, for whom the service is either 100% accessible or practically non-existent. This perspective is consistent with research emphasizing the role of vehicle design as a prerequisite for genuine inclusivity. Furthermore, by prioritizing public transport, the model aligns with broader sustainable development goals and findings from scholars highlighting the social and environmental benefits of favoring public over individual transport.

The practical implications of these findings go beyond the immediate objectives of the project “MOUNTAINS WITHOUT BARRIERS.” First, the developed methodology serves as a practical and replicable tool for decision-makers and planners, providing databased recommendations for Sustainable Urban Mobility Plans (SUMP). Second, the analytical framework is highly adaptable and can be applied to the location of other micromobility services (e.g., e-bike rentals) in rural or protected areas, where the “last mile” problem from the public transport stop remains a challenge. Third, the model actively promotes a shift in transport habits toward reducing reliance on private cars, which is consistent with European goals for green and sustainable tourism.

However, certain limitations of the study should be noted, which at the same time outline a clear roadmap for future work. The current GIS-based analysis does not account for micro-barriers (e.g., curbs, steep gradients, lack of sidewalks) along the critical door-to-stop path. Furthermore, the model uses a static snapshot of transport services and does not incorporate seasonal variations, such as reduced winter timetables, which can significantly impact accessibility in mountain regions. Finally, the current ranking does not yet include an economic feasibility screen for the top sites. Therefore, future research should proceed in three explicit steps: first, conducting on-site audits to map micro-accessibility; second, developing dynamic models that integrate seasonal service changes; and third, performing a detailed economic feasibility analysis for the highest-rated locations to ensure their long-term operational stability.

These limitations inspire future research, which should combine spatial modeling with a participatory approach, including surveys and consultations with people with disabilities to validate and, if necessary, adjust the weights. Field audits should also be carried out to verify the data, and dynamic models should include factors such as seasonal demand and weather conditions. Another logical step is to conduct an economic feasibility analysis for the highest-rated locations to ensure their long-term operational stability. It would also be worthwhile to explore the possibility of integrating the system with digital navigation platforms providing real-time information on vehicle availability, trail conditions, and timetables, thereby creating a fully inclusive travel chain (after conducting an accessibility audit of the approaches to the transit stop and the rental station).

Furthermore, while the current study employs a GIS-based spatial analysis framework focused on infrastructure placement, we acknowledge the broader trends in transport research that could complement our findings in the future. For instance, advanced predictive models, such as Spatiotemporal Graph Transformers, could enhance operational planning by forecasting demand. Similarly, micro-simulation studies on intelligent transport systems could assess the environmental benefits of directing users to these locations. Future research could also integrate economic models to explore user willingness-to-pay for such accessible services. While these areas fall outside the direct scope of our location-allocation methodology, they represent valuable avenues for subsequent, more holistic research on accessible tourism systems.

5. Conclusions

This study confirms that the thoughtful, systemic integration of specialized off-road wheelchair rental stations with public transport networks is a cornerstone for advancing accessible tourism. By identifying strategic trailheads reachable by public transport, this research provides a foundation for sustainable tourism development and the effective reduction in social exclusion, proving that true accessibility is achieved not through technological innovation alone, but through the seamless integration of technology, infrastructure, and inclusive planning.

The primary research contribution is the creation of a comprehensive, multi-stage, and replicable spatial analysis tool that moves beyond theoretical accessibility models to address real-world barriers. The novelty of this methodology lies in its unique synthesis of four key dimensions: the attractiveness of tourist resources, the specific transport requirements of people with disabilities, multimodal transport demands, and the technical parameters of the specialized vehicle. Crucially, it operationalizes the “personified measurement” approach by treating fundamental requirements—such as the availability of low-floor vehicles—as non-negotiable, “knock-out” criteria rather than simple variables. This ensures the final recommendations are grounded in the practical, lived reality of the end-users.

The application of this methodology in the Beskid Agglomeration successfully identified nine priority locations, validating the model’s effectiveness and providing immediate, implementable guidance for the “MOUNTAINS WITHOUT BARRIERS” project. Beyond this specific case, the findings have significant practical applications. The model serves as a diagnostic tool for crafting targeted interventions within Sustainable Urban Mobility Plans (SUMP). As demonstrated, it pinpoints specific deficits for highly ranked but imperfect sites—whether it is a lack of weekend service or an insufficient share of low-floor vehicles—thus transforming the ranked outputs into a practical roadmap for investment and upgrades. This framework is highly adaptable and can serve as a template for other tourist regions seeking to enhance accessibility, or even for locating other shared micromobility services in areas where the “last mile” remains a challenge.

Ultimately, this work provides decision-makers with a data-driven tool to create a coherent and uninterrupted accessibility chain. By prioritizing locations that are both attractive to tourists and genuinely accessible via sustainable transport modes, this approach empowers people with reduced mobility to fully participate in outdoor recreation while simultaneously promoting greener tourism patterns.