Next Article in Journal
Nutrient Status and Antioxidant Activity of the Invasive Amaranthus retroflexus L.
Previous Article in Journal
The Role of AI in Predictive Modelling for Sustainable Urban Development: Challenges and Opportunities
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 11
10.3390/su17115149
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Urban Commuting Preferences in Italy: Employees’ Perceptions of Public Transport and Willingness to Adopt Active Transport Based on K-Modes Cluster Analysis

by
Mahnaz Babapour
*,
Maria Vittoria Corazza
and
Guido Gentile
Department of Civil, Building and Environmental Engineering, Sapienza University of Rome, 00184 Rome, Italy
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(11), 5149; https://doi.org/10.3390/su17115149
Submission received: 17 March 2025 / Revised: 25 May 2025 / Accepted: 30 May 2025 / Published: 3 June 2025
Browse Figures
Versions Notes

Abstract

:
Commuting plays a critical role in shaping sustainable transport systems, yet understanding the diverse preferences of commuter groups remains a challenge for policymakers. As cities aim to promote sustainable transport, it is essential to better understand the factors influencing travel behaviors. This study investigates the commuting preferences and behaviors of urban employees in Italy, focusing on identifying distinct user profiles and their implications for policy development. Using a dataset of 2301 participants from Italian cities, the research analyzed transport mode choices, willingness to adopt sustainable transport options, and perceptions of public transport (PT) services, including factors such as travel time, proximity to PT stops, cost, and comfort, rated on a four-point Likert scale. K-modes clustering was employed to segment participants into three clusters based on their travel behaviors. The results revealed three distinct user profiles: (1) car-dependent users with negative perceptions of PT, driven by family obligations and dissatisfaction with PT services; (2) individuals who primarily use cars but are somewhat open to improvements in PT; (3) individuals willing to adopt alternative mobility options, including active and shared transport modes. Significant differences were found across clusters in terms of mode choices, willingness to use sustainable transport, and satisfaction with PT services. Notably, employees showed limited interest in alternative sustainable transport modes such as e-scooters and walking, with 73% and 66% of participants expressing little or no interest, respectively. Despite incentives such as company subsidies for purchasing bicycles or e-scooters, 58% of employees remained uninterested in adopting these alternatives. Additionally, employees’ perceptions of PT services revealed dissatisfaction with factors such as travel time, comfort, and punctuality, with over 70% rating these aspects as “Poor” or “Fair”. These findings suggest that improving the quality of PT services, particularly in terms of travel time, punctuality, comfort, and cost, should be a priority for enhancing user satisfaction. This research provides valuable insights for policymakers seeking to reduce car dependence and promote sustainable urban transport planning.

1. Introduction

Urbanization and population growth have significantly increased travel demand, posing challenges to transport systems worldwide [1]. The 2030 Agenda for Sustainable Development highlights the importance of sustainable urban mobility, particularly for vulnerable groups such as children, women, the elderly, and people with disabilities [2]. To promote sustainability, various policies have encouraged car-free travel modes such as walking, cycling, and public transport (PT). However, travel mode choice is shaped by both habit and intention [3,4], and is also influenced by socio-economic factors, cultural norms, urban design, and technological advancements [5], as well as individual perceptions and personal attitudes [6,7,8]. Analyzing citizen travel habits is crucial for addressing urban mobility challenges like traffic congestion and air pollution while ensuring transport systems meet the needs of diverse populations [9].
Efforts to promote sustainable transport have led to the development of mobility management (MM) strategies, supported by organizations such as the European Platform on Mobility Management (EPOMM) [10]. MM focuses on managing travel demand and encouraging sustainable transport options with minimal financial investment yet significant benefits in terms of congestion reduction and environmental impact [11]. Legislative efforts such as Italy’s Decree Ronchi (1998) introduced home-to-work travel plans, requiring large employers to implement strategies that promote alternatives to private car use [12]. Additionally, the European Directive 2014/95/EU underscores Corporate Social Responsibility (CSR), emphasizing the role of enterprises in mitigating their environmental and social impacts [13,14]. Such legislation led to the establishment of the mobility manager role in Italy, facilitating the adoption of sustainable transport initiatives through company-level policies and employee engagement [15].
Transport Demand Management (TDM) has become a key approach in urban transport planning, aiming to reduce reliance on private cars by promoting sustainable alternatives such as PT, cycling, walking, carpooling, and teleworking [16,17,18]. These strategies are aligned with global sustainability goals and play a crucial role in shifting commuter behavior toward more environmentally friendly transport modes [16]. Among TDM strategies, Workplace Travel Plans (WTPs) specifically focus on encouraging employees to adopt sustainable commuting options [19]. While these initiatives promote alternatives such as PT and cycling, the continued preference for private vehicles remains a challenge, often attributed to perceived convenience, independence, and comfort. Successful WTPs require a combination of infrastructure improvements, employer-led incentives, and technological advancements to make sustainable commuting more appealing [20,21,22]. However, poor infrastructure, including inadequate transfer points and limited accessibility for people with disabilities, discourages the use of sustainable transport [23,24]. Investing in high-quality, reliable transport services is essential for reducing car dependency, as improved infrastructure enhances accessibility, reduces emissions, and mitigates social exclusion [25,26]. The effective implementation of WTPs requires strong collaboration between employers, employees, and policymakers, along with supportive regulations to ensure their success [27,28].
The perception of urban mobility plays a vital role in shaping sustainable transport choices, as individual preferences are strongly influenced by service quality, intermodal connectivity, and user experience [29]. Despite efforts to promote sustainable transport, private car use remains dominant, partly due to its association with autonomy, predictability, and societal acceptance [30,31]. In cities where investments in road infrastructure exceeds spending on walking, cycling, and PT, car dependency becomes further entrenched [32]. PT perceptions, particularly regarding service quality attributes such as timetable reliability and intermodal connections, significantly influence user satisfaction and mode choice [33]. Comfort has been identified as a major determinant of transport preferences [34]. Research has categorized PT users based on their perceptions and satisfaction levels, demonstrating that expectations vary depending on overall service quality [35,36]. Understanding these perceptions is essential for designing policies that encourage sustainable transport and reduce dependence on private vehicles.
Given the persistent dominance of private car use and the complex interplay of behavioral, infrastructural, and perceptual factors, this study investigates commuting habits and preferences among urban employees in Italy through a novel clustering-based approach.
To the best of the authors’ knowledge, this is the first study in Italy to combine clustering techniques specifically suited for categorical data with a wide range of behavioral and attitudinal factors, such as employees’ perceptions of PT quality, willingness to adopt sustainable modes, and personal constraints like family obligations. This study offers novel contributions by segmenting Italian employees based on a comprehensive set of behavioral and perceptual variables. While previous research has often focused narrowly on mode choice or demographic characteristics, our approach captures a more holistic picture of commuting behavior. By contextualizing the findings alongside international studies, this study illustrates both parallels with known commuter profiles (e.g., from Northern Europe) and the distinct challenges facing Italian cities, such as lower satisfaction with PT and infrastructural limitations. This nuanced understanding supports the design of targeted and context-sensitive mobility policies.
This research explores the commuting habits and preferences of urban employees in Italy, aiming to identify distinct user groups and assess their impact on policy formulation. Specifically, the study seeks to answer the following questions: How do commuter clusters—classified based on travel behaviors, mode choices, and attitudes toward sustainable transport—differ in their willingness to adopt sustainable transport options and their satisfaction with PT services? How do family obligations and other personal responsibilities influence transport mode preferences? What are the requirements for effective WTPs?
This study builds upon our previous research [37], which analyzed commuting behavior in Italy using K-means clustering and spatial analysis. While that study focused primarily on clustering based on travel behaviors, the present research employs K-modes clustering, which is better suited for categorical data. Additionally, this study incorporates users’ perceptions of various aspects of PT, a dimension not explored in the previous analysis. These enhancements provide a more comprehensive understanding of commuting behavior and PT preferences.
This study starts with a literature review (Section 2) to emphasize the necessity of a more in-depth examination of commuter travel patterns. Next, the methodology (Section 3) outlines the case study, descriptive statistics, and clustering techniques used in the analysis. Section 4 presents the findings, while Section 5 provides a discussion, and Section 6 provides some concluding remarks.

2. Literature Review

In this section, key studies related to WTPs and corporate mobility management, PT perceptions and sustainable transport adoption, and the application of clustering techniques in travel behavior research are reviewed. First, strategies and policies promoting sustainable commuting within organizations are discussed. Next, users’ perceptions of PT and their influence on travel mode choices are examined. Finally, the use of clustering techniques in analyzing travel behavior is explored, highlighting their relevance to this study. The section concludes by identifying existing research gaps that serve as a foundation for the current study.

2.1. WTPs and Corporate Mobility Management

WTPs are strategic initiatives aimed at improving sustainable commuting options for employees. The success of WTPs depends on evaluating available transport choices, commuting behaviors, and the factors influencing mode selection. For example, when considering walking as a commuting option, research shows that factors such as individual characteristics (e.g., age, car ownership), employment conditions (e.g., availability of free parking), and psychosocial influences (e.g., social norms, intention to walk) significantly impact decision-making [38].
Similarly, PT usage is shaped by both personal and external factors. A study of suburban commuters traveling to a major hospital found that accessibility and convenience were critical determinants of PT use [39]. Other studies have examined the long-term impact of WTP interventions on active commuting, showing that strategies promoting walking, cycling, and PT can lead to modest but significant increases in sustainable travel behaviors [40]. On the other hand, research has also investigated the reasons for choosing private cars or motorcycles, highlighting obstacles to adopting more sustainable alternatives [41].
Studies suggest that effective WTPs cater to a range of employee needs, such as offering flexible work hours to alleviate time pressures, improving PT accessibility, and providing incentives for active commuting [42]. Additionally, research highlights that workplace policies can either support or undermine sustainable transport objectives, depending on whether they provide free parking or prioritize alternative modes [43]. Improving the accessibility and connectivity of PT networks has also been identified as a crucial factor for the success of WTPs [44].

2.2. PT Perceptions and Sustainable Transport Adoption

Recent research underscores the multifaceted factors that shape perceptions of PT and the adoption of sustainable transport. User satisfaction with PT is influenced by factors such as service quality, intermodal connectivity, and adherence to timetables [33]. Attitudes toward sustainable transport, subjective norms, perceived behavioral control, and egoistic values all play a role in determining the likelihood of opting for sustainable transport options [45]. For younger university students, the perceived environmental impact and customer-related factors are more influential than product-specific factors when choosing transport modes [46]. Although innovations and sustainability initiatives in PT are viewed positively, their actual impact on users remains limited, as they tend to prioritize speed, frequency, and safety [47]. Gender and age also affect perceptions, with men placing more value on technological advancements, while women emphasize sustainability [47]. These findings highlight the importance of aligning transport services with user expectations and values to foster the adoption of sustainable transport.

2.3. Application of Clustering Techniques in Travel Behavior Research

Clustering methods have been extensively used in travel behavior research to identify patterns and categorize individuals based on their mobility choices. Different techniques have been applied, including K-means clustering on activity–travel behavior time series [48], principal component analysis followed by K-means clustering to define user profiles [49], and clustering based on travel mode choice dynamics using GPS data [50]. These approaches have identified distinct groups of travelers, such as those who prefer stability, seek happiness, or enjoy routines [49]. One study employed latent class clustering analysis to identify distinct subgroups among e-scooter non-users in Helsinki and Tokyo. The analysis revealed five latent classes that differ in their attitudes and socio-demographic characteristics, ranging from those with highly negative perceptions to those with less negative views toward e-scooter deployment. The study highlighted that safety concerns were more prevalent among individuals with stronger negative attitudes, whereas cost barriers were more relevant for lower-income groups with milder negative perceptions [51]. Another study used K-means clustering and identified four distinct commuter groups based on demographic factors and transport preferences, including age, gender, family circumstances, vehicle ownership, and willingness to use sustainable modes [37].
These clustering methods provide important information to transport planners and policymakers, helping them better understand travel behavior patterns and develop strategies for promoting sustainable urban mobility.

2.4. Research Gap

The increasing adoption of shared mobility services, such as e-scooters and bike-sharing, has introduced new commuting options. However, further research is needed to assess employees’ desire to integrate these services into their daily routines. Studies should focus on identifying the specific barriers to adoption, including safety concerns, infrastructure limitations, and the compatibility of these services with work schedules. Moreover, the role of employers in facilitating the use of these services remains an area that has not been fully explored.
Another significant challenge is ensuring inclusiveness in WTPs. Research should investigate how these plans can better cater to employees with diverse needs, including those with disabilities or caregiving responsibilities. Understanding how different population groups respond to workplace interventions is important when designing more targeted and equitable policies.
By addressing these research gaps, this study aims to provide a more comprehensive understanding of the factors influencing commuting behavior, contributing to the creation of more effective and inclusive mobility policies.

3. Methodology

This section outlines the study’s methodology, starting with the data collection process. Descriptive statistics are provided to offer a comprehensive overview of the dataset, followed by a discussion of the selection of variables and the application of K-mode clustering. Figure 1 illustrates the research methodology framework.

3.1. Data Collection

An online survey covering multiple cities across Italy (Figure 2) was used to collect data for this study, yielding 2301 valid responses. Among these, the ten cities with more than thirty responses each accounted for approximately 84% of the total. As shown in Figure 3, Milan, Bari, and Rome had the highest responses collectively, making up 62.58% of the dataset (1940 valid responses). These cities represent a mix of northern, central, and southern Italy and include both large metropolitan areas and medium-sized cities, which contributed to a degree of geographical and contextual diversity in the sample. Slovin’s formula [52] (see Equation (1)) was applied to evaluate if these 1940 responses were sufficient for analyzing a population of 8,404,832 individuals. In this formula, n represents the sample size, N is the total population, and e denotes the margin of error. Assuming a commonly used 95% confidence level, the corresponding margin of error is 5%. For a population exceeding eight million, this level of confidence and margin of error requires a minimum sample size of approximately 385 responses. Since the study obtained 1940 responses—well above this threshold—the sample size was considered statistically sound, even under more stringent accuracy conditions.
n = N 1 + N   e 2 ,

3.2. Descriptive Statistics

The dataset analysis revealed several key patterns. The majority of respondents (about 60%) were male, with the 41–55 age group being the most prevalent (71%), while other age groups had lower representations. In terms of household composition, 17% lived alone, whereas 33% resided in households with three or more members. Additionally, half of the respondents had family members who required transport assistance.
Regarding employment status, the majority (85%) worked full-time, typically five days a week, while smaller shares were employed part-time (9%) or worked in shifts (6%). Travel habits indicated that 58% commuted by car, while 28% relied on PT. Notably, 77% of private car users reported being satisfied or very satisfied with their chosen mode of transport. Among the factors influencing travel decisions, 24% prioritized cost, 34% value independence, and only 10% cited parking difficulties as a concern.
In terms of vehicle ownership, 83% owned a private car, 15% had a motorcycle, and fewer than 4% owned a bicycle or e-scooter. When considering alternative transport options, 66% expressed a strong desire to walk, while opinions on company-provided bicycles were evenly divided. Additionally, 27% indicated a willingness to use e-scooters for commuting.

3.3. Variable Selection

Users’ perceptions were assessed across six dimensions: information, travel time, punctuality, comfort, cost, and proximity to PT stops. To select the most suitable variables, a correlation analysis was performed. Since punctuality and information exhibited a strong correlation (see Figure 4), they were excluded from the clustering process. Consequently, only travel time, comfort, cost, and proximity to PT stops were retained to represent perceptions of PT services. This selection ensured that each variable contributed unique insights, improving the identification of distinct passenger satisfaction patterns.
The retained perception-related variables were classified into four categories: Poor, Fair, Good, and Very Good. Incorporating these ratings into the clustering analysis provided a more profound understanding of users’ perspectives on PT services. By examining these ratings, distinct user groups could be identified based on their satisfaction levels, enabling the recognition of specific needs and expectations. This approach supported the development of targeted improvements in PT services.
In addition to these perception-related variables, the analysis also incorporated demographic variables (gender and age), travel behavior variables (mode of transport, travel cost, and experience with shared mobility services), reasons for choosing a particular travel mode (such as cost, travel time, independence during trips, lack of PT availability, reduced stress, parking issues, accompanying others, and the need for intermediate stops), as well as users’ willingness to walk, use e-scooters, adopt shared mobility services, or use a bicycle or e-scooter provided by their employer.
In total, forty-nine one-hot encoded (0,1) variables—detailed in Table 1 —were selected for clustering, as no significant correlations were detected among them.

3.4. K-Mode Clustering

Partition-based and hierarchical clustering methods are among the most widely used techniques for categorical data clustering, as evidenced by the frequent application of algorithms such as K-modes [53]. K-modes extends the K-means algorithm to handle categorical data [54], providing notable advantages in terms of simplicity and computational efficiency [55], particularly when managing large datasets. Instead of means, it utilizes modes and employs simple matching dissimilarity rather than Euclidean distance [56]. However, the main drawback of this method is the need to define the number of clusters [55,57].
In this study, Python 3.9.20 was used to determine the optimal number of clusters using the Elbow Method [58], the Davies–Bouldin Index (DBI) [59], and the Silhouette method [60]. The Elbow Method involves plotting the total Within-cluster Sum of Squares (WSS) against the number of clusters, where the “elbow” point represents the stage at which adding more clusters no longer leads to substantial improvements in data fit. The DBI evaluates clustering quality by measuring cluster compactness and separation, with lower values indicating better clustering performance. The Silhouette method assesses how well each data point fits within its assigned cluster compared to other clusters, with the highest score representing the optimal number of clusters [61].

4. Results

In this section, the results of the analysis are presented, beginning with the determination of the optimal number of clusters. A three-cluster analysis is then conducted, followed by an examination of the cluster centers and chi-square statistics. Finally, the sustainable transport preferences and satisfaction levels of the participants are explored.

4.1. Determining the Optimal Number of Clusters

The clustering evaluation methods suggest varying optimal numbers: the Silhouette method points to k = 2 or k = 5–6, while the Davies–Bouldin Index indicates k = 2 or k = 9. The Elbow Method recommends k = 4–5. K = 2 is the most consistent across the different methods, suitable for simpler clusters, but k = 4 or 5 may be better for more detailed segmentation. The choice depends on whether simplicity or more refined clustering is preferred.
The study initially applied k-modes clustering with four clusters. Figure 5, Figure 6 and Figure 7 display the results of the Elbow Method, Silhouette Score, and DBI, respectively.
Therefore, this study began by applying k-modes clustering with four clusters. The chi-square test revealed that several variables had p-values greater than 0.05, indicating that they might not play a significant role in differentiating the clusters. Specifically, the variables Age Group: 36–40 years; Age Group: 56–60 years; and Travel Cost: EUR 41–60 were found to be non-significant.
Next, the study proceeded with a k-modes clustering analysis using three clusters. The chi-square test results indicated that the same variables—Age Group: 36–40 years; Age Group: 56–60 years; and Travel Cost: EUR 41–60—remained non-significant.
To improve clustering validity, the non-significant variables were removed, and the analysis was rerun. The Elbow Method (Figure 8) suggests three clusters offer a satisfactory balance between simplicity and data structure. The highest Silhouette Score is for two clusters (Figure 9), but it drops with more clusters. The DBI (Figure 10) is lowest for two clusters but still acceptable for three. Beyond three clusters, DBI increases, indicating poorer clustering quality. Based on these metrics, three clusters are optimal, though two clusters could be chosen to maximize separation.

4.2. Three-Cluster Analysis—Cluster Centers

The previous section performed k-modes clustering, taking into account three clusters with the following characteristics:
Cluster 1: Car users with negative perceptions of PT. This group predominantly relies on private cars, largely influenced by family obligations and significant dissatisfaction with PT. They rate PT poorly in various aspects, including travel time, comfort, cost, and proximity to stops. The presence of family members who cannot travel independently further reinforces their dependence on private vehicles. Moreover, they exhibit minimal interest in alternative mobility options such as walking, e-scooters, or shared mobility services. Their negative perceptions of PT, combined with their family-related travel needs, make them unlikely to transition to other modes of transport.
Cluster 2: Car users with a slightly more positive view of PT. Although this cluster also prefers using private cars, their perception of PT is slightly more favorable compared to Cluster 1. They rate PT as “Fair” regarding travel time, comfort, cost, and proximity to stops, indicating moderate dissatisfaction rather than a complete rejection. Although this group does not currently use PT or show strong willingness to shift to other modes (e.g., walking or shared mobility), their moderate PT ratings indicate potential receptiveness to modal shift, especially if PT service quality is improved. Compared to Cluster 1’s resistance and Cluster 3’s strong openness to sustainable modes, Cluster 2 appears to be a middle-ground group, representing an important target for transition policies aimed at car-dependent commuters who are not completely averse to PT.
Cluster 3: Individuals open to alternative mobility options. Unlike the previous clusters, this group does not prioritize car use and maintains neutral opinions about PT. They are open to exploring alternative mobility options, such as walking and shared mobility services. Additionally, they are willing to use bicycles or e-scooters if provided by their employers. This flexibility may stem from personal lifestyle preferences or workplace incentives, making them strong candidates for initiatives that encourage sustainable and shared mobility solutions. Notably, Cluster 3 is the only group that clearly demonstrates a willingness to adopt sustainable transport modes—in contrast to Clusters 1 and 2, which show no such openness.

4.3. Three-Cluster Analysis—Chi-Square Statistics

The application of chi-square statistics in k-modes clustering has been explored to enhance the algorithm’s performance for categorical data. The chi-square test is a versatile statistical tool applied in transport research. It can be used to analyze road accidents in construction zones, ensuring the proper comparison of observed and expected data [62]. In traffic engineering, chi-square tests are employed alongside other statistical methods for data analysis, hypothesis testing, and model evaluation [63]. The test is also valuable in assessing consumer satisfaction, such as in studies on electric bikes, by determining if observed differences are due to chance or if they reflect genuine relationships between variables [64].
In this study, the chi-square statistics measure how significantly each variable differs across clusters, with higher values indicating a stronger influence in distinguishing between them. As shown in Table 2, the analysis highlights that PT service ratings—particularly negative assessments of travel time, proximity to stops, comfort, and cost—are the most influential factors in cluster formation. This result underscores the crucial role of PT perception in shaping commuter segments.
Additionally, the willingness to adopt alternative transport modes, such as bicycles, e-scooters, and shared mobility, strongly contributes to cluster differentiation, reflecting variations in openness to sustainable transport. Mode choice preferences—especially for cars, PT, or bicycles/e-scooters—also play a key role, along with the willingness to walk.
Factors with a moderate influence on cluster distinctions include motivations behind mode choice, such as cost considerations, limited PT availability, parking difficulties, and the desire for independence during the trip. These indicate that each cluster is shaped by distinct travel behavior drivers. Demographic characteristics like age and gender also contribute to cluster formation but have lower chi-square values compared to PT service ratings and willingness to use alternative mobility. Travel costs influence clustering to some extent, though specific cost categories, such as EUR 61–80, along with older age groups (over 60 years) and reduced stress as a travel motivation, have lower chi-square values despite remaining statistically significant. The non-significant variables are traveling costs in the EUR 61–80 range (p-value = 0.11) and travel time as a reason for mode choice (p-value = 0.32), suggesting that these do not play a meaningful role in defining clusters.
Overall, satisfaction with PT services and openness to sustainable transport modes are the primary drivers of cluster separation. Individuals with strong dissatisfaction, particularly those who rate travel time, stop proximity, and comfort poorly, tend to form a distinct group, often favoring car use. While demographic factors like age and gender contribute to differentiation, their impact is less pronounced than travel habits and transport attitudes. Although cost considerations influence travel choices, lower-income travelers do not emerge as the most distinct cluster-defining factor.
Figure 11 illustrates the distribution of cases among the three clusters. Cluster 1 consists of 827 cases (36%), Cluster 2 includes 832 cases (36%), and Cluster 3 comprises 642 cases (28%). This distribution suggests that the clusters are balanced in size.
Figure 12 presents the results of k-modes clustering, showing three distinct user groups projected onto two dimensions using multiple correspondence analysis (MCA). MCA is a statistical method for exploring and visualizing relationships among categorical variables [65]. It extends correspondence analysis to analyze multiway data [66] and can be used to investigate patterns, trends, and outliers in multi-dimensional categorical datasets [67]. MCA is particularly useful for analyzing questionnaire data and creating typologies of individuals based on their similarities [65]. Researchers have developed various approaches to MCA, including combining it with clustering techniques to capture heterogeneity in subgroups [68].
The figure below depicts each cluster in a different color. Although there is some overlap between clusters, indicating individuals with mixed or transitional commuting behaviors, there are also clear separations. These distinct areas highlight meaningful differences in transport mode preferences across the identified user groups.

4.4. Exploring Sustainable Transport Preferences and Satisfaction

Gaining insight into employees’ attitudes toward sustainable transport options is vital for encouraging environmentally conscious travel behaviors and decreasing reliance on private cars. Figure 13 illustrates employees’ willingness to adopt three sustainable options: bicycles provided by their company, e-scooters, and walking.
The results reveal a generally low level of interest in these alternatives. The lowest interest was observed in the use of e-scooters, with 73% of employees expressing no interest, followed by walking, where 66% of employees indicated limited interest. Additionally, despite incentives such as company subsidies for purchasing bicycles or e-scooters, 58% of employees still reported a lack of interest in adopting this mode.
This analysis includes both shared and privately owned e-scooters and bicycles, providing a comprehensive view of employees’ preferences regarding different forms of micromobility.
Previous studies have indicated that e-scooters are frequently used for trips to recreational, educational, and city center areas [69], which may limit their perceived suitability for daily commuting purposes. Additionally, existing research highlights that the main barriers for non-users are primarily external and infrastructural in nature. These include the convenience of other transport options, safety concerns while riding in traffic, inadequate road conditions, the absence of dedicated cycling infrastructure, and the distance to destinations being too far for bike or e-scooter use [70].
Overall, these findings suggest the importance of addressing both individual attitudes and structural challenges when encouraging the adoption of sustainable transport modes. Enhancing infrastructure, improving safety, and aligning workplace policies could play a crucial role in shifting commuting behaviors toward more sustainable alternatives.
Furthermore, we analyzed employees’ perceptions of various aspects of PT services. Table 3 presents employees’ perceptions of various aspects of PT, including information, travel time, comfort, cost, proximity to PT stops, and punctuality, using a four-point Likert scale (Poor, Fair, Good, Excellent). The results indicate significant dissatisfaction with travel time, comfort, and punctuality, with over 70% of respondents rating these factors as Poor or Fair. Information and cost also show high levels of dissatisfaction, though they are rated somewhat better than the other aspects. Proximity to PT stops received relatively better ratings, but most respondents still consider it inconvenient.
These results imply that improving PT travel time, punctuality, comfort, and cost should be key priorities to enhance user satisfaction. Other studies support this, suggesting that PT usage would increase if service quality aligned with users’ expectations. Specifically, improvements should focus on the better integration of intermodal options, more reliable adherence to timetables, and a more responsive approach to users’ needs [33].

5. Discussion

Table 4 presents a summary of the key characteristics of the three distinct commuter clusters identified in the analysis. The clusters were formed based on individuals’ transport preferences, perceptions of PT, and willingness to shift to alternative mobility options.
The results of the clustering analysis reveal three distinct groups of individuals with varying attitudes toward transport modes and PT services, each offering important lessons in into urban mobility planning and policy development.
The first cluster consists of individuals who are highly reliant on private cars, primarily due to family obligations, and exhibit a strong preference for car use. They express significant dissatisfaction with PT, citing issues related to travel time, comfort, cost, and the proximity of PT stops. These negative perceptions reinforce their dependence on cars and limit their adoption of alternative transport modes. Without substantial improvements in PT services, members of this cluster are unlikely to adopt more sustainable transport options. Therefore, strategies must not only enhance PT but also address broader socio-economic and lifestyle constraints.
Initiatives aimed at improving PT quality, such as reducing travel times, enhancing comfort, and optimizing PT stop locations, are crucial. Studies indicate that acceptable walking distances to PT stops vary depending on urban density and transport mode, with users willing to walk further to train stations than to bus or tram stops [71]. Additionally, optimizing stop locations using mathematical models can help reduce travel time and improve transit accessibility [72]. Geographical features, such as elevation differences and footpath gradients, should also be considered when assessing stop reachability, especially in rural areas [73]. These insights can guide future strategies for more sustainable PT systems. However, improving infrastructure alone is insufficient; cultural shifts and changes in behavior toward car dependency must also be part of the solution.
Tailored solutions for this group could include car-sharing options and family-oriented mobility services. Providing mobility solutions for individuals caring for dependent family members is essential for ensuring accessibility and reducing social exclusion, particularly in rural areas [74]. An inclusive mobility system should accommodate the needs of all individuals, regardless of disability status, through both technological and non-technological innovations [75]. Achieving this goal requires collaboration between the government, the private sector, academia, and civil society to develop comprehensive strategies.
Cluster Two consists of individuals who primarily use cars but have a slightly more favorable perception of PT compared to Cluster One. While they still prefer cars, their moderate dissatisfaction with PT services suggests the potential for modal shift if services are substantially improved. Enhancing travel time, comfort, and reliability could make PT a more competitive alternative. Moreover, targeted communication strategies could raise awareness among PT providers and users through real-time information, and user feedback mechanisms may also help build trust and reshape attitudes toward PT.
The third cluster consists of individuals who are more open to adopting alternative mobility options, such as walking, cycling, shared mobility, and company-provided bicycles or e-scooters. They have a neutral perception of PT and are not heavily reliant on cars, suggesting greater adaptability to changes in urban mobility. This cluster shows a high willingness to shift toward sustainable transport modes, which points to the potential to encourage the adoption of alternative transport options. While this group may not represent the majority, their willingness to adopt sustainable transport presents a promising target for future interventions.
Policies targeting this group could include expanding the infrastructure for walking and cycling and promoting shared mobility services. Additionally, policies that support company-provided bicycles or e-scooters could further encourage sustainable transport practices in urban environments. One study found that offering subsidized commuter tickets, bicycle benefits, and shared e-bikes/e-scooters successfully introduced more sustainable commuting options [76]. These initiatives not only reduced car usage but also improved employee well-being and increased physical activity [77]. However, the success of such programs depends on employers’ commitment to implementation and ongoing support, including providing appropriate facilities such as secure bike parking and showers [77]. Employers and municipalities should therefore work together to create an environment that facilitates and supports these sustainable mobility options.
Given the high willingness to shift among individuals in this cluster, initiatives aimed at increasing the availability and convenience of sustainable transport options could significantly impact overall mobility patterns. The long-term goal should be to establish a balanced mobility ecosystem where users have diverse options to choose from based on convenience, affordability, and sustainability.
The findings underscore the need for tailored policy interventions that address the distinct needs and preferences of different user groups. For those in Cluster One, improving PT quality and offering inclusive, family-oriented solutions is key. For Cluster Two, enhancing PT services and providing targeted outreach on the benefits of sustainable transport could increase the willingness to adopt alternative modes. Finally, Cluster Three represents a promising target group for promoting sustainable transport initiatives, with a clear opportunity to expand infrastructure and services that support walking, cycling, and shared mobility. These strategies should contribute to the overarching goal of reducing car dependency and fostering more accessible and sustainable urban environments.
Furthermore, the findings align with the existing literature, reinforcing that external and infrastructural barriers significantly influence users’ willingness to adopt sustainable transport modes. Users’ reluctance to use alternatives such as e-scooters, bicycles, and walking is often driven by concerns over safety [51], inadequate infrastructure, and the overall convenience of existing transport options. Specifically, despite incentives like company subsidies for bicycle or e-scooter purchases, a considerable number of employees remain uninterested. Additionally, dissatisfaction with PT travel time, punctuality, and comfort remains a key deterrent to increased usage. These insights reinforce the importance of integrating intermodal solutions, ensuring better adherence to timetables, and adopting a user-centered approach to urban mobility planning. Addressing these infrastructural and safety barriers is essential to creating a mobility system that supports both sustainable transport and the broader goals of reducing carbon emissions and improving quality of life.
The commuting behaviors observed in this study share similarities with segmentations found in other urban mobility studies. For instance, Anable identified a group of “Complacent Car Addicts” in the UK, who are resistant to mode shift—like our first cluster—highlighting how car dependency is often rooted in convenience and necessity, such as family obligations [78]. Conversely, our third cluster aligns with the “Aspiring Environmentalists” or “Active Switchers” observed in studies from Denmark and Germany, where individuals are more open to using sustainable transport if appropriate infrastructure and incentives exist [79]. However, compared to Northern European cases, Italian participants demonstrated lower baseline satisfaction with PT and more significant concerns about reliability and accessibility [80], suggesting that service quality may be a stronger barrier in the Italian context. This finding highlights the importance of tailoring mobility policies to specific cultural, infrastructural, and socio-economic contexts.
Policy Recommendations:
  • Cluster One (car-dependent group with negative PT perceptions):
    ○
    Improve PT services: Focus on reducing travel times, enhancing comfort, and optimizing stop locations.
    ○
    Providing family-oriented mobility solutions (e.g., car-sharing, accessible transport for caregivers).
    ○
    Incentives for car-sharing schemes in areas with limited PT infrastructure.
  • Cluster Two (car users with moderate PT satisfaction):
    ○
    Enhance PT quality to increase competitiveness with private vehicles.
    ○
    Launch communication campaigns highlighting PT’s benefits.
    ○
    Improve real-time information and user-provider communication.
  • Cluster Three (open to sustainable transport options):
    ○
    Expand infrastructure for walking, cycling, and shared mobility.
    ○
    Encourage employer-provided bicycles and e-scooters.
    ○
    Foster shared mobility systems to facilitate mode shift.
These tailored strategies aim to reduce car dependency, support sustainable commuting habits, and improve PT systems, contributing to the development of more inclusive, efficient, and environmentally responsible urban mobility systems.

6. Conclusions

This study emphasizes the value of understanding employees’ transport preferences and perceptions to design effective urban mobility policies. The clustering analysis identified three distinct groups with varying attitudes toward transport modes, emphasizing the need for targeted interventions rather than a one-size-fits-all approach. This tailored strategy can enhance the effectiveness of urban mobility initiatives, ensuring that policies resonate with the specific needs and preferences of each group. By acknowledging these diverse attitudes, city planners can foster a smoother transition towards more sustainable transportation options.
The first cluster consists of individuals heavily reliant on private cars, mainly due to family obligations, with strong dissatisfaction toward PT in terms of travel time, comfort, cost, and accessibility. Their low desire to shift to sustainable transport modes underscores the necessity of substantial PT improvements alongside complementary measures such as car-sharing and family-oriented mobility solutions. Addressing these concerns requires not only infrastructural enhancements but also tailored policies that consider caregiving responsibilities and accessibility challenges.
The second cluster, while also car-oriented, exhibits slightly more favorable perceptions of PT but remains reluctant to change unless services become significantly more competitive. Enhancing service reliability, reducing travel time, and improving comfort, along with targeted outreach efforts to communicate PT benefits, could encourage a gradual shift toward sustainable transport. Bridging the communication gap between PT operators and users may also foster more positive perceptions and engagement.
The third cluster represents individuals more open to alternative mobility options, including walking, cycling, shared mobility, and company-provided bicycles or e-scooters. Their neutral perception of PT, combined with a high willingness to adopt sustainable transport, presents an opportunity for policy interventions focused on expanding active mobility infrastructure, promoting shared mobility services, and implementing workplace initiatives.
Employer-backed incentives, such as subsidized bicycle programs, have the potential to promote sustainable commuting and improve employee well-being, though their effectiveness may be limited by other factors influencing employees’ mobility choices.
Addressing the barriers mentioned in the discussion section requires an integrated approach that combines PT enhancements, intermodal solutions, and investments in active mobility infrastructure to create a more sustainable and accessible urban transport system.
To deepen the understanding of commuting behavior, future research could benefit from incorporating psychological and social frameworks. Although prior studies have explored how major events such as the COVID-19 pandemic influence PT use [81,82], the effects of other disruptions—such as energy or social crises—remain underexplored and offer valuable avenues for further investigation. Applying models like the Theory of Planned Behavior [83] offers a useful lens for examining how individual attitudes, social pressures, and perceived behavioral control contribute to travel mode choices. Employing such approaches may offer more comprehensive insights into strategies for promoting behavioral change.
This study provides valuable insights but is subject to certain limitations. It is based on self-reported information, particularly concerning participants who stated a willingness to change their mode of transport. Response biases may influence such data, as individuals may misjudge or inaccurately express their readiness to adopt alternative travel behaviors. Additionally, responses may be affected by recall bias or a desire to provide socially acceptable answers. The absence of objective behavioral data limits the extent to which these self-reported intentions can be verified. Future research should consider alternative approaches to address these limitations and strengthen the validity of the findings.
Overall, these findings reinforce that fostering a shift toward sustainable transport requires a multifaceted strategy tailored to different commuter groups. While PT improvements are essential, complementary initiatives such as car-sharing, investments in active mobility, family-oriented mobility services, and behavioral incentives are equally crucial. A holistic approach that considers both infrastructural and socio-economic factors is necessary to reduce car dependency and support more sustainable urban mobility choices.
While this study focuses on commuting behaviors in Italy, its findings offer broader insights that may be applicable to other countries facing similar urban transport challenges, such as car dependency, limited PT satisfaction, and increasing interest in active and shared mobility. The identified clusters—ranging from car-dependent individuals with caregiving responsibilities to those more open to alternative modes—reflect behavioral patterns that are not unique to the Italian context. In many European and international cities, factors such as family obligations, service quality perceptions, and personal attitudes similarly influence mode choice [78,84]. However, the transferability of these results depends on contextual elements, including cultural attitudes toward mobility, the quality and integration of PT systems, urban density, and national transport policies. Countries with well-established cycling cultures or highly integrated transport networks, such as the Netherlands and Denmark, may exhibit different commuter segmentations and greater readiness for modal shift [79]. Therefore, while our methodological approach and policy implications offer a useful framework, local adaptations are essential for effectively applying these insights in other settings. Comparative studies are encouraged to validate and refine these findings across diverse geographic and socio-economic contexts.

Author Contributions

Conceptualization, M.B. and M.V.C.; methodology, M.B.; software, M.B.; validation, M.B.; formal analysis, M.B.; investigation, G.G.; resources, G.G.; data curation, G.G.; writing—original draft preparation, M.B.; writing—review and editing, M.B. and M.V.C.; visualization, M.B.; supervision, G.G. and M.V.C.; project administration, G.G.; All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Ethical review and approval were waived for this study in accordance with Sapienza University’s Code of Ethics and Conduct, as the research did not involve human health interventions and complied with applicable legal regulations on informed consent and data protection.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to privacy and confidentiality concerns. The dataset contains sensitive information, such as the geographical locations of employees and companies, which could potentially identify individuals or reveal confidential business information.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Diao, M.; Kong, H.; Zhao, J. Impacts of transportation network companies on urban mobility. Nat. Sustain. 2024, 4, 494–500. [Google Scholar] [CrossRef]
  2. United Nations. Goal 11: Make Cities Inclusive, Safe, Resilient, and Sustainable. United Nations, n.d. Available online: https://www.un.org/sustainabledevelopment/cities/ (accessed on 11 February 2025).
  3. Triandis, H.C. Values, attitudes, and interpersonal behavior. Neb. Symp. Motiv. 1980, 27, 195–259. [Google Scholar]
  4. Havlíčková, D.; Zámečník, P. Considering habit in research on travel mode choice: A literature review with a two-level methodology. Trans. Transp. Sci. 2020, 11, 18–32. [Google Scholar] [CrossRef]
  5. Pourramazani, H.; Miralles-Garcia, J.L. Exploring the link between travel behaviour and sustainable mobility. Sustain. Cities Soc. 2022, 260, 217–228. [Google Scholar]
  6. Willis, D.P.; Manaugh, K.; El-Geneidy, A. Cycling under influence: Summarizing the influence of perceptions, attitudes, habits, and social environments on cycling for transportation. Int. J. Sustain. Transp. 2015, 9, 565–579. [Google Scholar] [CrossRef]
  7. Piras, F.; Sottile, E.; Meloni, I. Do psycho-attitudinal factors vary with individuals’ cycling frequency? A hybrid ordered modeling approach. Travel Behav. Soc. 2021, 22, 186–198. [Google Scholar] [CrossRef]
  8. Trolese, M.; De Fabiis, F.; Coppola, P. A walkability index including pedestrians’ perception of built environment: The case study of Milano Rogoredo Station. Sustainability 2023, 15, 15389. [Google Scholar] [CrossRef]
  9. Renne, J.L.; Bennett, P. Socioeconomics of urban travel: Evidence from the 2009 National Household Travel Survey with implications for sustainability. World Transp. Policy Pract. 2014, 20, 7–27. [Google Scholar]
  10. European Platform on Mobility Management (EPOMM). POMM Strategy Book-A Strategy for ECOMM and EPOMM; EPOMM: Brussels, Belgium, 2018; Available online: https://epomm.eu/sites/default/files/files/EPOMM_strategy_book.pdf (accessed on 11 February 2025).
  11. European Platform on Mobility Management (EPOMM). Mobility Management; EPOMM: Brussels, Belgium, 2008; Available online: https://epomm.eu/about/mobility-management (accessed on 11 February 2025).
  12. Ministero dell’Ambiente. Decreto 27 marzo 1998—Programma di Interventi per il Miglioramento Della Mobilità Urbana. M. dell’Ambiente. 1998. Available online: http://www.cittametropolitana.torino.it/cms/risorse/ambiente/dwd/mobilita-sostenibile/moma/Decreto27_Mar_1998.pdf (accessed on 11 February 2025).
  13. European Parliament & Council of the European Union. Directive 2014/95/EU on Non-Financial and Diversity Information. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32014L0095 (accessed on 11 February 2025).
  14. European Commission. Commission Staff Working Document: Corporate Social Responsibility, Responsible Business Conduct, and Business and Human Rights: Overview of Progress. European Commission. 2019. Available online: https://ec.europa.eu/docsroom/documents/34482 (accessed on 12 February 2025).
  15. EU Urban Mobility Observatory. Having a Mobility Manager Is Now Mandatory for Companies and Municipalities in Italy. EU Urban Mobility Observator. 12 July 2021. Available online: https://urban-mobility-observatory.transport.ec.europa.eu/news-events/news/having-mobility-manager-now-mandatory-companies-and-municipalities-italy-2021-07-12_en?prefLang=et (accessed on 12 February 2025).
  16. Hung, P.-H.; Doi, K.; Inoi, H. User retention tendency of bus routes based on user behavior transition in an area with low mode share of public transport. IATSS Res. 2020, 44, 111–124. [Google Scholar] [CrossRef]
  17. Meyer, M.D. Demand management as an element of transportation policy: Using carrots and sticks to influence travel behavior. Transp. Res. Part A 1999, 33, 575–599. [Google Scholar] [CrossRef]
  18. Black, C.S.; Schreffler, E. N Understanding transport demand management and its role in delivery of sustainable urban transport. Transp. Res. Rec. 2010, 2163, 81–88. [Google Scholar] [CrossRef]
  19. Esztergár-Kiss, D.; Braga Zagabria, C. Method development for workplaces using mobility plans to select suitable and sustainable measures. Res. Transp. Bus. Manag. 2021, 40, 100544. [Google Scholar] [CrossRef]
  20. Jiménez, P.; María-Dolores, D.; Beltrán, S. An integrative and sustainable workplace mobility plan: The case study of Navantia-Cartagena (Spain). Sustainability 2020, 12, 10301. [Google Scholar] [CrossRef]
  21. Wake, D. Engaging Perth Workplaces for Sustainable Transport: An Evaluation of the TravelSmart Workplace Program. 2012. Available online: https://www.semanticscholar.org/paper/Engaging-Perth-workplaces-for-sustainable-an-of-the-Wake/3ca92e2c78b25a7c458e64cc728a60008a555a9a (accessed on 12 February 2025).
  22. Mathew, S.; Iyyappan, M.; Kiran, S.S. Green commuting strategies for corporate sustainability for optimized reduction of CO2 emissions. In Proceedings of the 2023 International Conference on New Frontiers in Communication, Automation, Management and Security (ICCAMS), Bangalore, India; 2023; pp. 1–5. [Google Scholar]
  23. Batty, P.; Palacin, R.; González-Gil, A. Challenges and opportunities in developing urban modal shift. Travel Behav. Soc. 2015, 2, 109–123. [Google Scholar] [CrossRef]
  24. Saif, M.A.; Zefreh, M.M.; Torok, A. Public transport accessibility: A literature review. Period. Polytech. Transp. Eng. 2019, 47, 36–43. [Google Scholar] [CrossRef]
  25. United Nations Economic Commission for Europe. Sustainable urban mobility and public transport development. In Sustainable Urban Mobility and Public Transport in UNECE Capitals; United Nations: New York, NY, USA, 2016. [Google Scholar]
  26. Etukudoh, E.A.; Adefemi, A.; Ilojianya, V.I.; Umoh, A.A.; Ibekwe, K.I.; Nwokediegwu, Z.Q.S. A review of sustainable transportation solutions: Innovations, challenges, and future directions. World J. Adv. Res. Rev. 2024, 21, 1440–1452. [Google Scholar]
  27. Babapourdijojin, M.; Gentile, G. Assessing the mobility impact on corporate social responsibility. In Reliability and Statistics in Transportation and Communication; Kabashkin, I., Yatskiv, P.O., Eds.; Springer International Publishing: Berlin/Heidelberg, Germany, 2023; pp. 320–335. [Google Scholar]
  28. Babapourdijojin, M.; Gentile, G. Assessing the benefits and external costs of road transport in Italy. In Proceedings of the 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems, Nice, France, 14–16 June 2023; pp. 1–6. [Google Scholar]
  29. Canitez, F. Sustainable urban mobility in Istanbul: Transition and planning. J. Intell. Decis. Making Inform. Sci. 2024, 1, 45–54. [Google Scholar] [CrossRef]
  30. Marletto, G.E. Structure, agency and change in the car regime: A review of the literature. Eur. Transp. Trasp. Eur. 2011, 47, 71–88. [Google Scholar]
  31. Kent, J.L. Secured by Automobility: Why Does the Private Car Continue to Dominate Transport Practices? Ph.D. Thesis, University of New South Wales, Sydney, Australia, 2013. Available online: https://www.be.unsw.edu.au/sites/default/files/upload/pdf/cf/hbep/research/JK%20Thesis_final_encrypted.pdf (accessed on 12 February 2025).
  32. Lee, D.; Matthews, T.; Imran, M. Discursive Barriers to Sustainable Urban Transport in Auckland. 2010. Available online: https://www.semanticscholar.org/paper/Discursive-barriers-to-sustainable-urban-transport-Lee-Matthews/8174e8bf78e8a2269ca9169851eb880727326690 (accessed on 12 February 2025).
  33. Ramos, S.; Vicente, P.; Passos, A.M.; Costa, P.L.; Reis, E. Perceptions of the public transport service as a barrier to the adoption of public transport: A qualitative study. Soc. Sci. 2019, 8, 150. [Google Scholar] [CrossRef]
  34. Glerum, A.; Atasoy, B.; Monticone, A.; Bierlaire, M. Adjectives qualifying individuals’ perceptions impacting on transport mode preferences. In Proceedings of the International Conference on Management of Change, Leeds, UK, 4 July 2011; Available online: https://transp-or.epfl.ch/documents/proceedings/GleAtaMonBie_ICMC2011.pdf (accessed on 12 February 2025).
  35. Vicente, P.; Reis, E. Profiling public transport users through perceptions about public transport providers and satisfaction with the public transport service. Public Transp. 2016, 8, 387–403. [Google Scholar] [CrossRef]
  36. Sinha, S.; Swamy, H.M.S.; Modi, K. User perceptions of public transport service quality. Transp. Res. Procedia 2020, 48, 3310–3323. [Google Scholar] [CrossRef]
  37. Babapourdijojin, M.; Corazza, M.V.; Gentile, G. Systematic Analysis of Commuting Behavior in Italy Using K-Means Clustering and Spatial Analysis: Towards Inclusive and Sustainable Urban Transport Solutions. Future Transp. 2024, 4, 1430–1456. [Google Scholar] [CrossRef]
  38. Adams, E.J.; Esliger, D.W.; Taylor, I.M.; Sherar, L.B. Individual, employment, and psychosocial factors influencing walking to work: Implications for intervention design. PLoS ONE 2017, 12, e0171374. [Google Scholar] [CrossRef] [PubMed]
  39. Bina, M.; Biassoni, F. Travel experience and reasons for the use and nonuse of local public transport: A case study within the Community Interregional Project SaMBA (Sustainable Mobility Behaviors in the Alpine Region). Sustainability 2023, 15, 24. [Google Scholar] [CrossRef]
  40. Petrunoff, N.; Wen, L.M.; Rissel, C. Effects of a workplace travel plan intervention encouraging active travel to work: Outcomes from a three-year time-series study. Public Health 2016, 135, 38–47. [Google Scholar] [CrossRef]
  41. Guzman, L.A.; Arellana, J.; Alvarez, V. Confronting congestion in urban areas: Developing sustainable mobility plans for public and private organizations in Bogotá. Transp. Res. Part A Policy Pract. 2020, 134, 321–335. [Google Scholar] [CrossRef]
  42. Šucha, M. Car-free life and factors influencing travel mode choice. Eur. Transp. Trasp. Eur. 2023, 93, 9. [Google Scholar] [CrossRef]
  43. Ramus, C.A. Organizational support for employees: Encouraging creative ideas for environmental sustainability. Calif. Manag. Rev. 2001, 43, 105–125. [Google Scholar] [CrossRef]
  44. Petrunoff, N.; Rissel, C.; Wen, L.M. ‘If you don’t do parking management, forget your behaviour change, it’s not going to work’: Health and transport practitioner perspectives on workplace active travel promotion. PLoS ONE 2017, 12, e0170064. [Google Scholar] [CrossRef]
  45. Nogueira, M.; Dias, F.; Santos, V. Sustainable mobility choices: Exploring the impact of consumers’ values, attitudes, perceived behavioural control and subjective norms on the likelihood to choose sustainable mobility options. J. Consum. Behav. 2023, 22, 511–528. [Google Scholar] [CrossRef]
  46. Rodríguez-Rad, C.J.; Revilla-Camacho, M.-Á.; Sánchez-del-Río-Vázquez, M.-E. Exploring the intention to adopt sustainable mobility modes of transport among young university students. Int. J. Environ. Res. Public Health 2023, 20, 3196. [Google Scholar] [CrossRef] [PubMed]
  47. Gómez-Ortega, A.; Flores-Ureba, S.; Gelashvili, V.; Jalón, M.L.D. Users’ perception for innovation and sustainability management: Evidence from public transport. Rev. Manag. Sci. 2024, 18, 859–882. [Google Scholar] [CrossRef]
  48. Chen, R.; Zhang, J.; Ravishanker, N.; Konduri, K. Clustering activity–travel behavior time series using topological data analysis. J. Big Data Anal. Transp. 2019, 1, 109–121. [Google Scholar] [CrossRef]
  49. Zamprogno, M.M.; Esztergár-Kiss, D. Applying cluster analysis for the investigation of travel behavior and user profiles. Period. Polytech. Transp. Eng. 2024, 52, 362–371. [Google Scholar] [CrossRef]
  50. Zhao, P.; Bucher, D.; Martin, H.; Raubal, M. A clustering-based framework for understanding individuals’ travel mode choice behavior. In Proceedings of the 22nd AGILE Conference on Geographic Information Science, Limassol, Cyprus, 17–21 June 2019. [Google Scholar]
  51. Dibaj, S.; Hosseinzadeh, A.; Mladenović, M.N.; Kluger, R. Where Have Shared E-Scooters Taken Us So Far? A Review of Mobility Patterns, Usage Frequency, and Personas. Sustainability 2021, 13, 11792. [Google Scholar] [CrossRef]
  52. Tejada, J.J.; Punzalan, J.R.B. On the misuse of Slovin’s formula. Philipp. Stat. 2012, 61, 129–136. [Google Scholar]
  53. Dinh, T.; Hauchi, W.; Fournier-Viger, P.; Lisik, D.; Ha, M.-Q.; Dam, H.-C.; Huynh, V.-N. Categorical data clustering: 25 years beyond K-modes. Expert Syst. Appl. 2025, 272, 126608. [Google Scholar] [CrossRef]
  54. Huang, Z. Extensions to the k-Means Algorithm for Clustering Large Data Sets with Categorical Values. Data Min. Knowl. Discov. 1998, 2, 283–304. [Google Scholar] [CrossRef]
  55. Goyal, M.K.; Aggarwal, S. A review on K-mode clustering algorithm. Int. J. Adv. Res. Comput. Sci. 2017, 8, 725–729. [Google Scholar] [CrossRef]
  56. Chadha, A.; Kumar, S. Extension of K-modes algorithm for generating clusters automatically. Int. J. Inf. Technol. Comput. Sci. 2016, 8, 51–57. [Google Scholar] [CrossRef]
  57. Yin, H.; Aryani, A.; Petrie, S.; Nambissan, A.; Astudillo, A.; Cao, S. A rapid review of clustering algorithms. arXiv 2024, arXiv:2401.07389. [Google Scholar]
  58. Bholowalia, P.; Kumar, A. EBK-means: A clustering technique based on elbow method and K-means in WSN. Int. J. Comput. Appl. 2014, 105, 17–24. [Google Scholar]
  59. Davies, D.L.; Bouldin, D.W. A cluster separation measure. IEEE Trans. Pattern Anal. Mach. Intell. 1979, 1, 224–227. [Google Scholar] [CrossRef] [PubMed]
  60. Shahapure, K.R.; Nicholas, C.K. Cluster quality analysis using silhouette score. In Proceedings of the 7th IEEE International Conference on Data Science and Advanced Analytics (DSAA), Sydney, Australia, 6–9 October 2020; IEEE: New York, NY, USA, 2020; pp. 747–748. [Google Scholar]
  61. Rousseeuw, P.J. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J. Comput. Appl. Math. 1987, 20, 53–65. [Google Scholar] [CrossRef]
  62. Ceder, A.; Dressler, O. A note on the χ2 test with applications to road accidents in construction zones. Accid. Anal. Prev. 1980, 12, 7–10. [Google Scholar] [CrossRef]
  63. Spiegelman, C.; Park, E.S.; Rilett, L.R. Transportation Statistics and Microsimulation; Chapman and Hall/CRC: Boca Raton, FL, USA, 2016. [Google Scholar]
  64. Sankari, J.; Senthilnathan, M.D.; Parasakthi, D. A study on consumer satisfaction about electric bikes (E-bikes) with special reference to Coimbatore city. Int. J. Res. Trends Innov. 2022, 7, 867–874. [Google Scholar]
  65. Husson, F.; Josse, J. Multiple Correspondence Analysis. In Visualization and Verbalization of Data; Blasius, J., Greenacre, M., Eds.; CRC Press: Boca Raton, FL, USA, 2014; pp. 165–184. [Google Scholar]
  66. Kamalja, K.K.; Khangar, N.V. Multiple Correspondence Analysis and its applications. Electron. J. Appl. Stat. Anal. 2017, 10, 432–462. [Google Scholar]
  67. Broeksema, B.; Telea, A.C.; Baudel, T. Visual Analysis of Multi-Dimensional Categorical Data Sets. Comput. Graph. Forum 2013, 32, 158–169. [Google Scholar] [CrossRef]
  68. Hwang, H.; Dillon, W.R.; Takane, Y. Fuzzy Cluster Multiple Correspondence Analysis. Behaviormetrika 2010, 37, 111–133. [Google Scholar] [CrossRef]
  69. Dibaj, S.; Keiichi, H.; Goto, R.; Taniguchi, A.; Mladenović, M.N. Latent class approach to classify e-scooter non-users: A comparative study of Helsinki and Tokyo. Travel Behav. Soc. 2025, 39, 100941. [Google Scholar] [CrossRef]
  70. Teixeira, J.F.; Diogo, V.; Bernát, A.; Lukasiewicz, A.; Vaiciukynaite, E.; Sanna, V.S. Barriers to bike and e-scooter sharing usage: An analysis of non-users from five European capital cities. Case Stud. Transp. Policy 2023, 13, 101045. [Google Scholar] [CrossRef]
  71. Sarker, R.I.; Mailer, M.; Sikder, S.K. Walking to a Public Transport Station: Empirical Evidence on Willingness and Acceptance in Munich, Germany. Smart Sustain. Built. Environ. 2020, 9, 38–53. [Google Scholar] [CrossRef]
  72. Ziari, H.; Keymanesh, M.R.; Khabiri, M.M. Locating stations of public transportation vehicles for improving transit accessibility. Transport 2007, 22, 99–104. [Google Scholar] [CrossRef]
  73. Truden, C.; Kollingbaum, M.J.; Reiter, C.; Schasché, S.E. GIS-Based Analysis of Reachability Aspects in Rural Public Transportation. Case Stud. Transp. Policy 2022, 10, 1827–1840. [Google Scholar] [CrossRef]
  74. Unbehaun, W.; Uhlmann, T.; Hössinger, R.; Leisch, F.; Gerike, R. Women and Men with Care Responsibilities in the Austrian Alps: Activity and Mobility Patterns of a Diverse Group. Mountain Res. Dev. 2014, 34, 276–290. [Google Scholar] [CrossRef]
  75. Fian, T.; Hauger, G. Composing a Conceptual Framework for an Inclusive Mobility System. IOP Conf. Ser. Mater. Sci. Eng. 2020, 960, 032089. [Google Scholar] [CrossRef]
  76. Villanen, M.; Vanhamäki, S.; Hämäläinen, R.-M. Encouraging Sustainable Mobility: Community Case Study on Workplace Initiatives in Lahti, Finland. Front. Sustain. 2023, 4, 1158231. [Google Scholar] [CrossRef]
  77. Hämäläinen, R.-M. Promoting Mobility Among Wellbeing Employees in Lahti, Finland. Eur. J. Public Health 2023, 33 (Suppl. S1), ckad133.240. [Google Scholar] [CrossRef]
  78. Anable, J. ‘Complacent car addicts’ or ‘aspiring environmentalists’? Identifying travel behaviour segments using attitude theory. Transp. Policy 2005, 12, 65–78. [Google Scholar] [CrossRef]
  79. Pucher, J.; Buehler, R. Making cycling irresistible: Lessons from the Netherlands, Denmark, and Germany. Transp. Rev. 2008, 28, 495–528. [Google Scholar] [CrossRef]
  80. Dell’Olio, L.; Ibeas, A.; Cecin, P. The quality of service desired by public transport users. Transp. Policy 2011, 18, 217–227. [Google Scholar] [CrossRef]
  81. Corazza, M.V.; Moretti, L.; Forestieri, G.; Galiano, G. Chronicles from the new normal: Urban planning, mobility and land-use management in the face of the COVID-19 crisis. Transp. Res. Interdiscip. Perspect. 2021, 12, 100503. [Google Scholar] [CrossRef] [PubMed]
  82. Corazza, M.V.; Musso, A. Urban transport policies in the time of pandemic, and after: An ARDUOUS research agenda. Transp. Policy 2021, 103, 31–44. [Google Scholar] [CrossRef]
  83. Ajzen, I. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 1991, 50, 179–211. [Google Scholar] [CrossRef]
  84. Gärling, T.; Axhausen, K.W. Introduction: Habitual travel choice. Transportation 2003, 30, 1–11. [Google Scholar] [CrossRef]
Sustainability 17 05149 g001
Figure 1. Research methodology workflow.
Figure 1. Research methodology workflow.
Sustainability 17 05149 g001
Sustainability 17 05149 g002
Figure 2. Cities with at least 30 responses each. (Source: Authors).
Figure 2. Cities with at least 30 responses each. (Source: Authors).
Sustainability 17 05149 g002
Sustainability 17 05149 g003
Figure 3. Proportion of responses by city in study dataset.
Figure 3. Proportion of responses by city in study dataset.
Sustainability 17 05149 g003
Sustainability 17 05149 g004
Figure 4. Correlations among different PT service factors.
Figure 4. Correlations among different PT service factors.
Sustainability 17 05149 g004
Sustainability 17 05149 g005
Figure 5. Elbow Method applied to all variables.
Figure 5. Elbow Method applied to all variables.
Sustainability 17 05149 g005
Sustainability 17 05149 g006
Figure 6. Silhouette Score applied to all variables.
Figure 6. Silhouette Score applied to all variables.
Sustainability 17 05149 g006
Sustainability 17 05149 g007
Figure 7. Davies–Bouldin Index (DBI) applied to all variables.
Figure 7. Davies–Bouldin Index (DBI) applied to all variables.
Sustainability 17 05149 g007
Sustainability 17 05149 g008
Figure 8. The Elbow Method applied to significant variables.
Figure 8. The Elbow Method applied to significant variables.
Sustainability 17 05149 g008
Sustainability 17 05149 g009
Figure 9. The Silhouette Score applied to significant variables.
Figure 9. The Silhouette Score applied to significant variables.
Sustainability 17 05149 g009
Sustainability 17 05149 g010
Figure 10. The Davies–Bouldin Index (DBI) applied to significant variables.
Figure 10. The Davies–Bouldin Index (DBI) applied to significant variables.
Sustainability 17 05149 g010
Sustainability 17 05149 g011
Figure 11. Number of cases in each cluster.
Figure 11. Number of cases in each cluster.
Sustainability 17 05149 g011
Sustainability 17 05149 g012
Figure 12. K-modes clustering results with 3 groups (MCA projection).
Figure 12. K-modes clustering results with 3 groups (MCA projection).
Sustainability 17 05149 g012
Sustainability 17 05149 g013
Figure 13. Employees’ willingness to walk, use e-scooters, or ride bicycles.
Figure 13. Employees’ willingness to walk, use e-scooters, or ride bicycles.
Sustainability 17 05149 g013
Table 1. List of used dummy variables.
Table 1. List of used dummy variables.
CategoryOptions
Demographic VariableGender (Female)
Age Group: Under 35 years
Age Group: 36–40 years
Age Group: 41–55 years
Age Group: 56–60 years
Age Group: Over 60 years
Family Members with Independent Travel Limitations
Travel Behavior VariablesMode Choice: Motorcycle
Mode Choice: PT
Mode Choice: Car
Mode Choice: Bicycle or E-Scooter
Mode Choice: Walking
Mode Choice: Multimodal Transport
Travel Cost: Less than EUR 20
Travel Cost: EUR 21–40
Travel Cost: EUR 41–60
Travel Cost: EUR 61–80
Travel Cost: EUR 81–100
Travel Cost: More than EUR 100
Experience with Shared Mobility Services
Reasons for Travel Mode ChoiceReason for Mode Choice: Cost
Reason for Mode Choice: Travel Time
Reason for Mode Choice: Independence During Trip
Reason for Mode Choice: Lack of PT Availability
Reason for Mode Choice: Reduced Stress
Reason for Mode Choice: Parking Difficulties
Reason for Mode Choice: Accompanying Others
Need for Stops During Trip
Perceptions of PT ServicesPT Service Rating: Travel Time (Poor)
PT Service Rating: Travel Time (Fair)
PT Service Rating: Travel Time (Good)
PT Service Rating: Travel Time (Very Good)
PT Service Rating: Comfort (Poor)
PT Service Rating: Comfort (Fair)
PT Service Rating: Comfort (Good)
PT Service Rating: Comfort (Very Good)
Perceptions of PT Services PT Service Rating: Cost (Poor)
PT Service Rating: Cost (Fair)
PT Service Rating: Cost (Good)
PT Service Rating: Cost (Very Good)
PT Service Rating: Proximity to Stops (Poor)
PT Service Rating: Proximity to Stops (Fair)
PT Service Rating: Proximity to Stops (Good)
PT Service Rating: Proximity to Stops (Very Good)
Sustainable Transport PreferencesWillingness to Walk
Willingness to Walk with Colleagues
Willingness to Use E-Scooter
Willingness to Use Shared Mobility Services
Willingness to Use Bicycle or E-Scooter Provided by Employer
Table 2. Chi-square test results.
Table 2. Chi-square test results.
VariablesChi-Squarep-Value
PT Service Rating: Travel Time (Poor)836.350.000
PT Service Rating: Proximity to Stops (Poor)815.680.000
PT Service Rating: Comfort (Poor)803.690.000
PT Service Rating: Cost (Poor)619.300.000
PT Service Rating: Comfort (Fair)557.090.000
Willingness to Use Bicycle or E-Scooter Provided by Employer529.060.000
PT Service Rating: Travel Time (Fair)526.80.000
PT Service Rating: Cost (Fair)502.150.000
Mode Choice: Car482.440.000
Willingness to Walk449.10.000
PT Service Rating: Proximity to Stops (Fair)439.660.000
Willingness to Use Shared Mobility Services328.160.000
PT Service Rating: Travel Time (Good)256.270.000
PT Service Rating: Comfort (Good)246.870.000
Willingness to Use E-Scooter217.260.000
Mode Choice: PT196.10.000
PT Service Rating: Proximity to Stops (Good)189.750.000
Reason for Mode Choice: Cost158.550.000
Family Members with Independent Travel Limitations131.780.000
PT Service Rating: Proximity to Stops (Very Good)121.210.000
PT Service Rating: Cost (Good)104.890.000
Reason for Mode Choice: Lack of PT Availability96.380.000
PT Service Rating: Travel Time (Very Good)88.320.000
Mode Choice: Bicycle or E-Scooter75.470.000
PT Service Rating: Comfort (Very Good)75.280.000
Mode Choice: Walking70.330.000
PT Service Rating: Cost (Very Good)65.650.000
Travel Cost: More than EUR 10060.270.000
Willingness to Walk with Colleagues50.860.000
Reason for Mode Choice: Independence during Trip50.560.000
Reason for Mode Choice: Parking Difficulties48.520.000
Age Group: Under 35 years48.220.000
Gender (Female)43.550.000
Travel Cost: EUR 21–4038.720.000
Age Group: 41–55 years35.50.000
Experience with Shared Mobility Services35.450.000
Mode Choice: Multimodal Transport35.050.000
Travel Cost: Less than EUR 2033.250.000
Need for Stops during Trip25.730.000
Reason for Mode Choice: Accompanying Others23.830.000
Mode Choice: Motorcycle18.720.000
Reason for Mode Choice: Reduced Stress13.30.001
Travel Cost: EUR 81–10011.780.003
Age Group: Over 60 years11.370.003
Travel Cost: EUR 61–804.420.110
Reason for Mode Choice: Travel Time2.270.320
Table 3. Employees’ perceptions about various aspects of PT.
Table 3. Employees’ perceptions about various aspects of PT.
Information Travel Time Comfort
Count% Count% Count%
Poor65128.3%60%105445.8%76%92540.2%80%
Fair72231.4%70130.5%90839.5%
Good71931.2% 41918.2% 40517.6%
Excellent2099.1% 1275.5% 632.7%
Cost Proximity to PT StopsPunctuality
Count% Count% Count%
Poor72531.5%73%77133.5%67%94441%72%
Fair95041.3%77233.6%71431%
Good51822.5% 54423.6% 49921.7%
Excellent1084.7% 2149.3% 1446.3%
Table 4. Characteristics, transport preferences, PT perceptions, and willingness to shift across clusters.
Table 4. Characteristics, transport preferences, PT perceptions, and willingness to shift across clusters.
Cluster OneCluster TwoCluster Three
Key CharacteristicsCar users with negative PT perceptions who rely on private cars due to family obligationsCar users with slightly better PT perceptionsIndividuals open to alternative mobility
Transport PreferencesStrong preference for car use; family travel needs limit alternativesPrimarily car users but with slightly improved PT perceptionPrefer walking, cycling, shared mobility, and company-provided bicycles or e-scooters
PT PerceptionsHigh dissatisfaction with PT (travel time, comfort, cost, proximity to stops)Fair ratings for PT services; moderate dissatisfactionNeutral PT perception; do not rely on cars
Willingness to ShiftVery low willingness to adopt alternative transportNo willingness to shift unless PT improvesHigh willingness to shift to sustainable transport
Cluster Size36%36%28%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Babapour, M.; Corazza, M.V.; Gentile, G. Urban Commuting Preferences in Italy: Employees’ Perceptions of Public Transport and Willingness to Adopt Active Transport Based on K-Modes Cluster Analysis. Sustainability 2025, 17, 5149. https://doi.org/10.3390/su17115149

AMA Style

Babapour M, Corazza MV, Gentile G. Urban Commuting Preferences in Italy: Employees’ Perceptions of Public Transport and Willingness to Adopt Active Transport Based on K-Modes Cluster Analysis. Sustainability. 2025; 17(11):5149. https://doi.org/10.3390/su17115149

Chicago/Turabian Style

Babapour, Mahnaz, Maria Vittoria Corazza, and Guido Gentile. 2025. "Urban Commuting Preferences in Italy: Employees’ Perceptions of Public Transport and Willingness to Adopt Active Transport Based on K-Modes Cluster Analysis" Sustainability 17, no. 11: 5149. https://doi.org/10.3390/su17115149

APA Style

Babapour, M., Corazza, M. V., & Gentile, G. (2025). Urban Commuting Preferences in Italy: Employees’ Perceptions of Public Transport and Willingness to Adopt Active Transport Based on K-Modes Cluster Analysis. Sustainability, 17(11), 5149. https://doi.org/10.3390/su17115149

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop