Sustainable Mobility in Barcelona: Trends, Challenges and Policies for Urban Decarbonization

Abstract

The Barcelona Metropolitan Area (AMB) has implemented various policies to reduce car use and promote more sustainable mobility. Initiatives such as superblocks, Low Emission Zones (LEZs), and the Bicivia network aim to transform the urban model in response to environmental and congestion challenges. However, the high reliance on private vehicles for intermunicipal travel, uneven infrastructure, and social resistance to certain changes remain significant issues. This study examines the evolution of mobility patterns and assesses the effectiveness of the above policies in fostering real and sustainable change. A mixed-methods approach was adopted, which combined an exploratory factor analysis (EFA) of 2011–2024 data, trend linear regression, and a comparative international analysis. The EFA identified four key structural dimensions: traditional transport infrastructure, active mobility and bus lines, public bicycles and mixed use, and transport efficiency and punctuality. The findings reveal a clear reduction in private car use and an increase in sustainable modes of transport. This indicates that there are prospects for future transformation. Nonetheless, challenges persist in intermunicipal mobility and the public acceptance of the measures. This study provides empirical and comparative evidence and emphasizes the need for integrated metropolitan governance to achieve a resilient and sustainable urban model.

1. Introduction

The phenomenon of global urbanization continues its rapid progression, with projections from UN-Habitat [1] indicating an increase from 56% to 68% of the world’s population residing in urban areas by 2050. This will represent a growth of approximately 2.2 billion inhabitants. Demographic growth poses significant challenges for urban mobility, especially in large metropolitan areas. In these contexts, the predominance of private vehicles exacerbates critical issues such as traffic congestion, air pollution, and the reduction in public space, which directly impacts citizens’ quality of life [2].

The transport sector is responsible for a substantial share of greenhouse gas emissions. According to the Intergovernmental Panel on Climate Change [3], this sector accounts for approximately 23% of global CO₂ emissions from energy-related sources. The prevalence of the ‘car-centered city’ model, characterized by an average occupancy of 1.2 passengers per vehicle, is not only intrinsically inefficient, but also a major contributor to air pollution and traffic accidents [4]. In this pressing context, sustainable mobility has emerged as a comprehensive solution to balance the environmental, social, and economic aspects of urban transport. It is structured around fundamental principles such as environmental sustainability, social equity, universal accessibility, economic viability, and technological integration, and promotes the design of cities that favor human interaction and active mobility [5,6]. These principles are also consistent with Jane Jacobs’ [7] historical critiques of the prioritization of motorized traffic and Jan Gehl’s [8] vision of cities designed on a human scale.

In Europe, various cities have successfully implemented sustainable mobility models, notably Amsterdam and Copenhagen. These cities have based their strategies on the strong promotion of cycling and the progressive restriction of car traffic. Amsterdam, for instance, has a consolidated cycling infrastructure that supports 36% of daily trips by bicycle [9], while Copenhagen has integrated cycling with its public transport system, achieving 21% daily bicycle use for commuting [10]. The AMB has implemented various policies to reduce car use and promote more sustainable mobility. Initiatives such as superblocks, green axes, Low Emission Zones (LEZs), and the Bicivia network seek to transform the urban model in the face of environmental and congestion challenges. However, according to previous research and despite the progress toward more sustainable mobility, the AMB continues to face a high reliance on private vehicles for intermunicipal journeys, territorial disparities in transport infrastructure, and considerable social resistance to certain policy interventions [11]. Moreover, current mobility challenges are shaped by shifts in travel behavior following the pandemic, gender-related factors, and household composition [12], along with persistent environmental concerns such as prolonged exposure to elevated nitrogen dioxide levels [13]. These multifaceted pressures highlight the need for more integrated and forward-looking analyses of urban mobility in metropolitan contexts.

2. Materials and Methods

2.1. The Field of Study

The study area of this research is the Barcelona Metropolitan Area (AMB), a complex urban region encompassing 36 municipalities with a population exceeding 3.2 million inhabitants and an approximate area of 636 km² (Figure 1). The city of Barcelona, its main center, has undergone an intense process of urban densification over the past 150 years. This has shaped an environment characterized by a strong dependence on automobiles. Despite efforts to diversify modal share, private vehicles continue to occupy a disproportionate amount of space. Currently, 60% of the metropolitan road space is allocated to private vehicle traffic and parking, even though this mode of transport accounts for only 25% of total trips [14]. The unequal distribution of urban road space has long contributed to critical structural issues such as chronic traffic congestion, elevated levels of air pollution, and the degradation of the quality of public space, all of which have significant implications for urban livability and public health. In the municipality of Barcelona, sidewalks represented only 34.4% of the street surface in 2000, a figure that had increased to 45.5% by 2023 [15], which signals ongoing efforts to reverse the historical dominance of private motorized transport and to reclaim urban space for pedestrians. However, despite these improvements, private vehicles continue to occupy a disproportionate share of public road infrastructure relative to their actual modal share. This mismatch is particularly evident at the metropolitan scale. According to the Pla Director Urbanístic Metropolità (Urban Metropolitan Master Plan, PDUM), 46.6% of total daily trips in the metropolitan area are made through active modes, predominantly walking (44.3%), while only 28.7% are carried out using private vehicles [16].

While notable progress has been made, particularly at the municipal level, toward reclaiming public space for active mobility through widened sidewalks, traffic calming, and bike infrastructure, these changes remain limited and slow, especially in densely populated areas. The development of Planes de Movilidad Urbana (Urban Mobility Plans, PMUs) has helped municipalities re-center their policies around walking and cycling, to promote the gradual reallocation of road space. Nevertheless, at the metropolitan scale, the road network continues to prioritize private motorized vehicles, with many intermunicipal corridors lacking basic urban features such as sidewalks, lighting, or bike lanes [17]. This contrast reflects an urban structure that is still skewed in favor of car mobility, despite the growing predominance of active modes in daily travel.

In this evolving context, Barcelona has implemented a series of strategic initiatives since the 1960s to promote more sustainable mobility. Early pedestrianization projects in central areas and the establishment of the Urban Ecology Agency marked the first steps toward a more pedestrian-centered city [18]. Among these early strategies, a particularly impactful measure was the creation of the AREA system in 2005, which reorganized on-street parking through a combination of blue and green zones. This initiative aimed to rationalize the use of curbside space by applying differentiated fees and time limits: blue zones encouraged short-term stays to support commercial activity, while green zones prioritized residents in areas with a high parking demand. By disincentivizing long-term parking and reducing the attractiveness of using private vehicles in central districts, AREA contributed to rebalancing public space in favor of more sustainable modes of transport. The implementation of this system was seen as a politically viable alternative to the congestion pricing models adopted in other cities, such as London, and became a foundational element of Barcelona’s integrated mobility policy [17]. Subsequently, the introduction of the orthogonal bus network in 2013, which optimized service coverage with fewer lines, demonstrated the city’s strong commitment to efficient, accessible public transportation. The implementation of the innovative superblock model, which restricts motorized vehicle circulation in specific residential blocks, has facilitated the recovery of public space for pedestrians and cyclists [18]. This concept has been expanded through the “Superblock Barcelona” project, involving the creation of green axes and new plazas. The aim is to transform the city on a broader metropolitan scale, and to prioritize active mobility and public transport as fundamental pillars [19].

Compiled with data from Barcelona City Council [20], Figure 2 shows an encouraging trend towards reduced private vehicle use and the increased adoption of public transportation. This suggests that the implemented policies have had a positive impact. Although the COVID-19 pandemic in 2020 temporarily altered these patterns, with a notable increase in walking and a temporary decrease in public transport use, the overall direction clearly points toward a more sustainable and resilient mobility model.

To complement this general trend, disaggregated data from Barcelona City Council [22] and the PDUM [16] reveal persistent structural inequalities in mobility patterns, particularly along gender and age lines. As shown in Figure 3, women in the city of Barcelona use private vehicles significantly less than men (13% vs. 27.4%). They walk more frequently (45.5% vs. 38.4%) and rely more on public transportation (39.5% vs. 28.5%) [22]. At the metropolitan level, Figure 3 reflects a similar pattern: 52.2% of women’s trips are made on foot or by bicycle, compared with 48.2% among men. Public transport use is also higher among women (27.4% vs. 20.0%), while men show a clear preference for private motorized transport (31.8% vs. 20.3%) [16].

The modal share is closely tied to the purposes of travel. As illustrated in Figure 4, women are more likely to make trips related to caregiving (11.7% vs. 8.4%) and routine shopping (7.9% vs. 2.4%), which contributes to a more fragmented, short-distance, and multimodal mobility pattern. In contrast, men report a higher share of work- and leisure-related trips [16]. Age-based analysis reveals additional dynamics. Among young people (ages 16 to 26), active modes and public transport are predominant, especially among women (52.3% of trips on foot vs. 48.2% for men). From age 30 onward, private vehicle use increases significantly, particularly among men aged 30 to 64 (43.4%), compared with their female counterparts (30.3%). Among people over 65, active mobility once again becomes dominant, with higher walking rates among women (38.3% vs. 31.8% for men) [16]. These disparities reflect an unequal distribution of access to urban space and mobility resources. This highlights the need for public policies that incorporate equity, accessibility, and spatial justice criteria. Recognition of the diverse mobility needs across gender and life stages is essential to building a truly inclusive, resilient, and sustainable transportation system.

Since the early twenty-first century, Barcelona has developed a set of strategic policies aimed at transforming its mobility model into one that is more sustainable, healthy, and equitable. A central component of this transformation has been the decisive promotion of cycling. In 2014, the city had only 116 km of bike lanes, accounting for 8.5% of the total street network [21]. In response, the municipality launched the Bicycle Strategy 2016–2018, which led to the significant expansion of the cycling network, with annual increases of 20 to 40 km, compared with just 5 to 10 km per year during the 2010–2015 period. The plan set a target of 308 km of bike lanes by 2018. However, by 2023, the network had reached 273 km of bike lanes and 1150 km of designated cycling itineraries. This illustrates the substantial progress that has been made, although the figures still fall short of the initial objectives [23].

In parallel with the municipal effort, the AMB launched the Bicivia network, a metropolitan-scale cycling infrastructure system aimed at enhancing intermunicipal connectivity and consolidating cycling as a key mode of transport [24]. According to the 4th Monitoring Report of the Pla de Mobilitat Urbana Metropolitana (Metropolitan Urban Mobility Plan, PMMU) 2023, Bicivia has been extended to 362 km, which represents a 58% increase from the 229 km recorded in 2016. Furthermore, the broader pedalable network, which includes bike lanes, greenways, and traffic-calmed streets, extended to 5405 km across the metropolitan area in 2022 [25]. These developments have been accompanied by broader urban and environmental policies. In January 2020, the Low Emission Zone (LEZ) Rondas de Barcelona was implemented, covering 36 metropolitan municipalities and restricting the circulation of the most polluting vehicles [26]. This measure complements comprehensive planning instruments such as the PDUM [16] and the PMMU [27], led by the AMB and the Autoritat del Transport Metropolità (Metropolitan Transport Authority, ATM). The PDUM promotes a polycentric territorial model articulated through metropolitan corridors and green axes, with the explicit goal of significantly reducing private car use. The PMMU establishes objectives for a healthy, sustainable, efficient, and equitable mobility system aligned with the European Union’s decarbonization targets [28,29].

In summary, both the city of Barcelona and its surrounding metropolitan region have undergone a profound transformation in their mobility model, progressively prioritizing sustainable modes such as cycling and public transport. Nonetheless, structural challenges remain, particularly in managing intermunicipal mobility, where private vehicles still dominate and where future strategic interventions must be focused.

2.2. Methodology

The proposed methodology is based on mixed methods, which combine a comparative case analysis with a comprehensive quantitative analysis and utilize robust multivariate statistical techniques. The methodology includes time series forecasting models, specifically linear regression and autoregressive integrated moving average (ARIMA), a model widely used in urban mobility forecasting due to its capacity to handle non-stationary and seasonal data patterns [30,31,32]. Additionally, exploratory factor analysis (EFA) was employed to identify latent dimensions in mobility patterns. This method is applied extensively in studies that analyze the structural dynamics of urban transport systems [33,34,35]. The objective was to analyze the evolution of travel patterns by mode of transport in the AMB to project trends toward the year 2050. The methodology was organized into four interconnected stages, which integrate multiple sources of information and employ analysis tools developed in Python (PyCharm 2022.3.2). The research methodology is summarized in the methodological framework shown in Figure 5.

2.2.1. Data Collection and Preprocessing

To conduct a robust trend and factor analysis of urban mobility in the AMB, a multi-source data collection strategy was adopted, focused on temporal consistency and relevance. The following official and validated datasets were utilized:

• Enquesta de Mobilitat en Dia Feiner (Weekday Mobility Survey, EMEF), 2011–2024: annual data of modal displacements (in thousands of daily workday trips), across all major transport modes, were extracted from EMEF. While official data is available for all of 2023, the value for 2024 was constructed using a projection based on the 2024 forecasts included in the PDUM to ensure the continuity and extension of the time series for modeling purposes [36]. To strengthen the methodological rigor of the trend analysis, the years 2020, 2021, and 2022 were excluded from the modeling process, due to the significant, well-documented distortions in mobility patterns caused by the COVID-19 pandemic. This exclusion criterion was consistently applied across all modeling techniques based on time series data. However, data from the pandemic period were retained in the general database for descriptive, contextual, and interpretative purposes.
• Institut d’Estadística de Catalunya (Catalan Institute of Statistics, IDESCAT), ATM and TMB: The data were incorporated from sources [37,38,39], including fare prices, demand volumes by operator, number of lines, and network length for all public transport subsystems. These variables were essential to understand the operational, economic, and structural dimensions of the mobility system.
• Observatori de la Mobilitat de Catalunya (Catalonia Mobility Observatory, OMC) and Bicing System Reports: infrastructure and performance indicators (e.g., bike-sharing use, bicycle infrastructure, pedestrian priority zones, and bus punctuality) were also included to represent policy and accessibility dimensions, particularly for active modes. The data source used was extracted from [40].

The variables considered in this study, along with their descriptions, units, and sources, are presented in

Table 1. Variables considered in the analysis. Data from [36,37,38,39,40].

Code	Variable	Description	Unit	Source
AP	On foot	Annual estimate of trips made by active mobility on an average working day, extrapolated to the year in the AMB	Thousands of trips/year	EMEF
BC	Bicycle
VMP	Personal mobility vehicles
TMA	Total active mobility
AB	Bus	Annual estimate of trips made by public transport on an average working day, extrapolated to the year in the AMB	Thousands of trips/year	EMEF
MT	Metro
TR	Train
RTP	Other public transportation
TTP	Total public transportation
AM	Automobile	Annual estimate of trips made by private vehicle on an average working day, extrapolated to the year in the AMB	Thousands of trips/year	EMEF
MC	Motorcycle/moped
OVP	Van, truck and other private vehicles
TVP	Total private vehicle
TTJ	Youth Transportation Fare	Quarterly transportation fare for young people	EUR/quarter	IDESCAT/ATM
TM	Monthly rate	Monthly transportation fee	EUR/month	IDESCAT/ATM
TI	Individual rate	Single trip rate	EUR/trip	IDESCAT/ATM
DM	Metro Demand	Total volume of public transport trips, registered in the transport system, reported by operators	Millions of trips/year	IDESCAT/ATM
DFGC	FGC Demand
DRD	Demand Rodalies
DTR	Tram Demand
DAB	Demand Bus TB
DBAMB	Demand AMB Bus
DTTP	Total Public Transportation Demand
LM	Metro Lines	Total number of metro lines in operation in the Barcelona metropolitan transport system	Lines	IDESCAT/ATM
LFGC	FGC lines	Number of FGC operating lines within the AMB	Lines	IDESCAT/ATM
LAB	TB Bus Lines	Number of urban bus lines operated by TMB in the city	Lines	IDESCAT/ATM
LBAMB	AMB Bus Lines	Number of intercity bus lines and metropolitan bus lines managed by the AMB	Lines	IDESCAT/ATM
TLTP	Total public transportation lines	Number of lines of total public transportation	Lines	IDESCAT/ATM
LRM	Length of Metro network	Total km of network in operation	Km	IDESCAT/ATM
LFGC	Length of FGC network
LRR	Length of Rodalies network
LATB	Network length Bus TB
LTAMB	Length of AMB bus network
LTRTP	Total length of public transportation network
EM	Metro Stations	Number of operational subway stations	Stations	IDESCAT/TMB
AEM	Accessibility of Metro stations	Accessible subway stations	Stations	IDESCAT/TMB
PAEM	Percentage of accessibility of Metro stations	Percentage of accessible subway stations	%	IDESCAT/TMB
ADB	Diesel/Biodiesel Buses	Number of buses with diesel/biodiesel propulsion in operation within the AMB	Buses	IDESCAT/TMB/AMB
AH	Hybrid Buses	Number of buses with hybrid propulsion in operation within AMB	Buses	IDESCAT/TMB/AMB
AE	Electric Buses	Number of buses with electric propulsion in operation within AMB	Buses	IDESCAT/TMB/AMB
AGN	Natural gas buses	Number of buses with natural gas propulsion in operation within the AMB	Buses	IDESCAT/TMB/AMB
TA	Total Buses	Total number of buses operating in the AMB, considering all propulsions.	Buses	IDESCAT/TMB/AMB
DB	Demand for bicycle service	Total annual bicycle trips recorded	Trips/year	Bicing
EBM	Mechanical bicycle stations	Number of mechanical stations available in the shared system	Stations	Bicing
BM	Mechanical bicycles	Number of mechanical bicycles available in the shared system	Bicycles	Bicing
BE	Electric bicycles	Number of electric bicycles available in the shared system	Bicycles	Bicing
EBE	Electric bicycle stations	Number of power stations available in the shared system	Stations	Bicing
PRP	Pedestrian priority	Pedestrian priority streets and zones	Ha	IERMB
VC	Cycle paths	Km of bicycle lanes	Km	OMC
C30	Streets 30 Km	Streets with speed 30 Km	Km	OMC
CP	Pedestrian streets	Single-platform streets with pedestrian priority	Km	OMC
PAB	Punctuality Bus	AMB bus service punctuality rate	Score	OMC

.

2.2.2. Exploratory Factor Analysis (EFA)

To identify the latent structural dimensions underlying the variability in urban mobility patterns, an exploratory factor analysis (EFA) was conducted using annual data from 2011 to 2024. This multivariate technique was used to reduce system complexity by grouping correlated observed variables into a smaller set of unobserved components. These components constitute an empirical foundation that is intended to inform subsequent stages of research, particularly in the future development of scenario-based analyses.

• Assessment of model adequacy
  Prior to factor extraction, the suitability of the data for EFA was evaluated using the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity. The KMO index assessed the proportion of variance that might be common among variables, while Bartlett’s test verified that the correlation matrix significantly departed from an identity structure. These procedures ensured the statistical feasibility of extracting meaningful latent dimensions.
• Factor extraction and rotation
  The extraction of components was performed using the principal components method. To enhance interpretability and facilitate the identification of distinct structural dimensions, an orthogonal Varimax rotation was applied.
• Interpretation of structural dimensions
  The resulting factor structure was interpreted by examining component loadings, to identify key thematic axes relevant to the urban mobility system. These dimensions reflected different operational, infrastructural, and behavioral aspects of mobility and are expected to serve as analytical inputs for future research focused on scenario construction and long-term mobility transitions.

2.2.3. Trend Analysis

To estimate the long-term evolution of urban mobility patterns, a trend analysis was carried out by applying ordinary least squares (OLS) regression models. The analysis was conducted using the Python programming environment. Specialized libraries were incorporated such as pandas for data structuring, SciPy for statistical correlation testing, and scikit-learn for the development and evaluation of regression models. Independent models were constructed for each transport mode, private vehicles, active mobility, and public transport, based on the historical time series from 2011 to 2024. The years 2020, 2021, and 2022 were excluded from the modeling process due to the abnormal distortions in travel behavior caused by the COVID-19 pandemic.

In order to complement the linear modeling approach and capture potential temporal dependencies or nonlinear dynamics, autoregressive integrated moving average (ARIMA) models were applied. These models can be used to represent autocorrelated structures and offer a robust approach to forecasting time series behavior. ARIMA-based projections were used to complement the linear estimations and to contrast potential modal trajectories under different structural assumptions. In parallel, non-parametric techniques for trend detection were explored, specifically the Mann–Kendall test and the Sen’s slope estimator. These methods are recognized for their robustness in short time series and their independence from assumptions of normality or homoscedasticity. However, due to the limited number of observations and insufficient statistical significance, their results were not incorporated into the main body of analysis and were retained solely as an internal methodological reference.

The primary objective of the trend analysis was to evaluate the extent to which structural transformations in travel behavior may influence the long-term configuration of the urban mobility system. The analysis focused on identifying patterns of modal change over time and understanding how these are shaped by policy interventions and evolving user preferences, thereby offering insights into the future dynamics of urban transport in the metropolitan context.

2.2.4. Comparative Analysis of Urban Mobility Models

Finally, the mobility strategies implemented in Barcelona were compared with those of two internationally recognized cities: Amsterdam and Copenhagen. Key indicators were considered, such as modal share, investment in sustainable transport infrastructure, and observed outcomes of public policies in each city. The purpose of this analysis was to contextualize the evolution and dynamics of mobility in Barcelona in an international reference framework, to facilitate the identification of best practices and potential areas for improvement that could inform future planning.

3. Results

3.1. Explanatory Factor Analysis (EFA)

To identify latent structures that synthesize the complexity of urban mobility patterns in the AMB, an EFA was conducted. This multivariate technique can be used to reduce a large set of observed variables into a smaller number of latent factors, revealing underlying dimensions that explain the covariation between them. In the context of this study, EFA offers an empirical approach to understanding the main structural axes of the metropolitan mobility system, which is especially useful when a predefined theoretical structure is not available.

Before proceeding with factor extraction, the suitability of the dataset for factor analysis was assessed. The KMO index yielded a value of 0.583, which is considered acceptable for the application of EFA according to established methodological standards. Although this value is at the lower threshold of acceptability, the specialized literature recognizes the possibility of obtaining significant factorial solutions even with moderate KMO values, provided that factor loadings are high and communalities are adequate. As Costello and Osborne point out [41], in situations where the factorial structure shows conceptual clarity and empirical coherence, KMO values between 0.5 and 0.6 can be accepted, especially in small samples or exploratory designs like the present case. Furthermore, Bartlett’s test of sphericity produced a χ² = 205.19 statistic with a significance level of p < 0.001, which indicates that the correlations between variables are sufficiently significant to justify the use of this technique. This combination of results supports the applicability of EFA in this study by demonstrating the existence of latent relationships among the selected indicators.

As observed in Figure 6, the correlation matrix reveals clear groupings of variables with high intercorrelations, especially among those linked to public transport infrastructure and usage. For example, the length of the metro, suburban train, and bus networks, as well as demand and accessibility variables, show significant positive associations, which indicates a strongly integrated transport system. Similarly, the variables related to active mobility, such as the demand for bicycle services, cycling infrastructure, and pedestrian priority, display consistent correlations with each other, and generally weak or negative relationships with the variables associated with car use. These patterns reflect functional groupings within the urban mobility system and suggest the existence of shared latent constructs. The visual density of the heatmap reinforces the relevance of applying dimensionality reduction techniques, which statistically and conceptually validate the use of EFA. In this sense, the matrix acts as a diagnostic and conceptual tool prior to factor extraction.

Building on this empirical foundation, the application of EFA enabled the extraction of four latent structural dimensions underlying Barcelona’s metropolitan mobility system. Together, these factors explain 92.3% of the shared variance among the analyzed variables. It is crucial to highlight that these dimensions do not correspond to specific geographical areas but rather represent functional patterns reflecting recurrent combinations of transport modes, infrastructure, and mobility behaviors. Figure 7 illustrates the grouping of variables within each factor, while

Table 2. Factors identified from the exploratory factor analysis (EFA).

Factor	Highest Loading Variables	Qualitative Interpretation	Explained Variance %	Cumulative Variance%
1. Conventional transport infrastructure and pedestrian priority	Metro lines (−0.34), Bus AMB lines (−0.28), FGC Length (−0.35), Pedestrian Streets (−0.34)	Relatively low presence of traditional mass transport and pedestrian infrastructure. Reflects system configurations where these elements are not predominant.	54%	54.0%
2. Integrated and active urban mobility	Bus lines (0.47), Bike Service Demand (0.42), Bus Length (0.15)	Pattern where collective transport supply aligns functionally with active mode demand. Reflects proximity-based, intermodal mobility.	27.1%	81.1%
3. Low penetration of public micromobility	Bike Station for Electrics (−0.38), Bike Service Demand (−0.54)	Dimension where public micromobility services are underrepresented. May indicate dependence on other modes or limited system integration.	6.8%	87.9%
4. Operational effi-ciency of bus transport	Bus Punctuality (0.62))	Axis focused on the reliability of public transport, where punctuality stands out as a key attribute of operational effi-ciency.	4.4%	92.3%

synthesizes their conceptual interpretation.

• Conventional transport infrastructure and pedestrian priority (54.0% of explained variance).
  This factor groups structural variables such as the length of metro lines, the Ferrocarrils de la Generalitat de Catalunya (Catalan Government Railways, FGC) network, interurban buses (AMB), and pedestrian streets. All these variables exhibit prominent negative loadings. Notably, this negativity does not imply an adverse effect but rather indicates that these variables hold less relative weight in defining this specific dimension. In other words, this system configuration is characterized by a low prominence of traditional mass transport infrastructure and consolidated pedestrian spaces, which could suggest the predominance of other mobility models or a different stage of development. Its interpretation is crucial to identify areas within the system in which conventional forms of public transport and pedestrian infrastructure are not yet predominant or have been superseded by new functional schemes.
• Integrated and active urban mobility (27.1% of explained variance).
  This component is characterized by positive loadings on variables such as urban bus line coverage, the total length of the bus network, and public bicycle demand. This pattern reflects a synergistic functional relationship between collective transport and active bicycle use, emphasizing their functional complementarity rather than merely installed infrastructure. This dimension suggests mobility patterns where both systems operate in a coordinated manner, facilitating short-distance trips and configuring a flexible intermodal model. These results are consistent with previously observed correlations between active mobility, electric buses, and bicycles, and align with strategic urban policies such as superblocks, green axes, or the Bicivia network, which promote accessible and sustainable mobility environments.
• Low penetration of public micromobility (6.8% of explained variance).
  This component exhibits negative loadings on variables such as electric bicycle stations and the demand for these services. The loadings reflect the fact that, within this dimension, public micromobility has limited or no representation. This factor can be interpreted as a modal configuration where shared bicycles do not constitute a relevant component of the overall transport system, either due to the dominant presence of other modes or the still-limited integration of these solutions. These results also relate to observed correlations concerning car use, which suggests an opportunity to expand or strengthen micromobility infrastructure and coverage in certain areas.
• Operational efficiency of bus transport (4.4% of explained variance).
  This factor is almost exclusively defined by a high positive loading on the bus punctuality variable, which allows its interpretation as a dimension focused on operational performance and service quality. Unlike factors reflecting structural infrastructure or demand patterns, this dimension captures a fundamental qualitative attribute: the reliability of the public transport service. Its emergence as an independent factor underscores the critical importance of efficiency in transport management, particularly in the context of urban policies aimed at optimizing traffic flow, such as the implementation of bus lanes, traffic signal prioritization, or regulation through the LEZ.
  The factors extracted through EFA not only synthesize the variables with the highest loadings and their conceptual interpretation, but also facilitate the understanding of how these dimensions articulate with the projected trends of the mobility system. It is crucial to highlight that negative loadings, in this context, indicate inverse relationships and must be interpreted carefully within the specific framework of the AMB metropolitan system, without being assumed as negative effects per se. This qualitative reading establishes a robust bridge between statistical analysis and the complexity of urban reality, which is fundamental for the formulation of effective strategies and interventions in mobility. To gain further insight into the patterns revealed by both the factor analysis and the correlation matrix, a complementary analysis of direct correlations was conducted between key mobility system variables and the main transport modes. This exercise provides a more granular perspective on the functional relationships that structure the distinct modal patterns observed in the AMB.
• Active mobility (Figure 8): The highest correlations with the active mobility variable were recorded for bus line length, electric bicycle use, and electric bus availability. This finding suggests that robust public transport infrastructure, combined with sustainable mobility technology options, acts as a catalyst for non-motorized travel. Furthermore, pedestrian streets and personal mobility vehicles are highlighted, reinforcing the creation of urban environments oriented towards active accessibility. These relationships are closely aligned with the urban strategies promoted by Barcelona City Council, such as the expansion of the Bicivia network, the implementation of segregated bike lanes, smart parking programs like Bicibox, and the consolidation of green axes in densely urbanized areas, which prioritize pedestrians and cyclists.

• Public transport (Figure 9): the strongest correlations with the public transport variable were observed in bicycle use, electric bicycle stations, diesel/biodiesel buses, and the length of the Rodalies network. This pattern indicates consolidated intermodality, where sustainable modes like cycling effectively interact with collective transport modes. Additionally, the structural role of both railway networks and conventional bus fleets in providing a public service is reinforced. These synergies are consistent with the objectives of the metropolitan area’s LEZ, which seeks to discourage the use of polluting private vehicles in favor of public transport and shared mobility.

• Car use (Figure 10): significant positive associations were identified with motorcycle or moped use and, notably, with mechanical bicycles. This could indicate patterns of shared mobility or modal alternation in certain urban environments. Furthermore, correlations were observed with the young driver rate and monthly subscription, which might be linked to seasonal variations or specific sociodemographic profiles in private vehicle use. These patterns can be interpreted within the framework of measures such as the reorganization of urban space through superblocks, which aim to restrict motorized traffic in residential areas, and youth mobility policies that include differentiated fares and shared micromobility services.

Collectively, these relationships provide a solid empirical basis for understanding the articulation between different transport modes, available infrastructures, demographic profiles, and urban policies in the city. Moreover, they strengthen the validity of previously identified factors and offer key insights for the development of prospective sustainable mobility scenarios within Barcelona’s urban transformation context.

3.2. Trend Analysis and Projections of Mobility in the AMB

This section presents a detailed analysis of historical trends and future mobility projections in the AMB up to the year 2050, differentiated by transport mode. To estimate this evolution, two complementary methodologies were employed: simple linear regression and ARIMA models. Linear regression allows for the identification of clear and understandable long-term trajectories, but it presents significant limitations when faced with disruptive events, saturation effects, or sudden structural changes, such as the COVID-19 pandemic. In contrast, ARIMA models offer a greater capacity for adaptation by incorporating the time series’ temporal dynamics (such as seasonalities, trends, or recent shocks), which allows for the generation of projections that are more sensitive to recent system variations. However, to ensure methodological consistency between both approaches and to prevent the atypical period of the pandemic from distorting the results, whether by amplifying conjunctural effects or introducing statistical noise, it was decided to exclude the 2020–2022 period from the training of both models. In the case of linear regression, this exclusion was necessary to preserve the linearity of the long-term trend. For the ARIMA model, although it has a greater capacity to absorb variability, this segment was omitted based on the criteria of comparability and robustness, prioritizing a more stable database that reflects sustained structural behaviors.

3.2.1. Statistical Evaluation of Projection Models

The statistical quality of the projection models that were employed was evaluated using key metrics, differentiated by the nature of each methodology. In the case of linear regression, the model applied to public transport exhibits a robust and statistically significant fit (${\mathrm{R}}^2$= 0.53; p-value = 0.01). This result validates the reliability of the identified trend for this mode. For active mobility, the fit is more modest (${\mathrm{R}}^2$ = 0.29; p-value = 0.08). However, the model still reveals signs of growth that are consistent with the current urban context, which suggests a latent trend. Conversely, the linear regression model applied to private vehicles was not statistically significant (${\mathrm{R}}^2$ = 0.03; p-value = 0.63), which indicates that this simple method fails to adequately capture the complexity of the evolution. Furthermore, the ARIMA models demonstrate a superior performance regarding their dynamic fit and time series modeling. The absence of significant residual autocorrelation, confirmed by the Ljung–Box test (p > 0.05) for all models, validates the adequacy of the autoregressive and moving average structure models. Likewise, most models maintain acceptable residual distributions according to the Jarque–Bera test (p > 0.05), with the exception of the model for public transport. This finding suggests that ARIMA models are appropriate for projection purposes, although the deviation from normality in public transport residuals should be considered with caution during interpretation. Notably, the lack of statistical significance in individual coefficients, while common in ARIMA models, does not invalidate their predictive utility, especially when global fit tests and the absence of residual autocorrelation are satisfactory. Collectively, both methodological approaches prove complementary for trend analysis: linear regression provides a clear view of the underlying long-term general trend, while ARIMA models respond more sensitively to recent oscillations and the observed effects of implemented policies within the historical series.

Table 3. Comparative statistical summary of forecast models. Author’s elaboration.

Mode of Transport	R2 (Linear Regression)	p-Value (Linear Regression)	Ljung–Box p-Value (ARIMA)	Jarque–Bera p-Value (ARIMA)
Active Mobility	0.288	0.089	0.39	0.46
Public Transport	0.53	0.011	0.34	0.02
Private Vehicles	0.027	0.632	1.0	0.64

summarizes the main statistical quality indicators for each analyzed transport sub-mode, which encompasses both methodologies.

3.2.2. Global Comparison Between Methodologies and General Projections

To facilitate an integrated understanding of the projections, Figure 11 presents a direct comparison of the results obtained through linear regression and ARIMA models for the three main transport modes: active mobility, public transport, and private vehicles. This visualization synthesizes the primary divergences between both approaches and underscores how the ARIMA model, by incorporating more complex temporal dynamics and being more sensitive to recent changes, anticipates a more accelerated transformation of the mobility system.

In particular, active mobility shows more pronounced growth in the ARIMA model’s projections, which suggests greater potential for expansion given the effective implementation of targeted urban policies. These policies include green axes, the Bicivia network, and the consolidation of pedestrian-friendly environments. For public transport, the differences between both models are less pronounced but equally relevant, with slightly more optimistic growth projected by ARIMA. This could be attributed to a more dynamic post-pandemic recovery or the impact of operational improvements derived from investments in punctuality and service coverage. The most notable divergence between both approaches is observed in the projected decline for private vehicle use. While linear regression anticipates a moderate decrease, the ARIMA model predicts a more sustained and pronounced drop. This behavior is fully consistent with the progressive implementation of restrictive measures in the AMB, such as the LEZ or the redesign of superblocks and green axes that reduce the amount of through traffic. These strategies, whose effects are integrated into the ARIMA model through the recent dynamics of the historical series, allow for the more precise capture of the structural effects that are shaping an urban environment that is less dependent on the automobile.

3.2.3. Active Mobility (Walking, Cycling, and PMVs)

Individual projections for active modes, such as walking, cycling, and personal mobility vehicles (PMVs), provide a detailed insight into their expected evolution until 2050 (Figure 12). Collectively, both models predict sustained growth for active mobility, albeit with differences in the magnitude and dynamics of each sub-mode.

• Walking consolidates its position as the predominant active mobility mode. Both the linear regression and ARIMA models project a significant increase. However, the ARIMA model anticipates more pronounced acceleration, driven by recent trends and possibly influenced by urban policies such as superblocks and green axes, which have reallocated public space in favor of pedestrians. This result confirms that walking will remain central to the proximity city model promoted by the AMB.
• In the case of cycling, both methodologies project a notable positive evolution. The linear model indicates a doubling of trips by 2050, while ARIMA suggests an even steeper slope. This points to the greater sensitivity of cycling to short- and medium-term policies, such as the expansion of the Bicivia network, improvements in cycling connectivity, and the growing culture of shared use. The anticipated trend validates current investments and highlights the potential of cycling as a key component of the modal shift.
• Finally, personal mobility vehicles (PMVs) exhibit more moderate growth. The linear regression shows a constant progression, while ARIMA captures some variability that could be related to regulatory uncertainties, conflicts with other modes, or still nascent infrastructure. Nevertheless, both models agree that PMVs will increase their participation in the system, though the total volumes are still lower than walking or cycling.

Overall, these patterns reinforce the interpretation of active mobility as an expanding structural dimension within the AMB, aligned with the factors extracted in the EFA and with metropolitan policies focused on proximity, accessibility, and sustainability.

3.2.4. Public Transport (Metro, Bus, Train, and Other Modes)

The disaggregated analysis by public transport mode (Figure 13) shows a positive evolution across all sub-components of the system until 2050. Both metro and bus services exhibit a clear upward trend, with evident recovery post 2023. The ARIMA model projects more dynamic growth than the linear model, especially for the metro, which could be associated with the consolidation of strategic infrastructure investments and operational improvements.

• In the case of the metro, there is a constant increase in the projected number of trips, with particularly accelerated progression under the ARIMA model. This trend can be linked to the consolidation of strategic investments in line extensions, automation, and improved service frequency. Furthermore, it reflects a growing modal preference, driven by public transport prioritization policies, the optimization of punctuality, and intermodal integration strategies. The divergence observed between both models suggests that the metro, as the backbone of the transport network, could respond with greater dynamism and strength to active policies reinforcing the structural network.
• The bus network maintains more moderate, but sustained, growth according to both models. This behavior could reflect the positive effects of measures such as the implementation of the LEZ, superblocks, green axes, and the redesign of public space through parking regulation, which collectively favors the accessibility and use of collective transport over private cars.
• In the case of railway transport (FGC and Rodalies), growth is more gradual. This can be attributed to its high dependence on long-term structural investment plans, which typically have longer implementation cycles.
• The rest of the public transport group (including trams, funiculars, or special services) also shows slow but continuous expansion. Given their smaller relative weight in the overall system, their projected evolution is stable, although their role could be enhanced if they are more firmly integrated into intermodality strategies.

Collectively, ARIMA models offer a more sensitive reading of recent changes and project slightly more optimistic increases than linear regression. This highlights the continuous need for active policies to maintain and boost this trend: technological improvements, fare integration, fleet electrification, and infrastructure expansion are essential elements to sustain projected growth. These projections suggest that if current strategies are maintained and operational efficiency and territorial coverage axes are reinforced, public transport could consolidate itself as the structuring axis of sustainable metropolitan mobility by 2050.

3.2.5. Private Vehicle Use

The projections for private vehicle use reflect a clear downward trend towards 2050, in both the linear regression and ARIMA models, albeit with differences in the pace and intensity of this decline (Figure 14).

• In the case of automobiles, both models agree on projecting a sustained decrease. However, the ARIMA model anticipates a faster and more pronounced reduction, which suggests a greater sensitivity to recent transformations of the urban system. This behavior is consistent with the cumulative effects of restrictive policies such as the LEZ, the expansion of the superblock model, the creation of green axes and pacified zones, the limitation of parking through differentiated fees, and the promotion of more sustainable alternatives.
• For motorcycles and mopeds, the linear regression suggests a slightly upward evolution, while the ARIMA model projects stabilization or even a slight decrease in the long term. This difference may be associated with the recent growth of these vehicles as car substitutes in urban environments, although their future could be affected by their inclusion in future environmental regulations or by the rise in personal mobility vehicles (PMVs).
• Regarding vans, light trucks, and other private vehicles, both models project a gradual increase, though more moderate in the case of the ARIMA model. This pattern suggests that these modes, associated with urban logistics and distribution, might be less sensitive to general individual mobility restrictions, as they involve economic and work-related uses that are more difficult to replace.

An examination of these results reveals a structural transformation in private motorized mobility within the AMB. The downward trend in automobile use, reinforced by the ARIMA model, validates the direction of current public policies, while also underlining the importance of continuing to consolidate functional, accessible, and competitive alternatives for all social and territorial sectors.

3.3. Comparative Analysis of Urban Mobility Strategies

The comparative analysis of the mobility strategies implemented in Barcelona, Amsterdam, and Copenhagen reveals distinctive approaches, but with complementary underlying principles toward building sustainable urban mobility systems. Amsterdam, globally recognized for its deep-rooted cycling culture, has systematically integrated bicycle infrastructure into the urban fabric since the 1970s. This historical prioritization, combined with speed reduction policies and the development of efficient bike-sharing systems, has resulted in approximately 36% of all trips in the city being made by bicycle [9]. This percentage underscores the success of its long-term approach and the profound internalization of cycling as a primary mode of transport.

Similarly, Copenhagen has developed an exceptionally advanced cycling infrastructure. Its strategy began with the pedestrianization of key streets in 1962 and has progressed into an extensive network of cycle paths. What distinguishes Copenhagen’s approach is its ambitious objectives and exemplary integration of cycling with public transport, materialized in the “Cycle Superhighways” and the continuous improvement of intermodal connections. The outcomes of these policies are remarkable, with a 137% increase in bicycle use in the city center and a consequent 32% reduction in motorized traffic [10]. In contrast, Barcelona has adopted a more recent and accelerated trajectory toward sustainable mobility. The introduction of the superblock model and the creation of green corridors represent a paradigm shift in the reconfiguration of urban space. These initiatives specifically aim to reduce the road space dedicated to private vehicles, actively prioritizing active transportation modes and the use of public transport. While initial results show a promising trend in reducing car use, Barcelona faces unique challenges stemming from its high population density and persistent reliance on private vehicles for intermunicipal trips. Despite their successes, each city faces the following particular challenges in its pursuit of more sustainable mobility:

• Amsterdam must manage the continuous improvement and maintenance of its vast cycling infrastructure in a densely populated environment, while integrating new technologies without compromising its established cycling culture.
• Copenhagen faces the need to expand and optimize cycling infrastructure in its peripheral areas, further strengthening integration with public transport. Additionally, it must manage traffic growth and proactively adapt to new mobility technologies.
• Barcelona, in turn, faces the challenge of consolidating and expanding its superblock and green corridor projects. Achieving the effective integration of these new infrastructures with the existing public transport network is essential, as is managing the necessary cultural shift toward truly sustainable mobility, particularly in the context of its complex metropolitan area.

The most significant differences among the three analyzed cities can be seen in their metropolitan governance structures. These structures directly influence the capacity for the planning, funding, and effective implementation of mobility policies. Amsterdam is part of Metropoolregio Amsterdam (MRA), which includes a specialized regional authority, the Vervoerregio Amsterdam. This entity has clear executive competencies over the planning, funding, and coordination of public transport across 14 municipalities. Although each municipality retains a considerable degree of autonomy, a strong tradition of pragmatic cooperation exists. This has enabled the implementation of unified fare policies, the creation of cycling networks that seamlessly cross administrative boundaries, and robust integration between urban planning and mobility. Copenhagen is integrated into the Greater Copenhagen Area, coordinated within the Capital Region and comprising over 30 municipalities. Despite the absence of a single metropolitan authority with full executive powers over all aspects of mobility, there is an established institutional culture of collaborative regional planning, especially concerning transport and the environment. The Movia authority, which is responsible for managing buses and local transport services, facilitates public transport integration beyond administrative borders. Additionally, long-term planning instruments like the Finger Plan have structured urban and mobility development for decades, which has facilitated the implementation of ambitious strategies such as Cycle Superhighways.

In contrast, Barcelona operates within the framework of the AMB, a public entity that brings together 36 municipalities and has relevant strategic instruments such as the PDUM and the PMMU. However, the AMB faces significant institutional and operational limitations that restrict its capacity for effective action as follows:

• Administrative fragmentation: metropolitan governance largely depends on voluntary cooperation among municipalities and complex political alignment across the levels of government (local, autonomous, and state). This structure can slow down decision-making and coordinated implementation.
• Limited competencies: the AMB does not manage all infrastructure or transport modes. For example, the Rodalies network falls under state jurisdiction, numerous urban infrastructures depend on the Generalitat de Catalunya, and municipalities retain significant control over their local street networks.
• Territorial inequalities: substantial differences persist between central and peripheral municipalities regarding access to quality public transport, cycling infrastructure, and the application of restrictions on private car use, leading to heterogeneous mobility across the metropolitan area.
• Uneven policy implementation: key measures such as the ZBE, the superblock model, or green axes have not been adopted with the same degree of commitment or intensity in all metropolitan municipalities, which reduces their regional impact and effectiveness.

These limitations contrast sharply with the governance models of Amsterdam and Copenhagen, which benefit from functional authorities with clear competencies, established mechanisms for intermunicipal cooperation, and an ingrained institutional culture of long-term strategic planning. For Barcelona, overcoming these barriers will require substantially strengthening the AMB’s institutional capacity, improving multi-level coordination among administrations, and consolidating a shared vision and firm commitment among municipalities to achieve truly metropolitan and sustainable mobility.

The comparative analysis of metropolitan areas reveals clear differences in mobility management. Amsterdam and Copenhagen demonstrate stronger integration between their urban centers and surrounding regions, although defining administrative boundaries can be complex. Barcelona, with its metropolitan region, operates under a more intricate administrative framework, which nonetheless offers the potential for more efficient metropolitan mobility management. However, the urban sprawl of Barcelona extends beyond the formal boundaries of the metropolitan area, adding layers of complexity to coordination and planning. Figure 15 illustrates the spatial extent of the metropolitan areas of Amsterdam, Copenhagen, and Barcelona, as well as their continuous urban fabric as defined by Corine Land Cover and Urban Atlas. This comparison highlights the varying degrees of urban continuity and sprawl that influence mobility strategies and governance challenges across the three contexts.

These distinct metropolitan approaches underscore the need for a long-term vision in urban mobility planning, the seamless integration of cycling with public transport, and the proactive redesign of urban space as essential elements for creating sustainable mobility systems. Moreover, effective metropolitan cooperation and coordination are vital for addressing mobility challenges across the region. The experiences of Amsterdam, Copenhagen, and Barcelona provide valuable models and insights that can inform decision-making and strategic planning in other cities that aim for more sustainable urban mobility. These examples also emphasize the importance of well-planned infrastructure and inclusive policies to support active and sustainable transport [44,45].

4. Discussion

This study proposes a mixed-methods approach to analyze mobility evolution in the AMB between 2011 and 2024, and to project its evolution up to 2050. It integrates an exploratory factor analysis (EFA), regression and ARIMA models, and an international comparative approach. This methodology identified latent structures within the metropolitan mobility system, long-term trends, and transferable insights from other cities.

The EFA was constructed from a longitudinal database with annual indicators of the AMB’s mobility system between 2011 and 2024, including variables on the mode of travel, public transport demand, infrastructure, fares, accessibility, and operational conditions. This database enabled the extraction of underlying patterns that structure metropolitan mobility and are relevant for designing sustainable policies. The analysis revealed four fundamental structural dimensions of the AMB’s mobility system. The first, “conventional transport infrastructure and pedestrian priority”, encompasses traditional elements such as the metro, interurban buses, and pedestrian streets, and reflects the limited consolidation of conventional transport as a structural backbone within the current system. The second dimension, “integrated and active urban mobility”, highlights the complementarity between public transport (especially the bus network) and the use of public bicycles, reflecting proximity-based travel patterns and functional intermodality. The third dimension, “low penetration of public micromobility”, reveals an underdeveloped structural space for shared electric bicycle services, which suggests significant opportunities for expansion. Finally, the fourth dimension, “operational efficiency of bus transport”, emphasizes the relevance of punctuality as a crucial indicator of system reliability. These findings allow the AMB’s mobility system to be interpreted not merely as a sum of modes, but also as an interdependent configuration of technical, functional, and behavioral dimensions. Additionally, they provide a robust empirical framework for designing differentiated and evidence-based sustainable mobility policies.

Regarding temporal evolution, the trend analysis between 2011 and 2024 confirmed a sustained decrease in private vehicle use, while the regression and ARIMA models allowed for the projection of different scenarios up to 2050. The ARIMA models, which are more sensitive to the recent effects of public policies, project a more accelerated transformation than the linear models, particularly for walking and cycling modes. Nevertheless, both approaches converge on the conclusion that, without the significant intensification of current strategies, the pace of change will be insufficient to achieve the climate neutrality objectives set for 2050.

At the international level, a comparison with Amsterdam and Copenhagen revealed significant contrasts. Both of these cities have functional metropolitan governance structures, supported by specialized transport authorities, integrated planning, and long-term institutional cultures. Conversely, the AMB faces considerable challenges: administrative fragmentation, limited competencies over certain modes, marked territorial inequalities, and uneven policy implementation. These weaknesses hinder the consolidation of a coherent and equitable metropolitan mobility strategy. Collectively, the results suggest that, while Barcelona has progressed in its transition towards more sustainable mobility, the consolidation of this change not only demands investment and infrastructure, but also coordinated, efficient, and adaptive governance, capable of responding to the social, technological, and environmental transformations of the twenty-first century.

Furthermore, one of the main structural challenges is still intermunicipal mobility, where private vehicles maintain a dominant modal share. This limitation is related to the functional weakness of conventional public transport, as evidenced by the AFE results: the first dimension, conventional transport infrastructure and pedestrian priority, shows the partial consolidation of structuring modes like the metro and interurban buses. This lack of robustness limits the potential for a modal shift in journeys that require broader territorial connectivity. To address this barrier, intermunicipal articulation must be strengthened through improvements in the interurban network, fare integration, and the development of accessible and efficient intermodality hubs. These strategies are key to ensuring a just and effective transition towards car-free mobility across the entire AMB.

The structural factors and projected trajectories that emerge from this study constitute a solid technical basis for developing prospective mobility scenarios towards 2050, thereby facilitating the simulation of alternative policies and the prediction of their territorial effects. Moreover, the proposed methodology, based on the factor analysis of longitudinal indicators and the modeling of trends by transport mode, is replicable in other metropolitan contexts to diagnose structural mobility patterns and guide sustainable, evidence-based policies.

5. Conclusions

This research contributes to the urban sustainability debate by identifying the key structural dimensions and trends that allow for a better understanding of the evolution and challenges of the AMB’s mobility system. The policies implemented over the last decade have fostered a progressive reduction in car use, an improvement in air quality, and a boost to active modes. However, the pace of transformation is still insufficient to meet medium and long-term decarbonization targets.

The identification of four functional dimensions through factor analysis, based on a longitudinal database from the 2011–2024 period, provides a robust methodological foundation to guide future mobility policies. These dimensions enable the metropolitan territory to be segmented based on its supply and demand structure, guiding differentiated interventions. Projections up to 2050, generated using regression and ARIMA models, offer a framework for anticipating the cumulative effect of current policies if maintained or intensified.

From a practical standpoint, this study’s findings offer the following concrete implications for various stakeholders:

• Local governments and AMB: support for decision-making through evidence on which strategies are working and where adjustments are needed.
• Transport operators: need to invest in service quality, intermodality, and clean technologies.
• Citizens: access to healthier, more affordable, and equitable mobility, although this also entails cultural changes that require institutional support.

Nevertheless, this study presents some limitations. The number of observations available for the statistical models is limited, which may affect the robustness of certain estimations. Furthermore, while EFA allows for the synthesis of structural patterns, it does not capture spatial dynamics, which should be addressed with geospatial tools in future work. Additionally, the projected scenarios do not yet consider the impact of emerging innovations such as autonomous vehicles, mobility as a service (MaaS), or new urban delivery platforms.

In this regard, future lines of research could focus on the following:

• The spatial modeling of the identified factors;
• The simulation of prospective scenarios with governance and behavioral variables;
• The study of the impact of new technologies on modal distribution and equity;
• The evaluation of the distributive effects of current policies on vulnerable groups.

Moreover, the combined methodological strategy, which combines factor analysis and projection models, is applicable to other urban and metropolitan contexts facing similar challenges in the transition towards sustainable mobility. The flexibility of the model derives from the utilization of widely available longitudinal indicators pertaining to transport supply and demand, a feature which enables comparative diagnostics even within diverse institutional or spatial configurations. Consequently, this approach supports the development of context-sensitive, data-driven strategies for sustainable mobility planning.

In summary, this study reaffirms the importance of data-driven strategic planning, oriented towards the equity, efficiency, and resilience of the metropolitan mobility system. The transformation towards a post-car urban model is not only possible but necessary, and its success will depend both on the quality of policies and the institutional capacity to coordinate, finance, and sustain them over time.