2. Materials and Methods
Data collection;
Proximity analysis between stations, elements of interest, and infrastructure dedicated to soft mobility;
Calculation of proximity indices;
Weighting procedure to the different considered elements and calculation of the overall Soft Mobility Hub (SMH) Index.
2.1. Data Collection (Step 1)
In the first step, data related to railway stations, elements of touristic, cultural, and environmental interest, as well as the networks of soft mobility infrastructures were collected, organised, and georeferenced within a Geographic Information System. This activity was carried out in collaboration with the organizations participating in the Alliance for Soft Mobility (A.Mo.Do.).
In particular, the following data were collected (January 2024):
A total of 2962 active railway stations, with regular railway services in 2023, almost 70% of which (2040) are along railway lines managed by Rete Ferroviaria Italiana, with the remaining 922 along railway lines managed under concession by regional or local companies.
A total of 951 villages that received recognition for their historical/cultural and traditional heritage: Bandiere Arancioni (Orange Flags) of the Touring Club Italiano (TCI), Borghi più belli d’Italia (Italian Most Beautiful Villages), Borghi Autentici d’Italia (Authentic Italian Villages), and the municipalities participating in the Cittaslow and Comuni Virtuosi (Virtuous Municipalities) network. Among these, 164 villages have received multiple recognitions.
A total of 55 UNESCO World Heritage Sites, acknowledged for their outstanding cultural and/or environmental value at the international level. Among these, 40 have specific entry points (free or paid), while 15 are extensive sites without specific entry points (e.g., Dolomites).
A total of 1753 protected areas, including national and regional parks and reserves, Sites of Community Importance (SCIs), Special Protection Areas (SPAs), WWF oases, sub-merged natural parks, marine protected areas, other protected areas recognised by the Italian Ministry of the Environment and Energy Security, and sites participating to the Italian Network of Mining Parks and Museums (ReMI). In total, they cover 89,750 km2 of Italian land, highlighting the importance of biodiversity conservation, habitat preservation, and ecosystem protection. Additionally, 32,985 km2 of those areas are safeguarded under multiple protected area designations.
A total of 18,077 km of existing greenways, derived from the recovery of disused railways, and cycle routes (at national, regional, and provincial level), either completed or planned, which belong to the Eurovelo, Bicitalia, and the National Touristic Cycle Routes networks. About 50% of these (9073 km) belong only to a single category, while 44% (7884 km) are shared by two categories, and the remaining 6% (1094 km) are shared by three categories. The Eurovelo network consists entirely of cycling routes that are also included in Bicitalia.
A total of 84 historical/cultural routes of national and inter-regional level (25,629 km), including the 42 Walks of the National Atlas, accessible on foot and, for some sections, even by bicycle, and 10,871 km of nationally significant paths (Sentiero Italia and European Paths) which offer opportunities for hiking and for exploring nature and landscapes. About 10% of the historical/cultural routes (2154 km) and 11% of the national paths (1293 km) are shared with another route/path. About 1.5% (342 km) of the historical/cultural routes are shared among three or more of them. However, these data are approximate, as differences in GPS tracks from various sources may have prevented the identification of additional shared sections without on-site surveys.
2.2. Proximity Analysis (Step 2)
By foot: 4 km/h, corresponding to a distance of 2 km;
By bicycle: 8 km/h, corresponding to a distance of 4 km.
Naturally, pedestrians, moving at a slower speed, generate smaller areas of influence compared to cyclists, so the pedestrian areas of influence were considered to assess the proximity to almost all elements of interest, except for those primarily or exclusively intended for cyclists. These areas of influence were reduced to 0.5 km for large-scale resources without defined access points (such as protected areas, except mining parks, and some UNESCO sites), as the actual area of interest often concerns only a portion of the entire resource. Additionally, a tolerance proportional to the considered AoI (0.5 km or 1 km) was applied to prevent elements which are close to each other and located at the boundary distances from being included or excluded due to differences of a few metres.
2.3. Calculation of Proximity Indices (Step 3)
For each station and each category of elements, the index has been calculated with the following procedure:
- (a)
Definition, using the Near Table function of ArcGIS Pro 3.3, of the straight-line distance between the station and each individual element, up to the influence distance increased by the tolerance.
- (b)
- Calculation of the proximity index (Ix) for each of the x stations and each of the n categories of elements, through the formula
where
- –
Ix is the proximity index for the x station, for each category of elements;
- –
Dx is the straight-line distance between the x station and each individual element, for each of the n categories of elements;
- –
- AoIn is the area of interest of the n category (Table 2);
- –
- Tn is the tolerance of the n category (Table 2).
Calculation, when the x station is simultaneously close to more than one element within the same n category, of the cumulative proximity index Ix, by the sum of the values of each Ix calculated.
Normalization of the Ix values on a 0–1 scale, based on the maximum value of each of the n categories of elements.
2.4. Weighting Procedure to the Categories of Elements Considered and Calculation of the Overall Soft Mobility Hub (SMH) Index (Step 4)
The overall Soft Mobility Hub (SMH) Index should take into account the relative importance of the considered elements of interest. For this reason, a specific weight has been assigned to the proximity index (Ix) of each of the n categories, reflecting its contribution to tourist attractiveness and, therefore, to the potential role of nearby stations as intermodal hubs.
The relative importance of elements of interest and infrastructure for soft mobility was assessed using the Analytic Hierarchy Process (AHP), a multicriteria decision support technique developed in the 1970s by Thomas L. Saaty. The AHP is a flexible and quantitative method for selecting alternatives based on their relative performance with respect to different criteria. Alternatives are structured in a hierarchical treelike framework, where they are compared pairwise according to the Saaty nine-point individual judgment scale, where “one” indicates that the two criteria compared are equally important, and “nine” indicates that one criterion is absolutely more important than the other.
To assess the effect of different weights assigned to the various indices, three scenarios were considered:
Scenario 1, where all indices have the same weight;
Scenario 2, defined by a group of experts from the University of Milan, Rete Ferroviaria Italiana, and the Alliance for Soft Mobility, through the application of the AHP method;
Scenario 3, defined by a group of stakeholders through the application of the AHP method only to the indices related to the cultural and environmental resources, while keeping the weights of the indices related to pedestrian and cycling infrastructure unchanged compared to scenario 2.
The evaluation matrices of the AHP method were compiled by experts and stakeholders, comparing the various proximity indices in pairs, assigning a score based on the importance of an element relative to the other. The relative weights are then calculated as the geometric weighted averages of all the experts’ judgments.
Before calculating the overall Soft Mobility Hub (SMH) Index, in accordance with the objectives of the study, large urban areas have been excluded from the analysis. To this end, three additional steps have been performed:
First, only the stations located in municipalities with a resident population of less than 100,000 inhabitants have been selected for the calculation of the SMH Index.
Then, the values of all the indices were again normalised (if necessary) on a 0–1 scale, based on the maximum value of each index after the exclusion of stations located in municipalities with a population exceeding 100,000 inhabitants.
Finally, each index has been “winsorised” to reduce the effect of potential outliers and limit extreme values, by considering the following:
- –
Tight tail (upper threshold value): 75th quantile (Q3) + 1.5 × IQR;
- –
Left tail (lower threshold value): 25th quantile (Q1) − 1.5 × IQR;
where
- –
Q1 is the first quartile;
- –
Q3 is the third quartile;
- –
IQR is the interquartile range.
where
- –
SMHx is the overall Soft Mobility Hub Index of the x station;
- –
WHCV is the weight of historical/cultural village (HCV) elements in each of the three scenarios;
- –
WUNE is the weight of the UNESCO sites (UNE) in the three scenarios;
- –
WPA is the weight of protected areas (PAs) in each of the three scenarios;
- –
WCM is the weight of cycling mobility (CM) infrastructures in each of the three scenarios;
- –
WPM is the weight of pedestrian mobility (PM) infrastructures in each of the three scenarios;
- –
HVCx is the proximity index of the x station to historical/cultural villages (HCVs);
- –
UNEx is the proximity index of the x station to the UNESCO sites (UNE);
- –
PAx is the proximity index of the x station to protected areas (PAs);
- –
CMx is the proximity index of the x station to cycling mobility (CM) infrastructures;
- –
PMx is the proximity index of the x station to pedestrian mobility (PM) infrastructures.
3. Results
3.1. Proximity Analysis
Regarding the proximity relationships between the stations and the infrastructure for soft mobility, this study revealed that 61.0% of stations (1807) are within 5.0 km of greenways or cycle routes; only 118 of these are near existing greenways, while 1774 stations are close to existing or planned cycle routes that are part of the National Touristic Cycling, Bicitalia, or Eurovelo networks. In addition, 554 stations (18.7%) are near two or more types of cycling routes.
Concerning pedestrian mobility, 53.0% of stations (1570) are located within 3.0 km from cultural routes and national paths. In particular, 42.8% (1268) are close to cultural routes, 18.9% (561) to paths, and 7.2% (214) to both cultural routes and paths. Further, 16.6% of stations (492) are near two or more cultural routes or paths. In total, 76.1% of stations (2255) are within the area of influence of greenways, cycle routes, cultural routes, and paths. Also, 37.9% of stations (1122) are near both cycling and walking routes (cultural routes and/or paths), and 52.1% are located less than one kilometre from such routes.
Regarding infrastructures for soft mobility, there are 172 stations (5.8%) that are located simultaneously near cycle routes, cultural routes, and paths but not in the area of influence of the elements of interest.
Finally, when considering both the soft mobility network and the elements of interest, it emerged that 15.6% of stations (463) are within 5.0 km from greenways or cycle routes and within 1.0 km from protected areas (2.5 km for mining parks). Among these, 65 stations are also close to a UNESCO site.
Overall, 195 stations (6.6%) are located near greenways/cycle routes and UNESCO sites, 12 of which are simultaneously close to historical/cultural villages, and 3 of which are also close to protected areas.
Further, 178 stations (6.0%) are near greenways/cycle routes and villages of historical and cultural interest, 58 of which are also close to a protected area.
In addition, 14.7% of stations (434) are close to cultural routes/paths and protected areas; among these, 63 are also close to a UNESCO site, while 57 are near to one or more villages of historical and cultural interest (3 stations to both of them).
Moreover, 6.5% of stations (192) are near cultural routes/paths and UNESCO sites, and a similar number (193) is close to cultural routes/paths and historical–cultural villages. Only 14 stations are simultaneously close to cultural routes/paths, UNESCO sites, and historical–cultural villages.
Furthermore, 333 stations (11.2%) are in proximity to both pedestrian mobility routes (cultural routes and paths) and cycling routes (greenways and cycle paths) and simultaneously near one or more protected areas; 168 stations (5.7%) are near greenways/cycle route, cultural routes/paths, and UNESCO sites, while 128 (4.3%) are near soft mobility infrastructure for both pedestrians and cyclists and villages of historical–cultural interest.
A smaller number of stations are near both types of soft mobility infrastructure and at least two elements of interest: specifically, 59 (2.0%) are close to greenways/cycle routes, cultural routes/paths, protected areas, and UNESCO sites, while 50 (1.7%) are near protected areas and historical villages, and 10 are near historical villages and UNESCO sites. Finally, only three stations (Levanto, Orta Miasino, and Vernazza) are close to all types of elements of interest and soft mobility infrastructure that were considered in this study.
3.2. Calculation of Proximity Indices
The mean values of the five indices are generally quite low (ranging from 0.192 to 0.388), due to the fact that the highest index values (=1) typically correspond to stations near two or more elements/infrastructures. This situation implies that, in the standardization process, a station that has only one element of interest/infrastructure very close to it has a low proximity index value, compared to stations that have more than one element in the radius of influence. On the other hand, it would not have been correct not to adequately valorise the stations close to more than one element.
Because the distribution of values is asymmetric and there are outliers, it is more convenient to use the median scores to have a better representation of the real situation.
The historic–cultural villages (HCV Index = 0.421) and UNESCO World Heritage Sites (UNE Index = 0.394) registered the highest median values of the proximity index, highlighting a significant relationship between the railway stations and the nearby historic villages or UNESCO sites. The median scores of the other indices range between 0.174 and 0.243 (PM Index = 0.174, CM Index = 0.185, and PA Index = 0.243), because of a good proximity relationship between many stations and protected areas and soft mobility infrastructures. The lower median scores for soft mobility infrastructures do not mean a scarcity of soft mobility infrastructure near the stations or that these infrastructures are predominantly distant, but, on the contrary, several stations are simultaneously very close to three or more cycling and/or pedestrian routes: in fact, 1542 stations (52.1%) are located less than one kilometre from a cycling or pedestrian route.
Considering all the elements of interest at the same time, it emerges that a very limited number of stations have proximity indices scores exceeding 0.50 simultaneously for different elements: four stations (Corniglia, Manarola, Monterosso and Riomaggiore), serving the famous Cinque Terre villages (Liguria region), scored above 0.50 for both the UNE and PA indices; two stations (Alberobello, Puglia region, and Palmanova, Friuli region) achieved scores above 0.50 for both the HCV and UNE indices; one station (Vogogna Ossola, Piedmont region) scored above 0.50 for HCV and PA indices. Vernazza (Liguria region) is the only station to score above 0.50 for all the three indices.
Finally, only nine stations have scores higher than 0.50 for both the CM and PM indices; four of these, all located in the city of Rome, simultaneously present a score greater than 0.50 for the UNE index.
In addition, 62 stations (2.1% of the total) score PM and AP indices simultaneously greater than 0.25, while 56 stations (1.9% of the total) present CM and UNE indices simultaneously greater than 0.25; for all other combinations, less than 1.5% of stations have scores simultaneously greater than 0.25.
The calculated proximity indices, when considered individually, highlight the potential of trains as a sustainable means of transportation for accessing local resources and developing soft mobility. On the other hand, the limited number of stations with high scores for both soft mobility and indices for elements of interest suggests that integration between rail transport, soft mobility, and local resources often remains insufficient.
3.3. Calculation of the Overall Soft Mobility Hub Index (SMH)
Because they have no elements of interest or soft mobility infrastructures nearby, 532 of these stations (21.0%) have a SMH index value of 0. The remaining 79.0% have SMH index scores ranging from 0.01 to 0.74 in scenario 1, from 0.01 to 0.68 in scenario 2, and from 0.01 to 0.77 in scenario 3. The average (0.15) and median (0.13) scores are the same in scenarios 2 and 3, while they are lower in scenario 1 (average score, 0.13; median score, 0.11).
Most stations have an SMH score below 0.25: the highest percentage is observed in scenario 1 (63.6%), followed by scenario 3 (58.4%) and scenario 2 (57.0%). Half of the Italian regions have more than 80% of stations with SMH index scores below 0.25 or equal to 0. The highest percentages are found in some regions of central-southern Italy and on the islands (Sicily, Calabria, Sardinia, Marche, Basilicata, Lazio), but there are some in the north, too (Emilia-Romagna).
Vernazza (Liguria) resulted as the station with the highest SMH score in both scenario 1 and 2, while in scenario 3, it was the station of Mantua (Lombardy). Vernazza is a historic village in the Cinque Terre area, recognised as a UNESCO World Heritage Site and protected by a National Park and a Marine Protected Area. It is a highly attractive tourist destination, both for its historical–cultural heritage and its environmental significance. Additionally, it is connected to a cultural route (Sentiero Liguria) and is close to an under-construction national cycling route (Ciclovia Tirrenica). Mantua, instead, is a town in the Po Valley that is not recognised as a historical–cultural village but is protected as a UNESCO World Heritage Site and surrounded by a Special Protection Area (Valli del Mincio). It is particularly suited for soft mobility, being connected by two national cycling routes and two cultural routes.
4. Discussion and Conclusions
This study highlighted that more than one-third of Italian railway stations are located near one or more elements of interest, confirming the hypothesis that the Italian railway network can play an important role in improving access to local resources. To enhance the touristic and recreative function of the railway network and inland and peripheral areas as touristic destinations, it is fundamental to improve the connection between elements of interest and railway stations.
In times of limited economic resources, it is also important to decide where to primarily direct improvement interventions, in order to maximize the return on investment. Identifying soft mobility hubs, i.e., the railway stations closest to elements of interest and soft mobility infrastructures, connecting all the available territorial information, appears to be a suitable method. Identifying such hubs, in fact, allows establishing priorities regarding where to initially concentrate financial resources and enhancement interventions, maximizing the return and promoting sustainable development.
Furthermore, over 75% of the stations are located in proximity to soft mobility infrastructures (pedestrian and/or cycling), which not only represent a touristic attraction, but, at the same time, can facilitate access to other elements of interest through a sustainable transport mode. In this sense, the CM and PM index scores confirm a strong connection between these stations and nearby greenways, cycling routes, cultural routes, and paths, either existing or planned. More than a half of the stations are located less than one kilometre from these routes, providing a direct or almost-direct connection between the train and paths for exploring the surrounding areas by bike and on foot. This demonstrates how the planning of soft mobility infrastructure networks, particularly for cycling, has been carefully integrated with the railway network. There is the exception of some areas of the South and the major islands, where there is still a delay in the development of soft mobility infrastructure (both cycling and pedestrian) and in its integration with the similarly limited railway network. In any case, there are still significant areas for improvement, as 23.9% of stations lack pedestrian or cycling infrastructure within the considered areas of influence, and 22.8% are more than one kilometre away from the nearest piece of infrastructure. These latter stations, in particular, represent opportunities where, with relatively modest investments, the connection with the soft mobility network could be significantly enhanced, encouraging intermodality with the train.
Overall, approximately one-third of Italian railway stations (877) are located near pedestrian and/or cycling routes and simultaneously fall within the areas of influence of historical villages, UNESCO sites, and/or protected areas. These stations could serve as hubs for soft mobility, facilitating the interchange between the railway network and networks of greenways, cycling routes, cultural routes, and paths to access cultural and historical resources, visit villages and towns, and explore protected areas.
The developed methodology also allows for the identification of stations located near elements of interest but currently lacking soft mobility infrastructures. These stations could also become important hubs for intermodality and sustainable tourism if the missing soft mobility infrastructures were built in the surrounding areas. By assigning a “priority score”, the model can assist decision makers to identify the stations with the highest potential, helping to address the usually limited resources toward the creation of soft mobility infrastructure that is most effective for fostering intermodality and active tourism. Planning new soft mobility infrastructures without an easy and safe connection with railway stations cannot lead to a concrete increase in active tourism and the development of a more sustainable tourism model, which are essential elements for the enhancement of inner peripheries.
This study presents some limitations, especially due to the databases used for attractions and soft mobility infrastructures. The elements of interest we considered were limited to those collected for the creation of the “Atlas of soft mobility in Italy” by A.Mo.Do. and Rete Ferroviaria Italiana, excluding other potential categories of points of interest. Due to the lack of a national database for soft mobility infrastructures, data from various sources had to be used, sometimes with differing scales; these discrepancies may have introduced inaccuracies that affect the results of the analysis. Furthermore, due to inaccuracies in available road network graphs and the lack of a sufficiently accurate and comprehensive geographic database for soft mobility infrastructures, also including local-level routes, distances between stations and resources were calculated in a straight line; this may have influenced the actual proximity relationships between stations and some elements, but, on the other hand, this approach avoided the errors caused by the aforementioned limitations, which might have otherwise excluded some real proximity relationships. Lastly, it was not possible to validate the results by collecting information on practical examples that were already implemented.
Future developments of this work could include mapping additional elements of interest, such as monuments, lakes, landscapes of particular significance, etc., as well as cycling and pedestrian soft mobility infrastructure at the local or provincial level. Furthermore, more complex models could be developed to analyse the relationships between stations, elements of interest, and soft mobility infrastructure, in order to take into account physical barriers (such as main rivers and mountain ridges), as well as the altimetric profiles of routes, which are crucial factors influencing choices about walking or cycling travel. Moreover, it could be important to consider socioeconomic aspects (such as local population density, economic impact or accessibility for different demographic groups) and temporal variations in transportation demand (such as seasonal tourism fluctuations) in the data model, in order to better evaluate the effectiveness of the possible hubs. Finally, the assessment of the practicality and desirability of using railway stations as intermodal hubs could be investigated, also through qualitative data such as traveller feedback.
Source link
Giulio Senes www.mdpi.com
