The Road to 2030: Combining Life Cycle Assessment and Multi-Criteria Decision Analysis to Evaluate Commuting Alternatives in a University Context

Abstract

Institutions are increasingly being challenged to reduce the environmental impacts of daily commuting, while balancing complex and often conflicting sustainability goals. This study addressed the limitations of carbon-centric assessments by proposing a framework that integrated life cycle assessment (LCA) with multi-criteria decision analysis (MCDA) to evaluate seven prospective commuting alternatives for 2030, using a Portuguese university as a case study. Utilizing the PROMETHEE method across 16 environmental criteria, the analysis revealed that active mobility offered the most balanced and sustainable outcomes, consistently performing the best across all impact categories. In contrast, the electrification of private vehicles, although it reduced greenhouse gas emissions, was identified as the least favorable option, due to significant trade-offs in areas such as resource depletion and water use, as well as other environmental burdens. Public transport scenarios, particularly those involving electric bus systems, showed intermediate performance. In this context, the proposed LCA–MCDA framework provides policymakers and institutions with a comprehensive decision-support tool to navigate environmental trade-offs, promote low-impact mobility strategies, and meet evolving sustainability reporting requirements.

1. Introduction

The transportation sector is central in shaping global environmental outcomes, contributing approximately one-quarter of energy-related greenhouse gas (GHG) emissions [1]. As urban populations grow and the demand for mobility intensifies, transportation systems face increasing pressure to reduce their environmental footprint. In response, national and international agendas, such as the European Green Deal and the United Nations’ Sustainable Development Goals, promote sustainable mobility as a critical pillar of climate action [2]. Universities and other public institutions are also being called upon to lead the way in transitioning to low-impact mobility models [3].

However, designing effective and sustainable transportation systems is inherently complex. It involves navigating interrelated and sometimes conflicting priorities: reducing emissions, ensuring accessibility, maintaining operational viability, and aligning with evolving user needs [4]. Technological transitions, such as the electrification of transportation or the promotion of active mobility, can introduce environmental trade-offs, including shifting impacts from tailpipe emissions, to energy production and infrastructure demands [5]. These challenges are particularly pronounced in institutional settings, where diverse populations, constrained budgets, and the need for transparent sustainability reporting shape commuting patterns. Furthermore, current transportation planning and sustainability assessments primarily focus on GHG emissions. While carbon metrics are essential, they alone do not capture the full spectrum of environmental consequences associated with different mobility strategies, such as resource depletion, toxicity, land use change, and particulate emissions. Thus, there is a critical need to move towards more holistic evaluation frameworks that account for a broader range of environmental impacts.

This study bridges this gap by proposing a novel framework that integrates Life Cycle Assessment (LCA) and Multi-Criteria Decision Analysis (MCDA) to evaluate seven commuting alternatives across 16 environmental impact categories. LCA is a methodological approach that quantifies the environmental impacts of a system throughout its entire life cycle, from raw material extraction to disposal, providing a comprehensive understanding of resource use and emissions. MCDA, on the other hand, supports decision-making by comparing multiple alternatives across diverse and often conflicting criteria, incorporating both objective indicators and stakeholder preferences. The combination of LCA and MCDA enhances the robustness of the analysis, allowing for a multidimensional and prioritized assessment of sustainability trade-offs. Unlike previous works that often focused on specific technologies or carbon emissions alone, this framework adopts a systemic perspective tailored to institutional settings. Applied to a university community in Portugal, the approach explores prospective scenarios for 2030, encompassing baseline trends, enhanced public transport, promotion of active mobility, and transport electrification. Few studies have deployed such a comprehensive and forward-looking methodology in the context of university commuting, positioning this work as a valuable contribution to environmental assessment and strategic mobility planning.Within this context, the main contributions of this study are threefold:

• Development of an integrated LCA–MCDA framework for evaluating seven commuting alternatives across 16 environmental impact criteria.
• Application of the PROMETHEE (Preference Ranking Organization METHod for Enrichment of Evaluations) method to systematically rank alternatives and identify trade-offs between impact categories.
• Provision of actionable insights for institutional mobility planners relevant to sustainability reporting and strategic decision-making.

By combining quantitative environmental assessment with decision-support techniques, this study provides a replicable approach for institutions, pursuing truly sustainable commuting systems.

The remainder of this paper is organized as follows: Section 2 reviews the relevant literature on LCA and MCDA, highlighting the synergies between these methods and identifying potential gaps. Section 3 outlines the methodological framework, which includes a case study, alternative designs, and evaluation methods. Section 4 presents the findings, focusing on environmental performance and the implications for decision-making. Finally, Section 5 summarizes the paper and suggests opportunities for future research.

2. Literature Review

This literature review is organized into three main sections: (i) The first section presents studies related to LCA and its applications in the transportation sector; (ii) the second section evaluates studies that utilize MCDA approaches in transportation planning; and (iii) the final section discusses the combination of both methods, highlights the main gaps identified in the research, and outlines our contributions within this article.

2.1. LCA in Transportation Sustainability

The application of LCA in transportation sustainability has evolved significantly over the past decade, with research focusing on two primary domains: vehicle technology and transportation infrastructure. In transportation LCA methodologies, the system boundary definition is typically categorized as Well-to-Tank (WTT), Tank-to-Wheel (TTW), or Well-to-Wheel (WTW) approaches [6]. WTT encompasses upstream processes, including resource extraction and fuel production. TTW focuses on vehicle operation, while WTW provides an integrated evaluation. This distinction is critical when comparing conventional vehicles with alternatives like electric vehicles, which shift environmental burdens from the operation phase to the production phase [7].

In vehicle technology assessment, LCA has proven invaluable for evaluating the comparative environmental performance of conventional, electric, hybrid, and alternative fuel vehicles. Studies by Hawkins et al. [7] and Messagie et al. [8] demonstrated that while electric vehicles typically generate lower emissions in the use phase, their production phase impacts, particularly from battery manufacturing, can partially offset these advantages when evaluated from a life cycle perspective. This finding has prompted more nuanced discussions regarding vehicle electrification strategies, highlighting the importance of regional electricity mix and battery production improvements in determining net environmental benefits [9]. For transportation infrastructure, LCA research has illuminated the significant embedded impacts in the construction, maintenance, and end-of-life phases. Chester and Horvath [10] established that infrastructure-related emissions can constitute an essential category of transportation life cycle impacts, depending on system utilization patterns. Recent studies have refined these assessments by incorporating more comprehensive system boundaries and temporal dynamics, as seen in Saxe et al. [11] for public transit infrastructure. These analyses have highlighted how infrastructure longevity, material selection, and utilization intensity critically influence per-passenger environmental impacts.

While the environmental benefits of sustainable mobility transitions have been well-documented through LCA studies, significant knowledge gaps remain in understanding their implications within specific institutional contexts, such as universities and other public organizations. Most transportation LCA research has focused on municipal or regional scales, with relatively limited attention given to institution-specific issues, such as higher education institutions, mobility patterns, and improvement strategies [12]. This gap is particularly consequential, as universities function as ‘small cities’ with distinct commuting patterns, unique governance structures, and substantial potential for targeted mobility interventions.

2.2. MCDA in Transportation Planning

MCDA has emerged as a vital methodological approach for evaluating complex trade-offs in transportation planning, offering structured frameworks to incorporate environmental, economic, and social criteria. Across domains such as policymaking, logistics, public transportation, and infrastructure planning, methods like PROMETHEE, AHP (Analytic Hierarchy Process), ELECTRE (ELimination and Choice Translating Reality), and TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) have been extensively used to guide sustainable decision-making. In urban policy and governance, MCDA supports the evaluation of transportation alternatives aligned with long-term sustainability goals, balancing objectives such as economic viability, infrastructure reuse, and social equity [13,14]. The participatory nature of MCDA is particularly valuable for incorporating diverse stakeholder perspectives, including those of public authorities, residents, and private sector actors, when assessing multi-level governance challenges or public–private partnerships [15,16]. In logistics, MCDA has guided the selection of freight strategies and last-mile delivery solutions, assessing impacts like congestion, emissions, and delivery performance in urban settings [17].

Recent applications have further demonstrated the versatility of MCDA in addressing emerging challenges in sustainable mobility. Caetano et al. [18] proposed a decision-support tool for bus fleet management that combines AHP with an enumeration procedure to optimize service frequency and vehicle technology based on sustainability criteria. Aquile Junyent et al. [19] applied an MCDA-GIS integration to identify suitable locations for shared mobility hubs in Barcelona, incorporating spatial, environmental, and demographic indicators into a publicly accessible planning tool. Kundu et al. [20] introduced a fuzzy-based group decision-making framework to compare urban transport modes using expert-elicited criteria, enhancing methodological robustness in ambiguous decision environments. Similarly, Dong et al. developed a probabilistic linguistic MCDA method to assess urban green transport in Guangzhou, highlighting the role of new-energy vehicles and infrastructure configuration in shaping sustainable mobility trajectories. In public transportation, MCDA continues to support evaluations of rapid transit investments, mode integration, and user satisfaction, often incorporating GHG emissions among the core performance metrics [21,22,23]. Recently, Manzolli et al. [24] applied a stochastic acceptability framework to assess electric bus alternatives under uncertainty, providing insights into operational feasibility and long-term sustainability. Research on active mobility has used MCDA to assess walkability, cycling infrastructure, and intermodal connectivity, highlighting its utility in promoting equitable and sustainable urban mobility [25,26]. In university contexts, Oubahman et al. [27] adopted a hybrid AHP-PROMETHEE model to evaluate student transport preferences in Budapest, demonstrating that small expert-based panels can effectively inform institutional planning. Collectively, these developments underscore MCDA’s expanding role as a decision-support framework capable of bridging disciplinary boundaries, integrating stakeholder views, and supporting systemic approaches to sustainable transportation.

Despite its versatility, the application of MCDA in transportation planning still presents notable gaps and limitations. A key shortcoming lies in the predominant focus on current conditions or short-term evaluations, with relatively few studies adopting a forward-looking, scenario-based perspective to capture the dynamics of long-term mobility transitions. Moreover, although sustainability is a common theme, carbon emissions are often the only environmental metric considered, neglecting broader life cycle impacts such as resource depletion, human toxicity, or land use change. In this context, the integration of LCA into MCDA frameworks remains relatively underexplored, particularly regarding commuting behavior and institutional mobility planning, where consistent data, scenario comparability, and transparency of trade-offs are essential. Finally, existing studies rarely focus on institutional contexts, such as universities or municipalities, which face increasing pressure to align mobility planning with sustainability reporting standards and climate action plans.

2.3. Gaps and Contributions

As the literature reveals, integrating LCA and MCDA in transportation and commuting planning offers substantial potential for holistic sustainability assessment. Nevertheless, it also introduces key methodological, data-related, and institutional challenges. LCA quantifies environmental impacts across a product or system’s life cycle, while MCDA enables structured comparisons across diverse criteria, including environmental, social, and economic dimensions. Combining these tools requires bridging their differing assumptions: LCA offers absolute impact data, whereas MCDA typically relies on relative rankings and stakeholder-driven weighting, posing challenges in harmonizing objective metrics with subjective values [28,29]. Data availability and resolution present further barriers, as detailed geographic and temporal inputs are often insufficient, especially for institutional commuting systems, which undermines precision and comparability [30,31]. Furthermore, fragmented governance and regulatory inertia can hinder integrated assessments across inter-modal systems [32].

This study tackles these challenges by implementing a robust and novel LCA–MCDA framework to assess prospective commuting scenarios for a university community in Portugal. By combining 16 environmental impact categories with the PROMETHEE outranking method, the approach provides a multidimensional understanding of sustainability trade-offs that conventional carbon-centric assessments often overlook. Three key contributions distinguish this work: (i) it advances the methodological integration of LCA and MCDA, bridging objective impact modeling with value-driven decision support; (ii) it applies this integrated framework within the underexplored context of institutional commuting, offering a replicable model for universities and similar organizations; and (iii) it adopts a forward-looking perspective by evaluating strategic mobility pathways for 2030, thus aligning with the time frames of climate action and sustainability reporting. In doing so, the paper supports more informed, transparent, and environmentally balanced decision-making in pursuing sustainable mobility transitions.

3. Methodology

3.1. Framework Definition

The research followed a structured, four-step approach to develop and apply a sustainability-oriented decision-support framework for commuting (see Figure 1).

The study began by analyzing commuting patterns among a university community. The University of Coimbra served as our case study, a complex institution with over 32,000 community members, including students, faculty, researchers, and technical staff, distributed across four geographically distinct campuses that encompass more than 100 buildings throughout the city. A comprehensive survey was used to assess the community’s commuting habits, willingness to switch to different modes of transportation, and the potential for adopting new transportation options. This survey is publicly accessible, and relevant information can be found in Appendix A. Grounded in this survey, an integrated LCA–MCDA framework was developed, designed to assess environmental impacts across multiple criteria, enabling a comprehensive evaluation of mobility alternatives. The framework was then applied to evaluate seven prospective commuting alternatives, ranging from baseline continuity to enhanced public transportation, active mobility prioritization, and mobility electrification, allowing for a comparison of their sustainability performance. Finally, based on the assessment results, a sustainable commuting plan for 2030 was proposed, offering data-driven recommendations to support institutional decision-making and climate action planning.

3.2. Life Cycle Assessment

This research employed the Organizational Environmental Footprint (OEF) 3.1 method [33]. The OEF method was developed by the European Commission, adapting traditional LCA principles to organizational contexts. Unlike product-focused assessments, the OEF method evaluates sixteen aggregated environmental impacts of an entity’s activities and processes, including direct operations and value chain considerations [34]. This approach is particularly valuable for entities like universities, which function as complex systems with diverse environmental interactions [35].

In this study, the organizational boundary was defined using the operational control approach, as recommended by the OEF 3.1 method [34]. The organizational boundary encompassed all University of Coimbra community members (32,000 individuals, including students, faculty, researchers, and technical staff) based on official institutional census data. The system boundaries followed a WTW approach, including (i) upstream processes such as fuel extraction, fuel production, electricity generation, vehicle manufacturing, and battery production for electric vehicles; (ii) direct operational impacts including fuel consumption, emissions, and infrastructure use; and (iii) infrastructure life cycle impacts from construction, maintenance, and end-of-life, allocated based on passenger–kilometer utilization. Environmental characterization factors were obtained from the Ecoinvent database v3.9, ensuring coverage of the supply chain impacts. This systematic approach ensured that potential burden shifting between life cycle stages was captured, particularly relevant when comparing conventional and electric vehicle alternatives. The assessment framework, as illustrated in Figure 2, established clear parameters for evaluating commuting patterns within this institutional context. The reporting flow was defined as ‘one year of regular travel for classes, research, and administration’, which captured the entire academic cycle of mobility activities. This approach aligns with institutional reporting practices, whilst facilitating meaningful temporal trend analysis.

3.2.1. Inventory

The Life Cycle Inventory (LCI) systematically compiles and quantifies all relevant material and energy flows throughout the defined system. Activities were systematically treated following the ISO 14072:2024 (Requirements and guidance for organizational life cycle assessment) guidelines [36]. The mathematical relationship between activities and environmental impacts can be expressed as

$${\mathrm{EI}}_i=\sum _{j=1}^n{\mathrm{A}}_j\times {\mathrm{EL}}_{i,j}$$

(1)

where ${\mathrm{EI}}_i$ represents the environmental impact for category i, ${\mathrm{A}}_j$ represents the magnitude of activity j (e.g., kilometers traveled by car), and ${\mathrm{EL}}_{i,j}$ is the environmental load vector that converts activity j into impacts i (e.g., kg CO₂ eq per kilometer). This calculation was performed for each impact category across all activities within the organizational boundary.

Data quality was rigorously maintained throughout the assessment process, with primary commuting data collected over five years (2019–2023) through an institutional transportation survey within the University [37]. For the assessments, inventory data collection followed a process-based approach with data in passenger-kilometers (PKM). These survey results were systematically extrapolated based on the annual population figures of the University of Coimbra to provide representative commuting patterns for the entire institutional community. See Appendix B for the detailed calculation steps.

3.2.2. Environmental Impact Assessment

For modeling purposes, SimaPro v.10.1.0.5 with the Ecoinvent database (v3.9) was utilized to establish scientifically validated environmental loads across all transportation modes. The inventory encompasses seven distinct commuting options identified through institutional transportation surveys: electric scooters, conventional buses, motorcycles, bicycles, electric/hybrid vehicles, internal combustion engine vehicles, and walking. Each mode was characterized by its specific life cycle emissions profile, capturing direct operational impacts.

This study employed a midpoint-oriented impact assessment approach. Following the OEF 3.1, the criteria assessed relevant impacts in several environmental aspects. The environmental loads for all categories were kept constant in this study, using the most recent values (based on 2023 data). Active modes were disregarded due to their residual impact.

Table 1. OEF 3.1 Impact Categories.

Abbr.	Unit	Description
AP	mol H+ eq	Acidification potential in soil and water
GWP	kg CO2 eq	Global warming potential
FETP	CTUe	Freshwater ecotoxicity potential
PM	Disease incidence	Cases of disease linked to particulate matter exposure
MEP	kg N eq	Marine eutrophication potential
FEP	kg P eq	Freshwater eutrophication potential
TEP	mol N eq	Terrestrial eutrophication potential
HTPc	CTUh	Comparative toxic units for humans, increased morbidity (cancer effects)
HTPnc	CTUh	Comparative toxic units for humans, increased morbidity (non-cancer effects)
IRP	kBq U-235 eq	Radioactive impact on human health
LU	Pt	Land use impact reflecting soil quality and land transformation
ODP	kg CFC-11 eq	Stratospheric ozone depletion potential
POFP	kg NMVOC eq	Photochemical ozone formation potential
RUF	MJ	Fossil resource use, measured in megajoules
RUM	kg Sb eq	Mineral and metal resource depletion, in kg antimony equivalent
WU	m3 world depriv.	Water use, measured in cubic meters of world-equivalent water

Abbreviations: AP = Acidification Potential; GWP = Global Warming Potential; FETP = Freshwater Ecotoxicity Potential; PM = Particulate Matter; MEP = Marine Eutrophication Potential; FEP = Freshwater Eutrophication Potential; TEP = Terrestrial Eutrophication Potential; HTPc = Human Toxicity (cancer effects); HTPnc = Human Toxicity (non-cancer effects); IRP = Ionizing Radiation Potential; LU = Land Use; ODP = Ozone Depletion Potential; POFP = Photochemical Ozone Formation Potential; RUF = Resource Use, Fossils; RUM = Resource Use, Minerals and Metals; WU = Water Use.

presents the complete set of OEF 3.1 impact categories with their respective units of measurement.

Each transportation mode was associated with an average travel distance and life cycle emissions factor. Modes were characterized not only by tailpipe emissions, but also by full-life-cycle metrics, including fuel production, vehicle manufacturing, and infrastructure use. The environmental models assumed that internal combustion engine cars and motorbikes were propelled by petrol, and the bus transportation was modeled for diesel. The Pedigree matrix approach, a semi-quantitative method for evaluating data quality across multiple dimensions such as reliability, completeness, and temporal correlation, was applied to systematically assess the uncertainty in our inventory data. It estimated the standard deviation (SD) for each activity. To validate the statistical reliability of the 2023 baseline scenario, a Monte Carlo simulation with 1000 iterations was performed, providing a 95% confidence interval for the coefficients of variation (CV), a ratio between the mean and the SD, quantifying model uncertainty (see calculations in Appendix B.3).

3.3. MCDA Foundation and Framework

The PROMETHEE method was chosen for its appropriateness in addressing issues that involve multiple, often conflicting, environmental criteria, especially when both qualitative and quantitative assessments are necessary. Unlike compensatory methods, PROMETHEE enables pairwise comparisons of alternatives across individual criteria, maintaining specific performance trade-offs without assuming full compensability among dimensions. This is particularly important for sustainability evaluations, where improvements in one criterion (e.g., reduced GHG emissions) do not automatically counterbalance significant losses in another (e.g., resource depletion). Additionally, PROMETHEE generates net and partial preference flows, facilitating a more nuanced interpretation of the results, including outranking strengths and weaknesses. Its transparency and ease of integrating decision-maker preferences, while accommodating incomplete or uncertain data inputs, render it a robust and interpretable tool for institutional decision-making contexts. These characteristics have contributed to its effective application in various transport studies, where decision support must reconcile environmental complexity with operational clarity.

Below, we describe the mathematical foundations of the PROMETHEE method and present the steps involved in our MCDA framework assessment.

3.3.1. Method Overview

PROMETHEE is a well-established family of outranking methods used in MCDA, first introduced in the foundational paper by Brans et al. [38]. This method is favored because it can provide a comprehensive ranking of alternatives, while considering multiple, often conflicting criteria. Its ability to incorporate stakeholder-defined weights and preference thresholds makes it particularly effective for evaluating complex sustainability trade-offs in commuting. Among the various MCDA methods available, PROMETHEE was specifically chosen for this study due to its non-compensatory nature, which ensures that strong performance in climate-related metrics does not obscure poor outcomes in other categories [39]. A brief description of the method is provided below.

Let $A=\{a_1,a_2,\dots ,a_n\}$ be the set of alternatives and $G=\{g_1,g_2,\dots ,g_m\}$ the set of evaluation criteria. The PROMETHEE procedure consists of the following steps:

Step 1 (Pairwise performance difference): For each criterion $g_k$, the performance difference between alternatives $a_i$ and $a_j$ is computed as

$$d_k(a_i,a_j)=g_k\left(a_i\right)-g_k\left(a_j\right)$$

(2)

This measures how much alternative $a_i$ outperforms $a_j$ with respect to criterion $g_k$.

Step 2 (Preference function): The difference $d_k(a_i,a_j)$ is then transformed into a preference degree using a criterion-specific preference function $P_k$ (see Appendix C.1 for more details), which can incorporate indifference ($q_k$) and preference ($p_k$) thresholds:

$${\pi }_k(a_i,a_j)=P_k\left(d_k(a_i,a_j)\right),k=1,\dots ,m$$

(3)

Step 3 (Aggregated preference index): To obtain a global measure of preference, the individual preference degrees are aggregated using a weighted sum, where $w_k$ denotes the weight of criterion $g_k$:

$$\pi (a_i,a_j)=\sum _{k=1}^m{w}_k·{\pi }_k(a_i,a_j)$$

(4)

This global preference index $\pi (a_i,a_j)$ quantifies the overall strength of preference for $a_i$ over $a_j$.

Step 4 (Outranking flows): For each alternative $a_i$, the positive and negative outranking flows are computed:

$${\varphi }^{+}\left(a_i\right)={\displaystyle \frac{1}{n-1}}\sum _{x\in A,x\ne a_i}\pi (a_i,x)$$

(5)

$${\varphi }^{-}\left(a_i\right)={\displaystyle \frac{1}{n-1}}\sum _{x\in A,x\ne a_i}\pi (x,a_i)$$

(6)

Here, ${\varphi }^{+}\left(a_i\right)$ represents the degree to which alternative $a_i$ outranks all other alternatives (also referred to as the positive flow or outflow), while ${\varphi }^{-}\left(a_i\right)$ quantifies the extent to which $a_i$ is outranked by the others (the negative flow or inflow).

Step 5 (Net outranking flow and final ranking): The net outranking flow $\varphi \left(a_i\right)$ is obtained by subtracting the negative flow from the positive flow:

$$\varphi \left(a_i\right)={\varphi }^{+}\left(a_i\right)-{\varphi }^{-}\left(a_i\right)$$

(7)

3.3.2. Framework

Figure 3 illustrates the overall process of the MCDA framework applied in this study. The procedure began with the definition of commuting alternatives designed to reflect realistic and policy-relevant mobility scenarios. The next step involved structuring the environmental criteria, derived from the LCA results, ensuring that the evaluation was grounded in comprehensive environmental performance indicators. Once the criteria had been established, a performance matrix was created by assigning quantified impacts to each option. This matrix served as the foundation for the subsequent analysis. The PROMETHEE method was then applied using dedicated software to compute preference flows and rankings. Finally, the results were translated into actionable recommendations, guiding decision-makers toward the most sustainable commuting options for the university context.

We used the Visual PROMETHEE software [40] to enable a systematic comparison and visualization of the alternatives. For the MCDA implementation, equal weighting was applied across all environmental impact categories, assigning a weight of 1/16 to each criterion. This approach was deliberately chosen to avoid subjective bias in the evaluation process and maintain methodological transparency, as differential weighting without strong stakeholder consensus could potentially skew the results toward particular criteria.

Figure 4 illustrates a decision tree of the fundamental objectives used to guide the multi-criteria evaluation of commuting alternatives.

The tree of fundamental objectives presents the decision hierarchy operationalizing the MCDA. At the top of the hierarchy is the main objective: to select the commuting alternative with the lowest environmental impact. This objective was broken down into sixteen evaluation criteria, grouped into three environmental domains: health (e.g., human toxicity, particulate matter), conservation (e.g., climate change, eutrophication, acidification), and resource use (e.g., water and mineral consumption). This structure ensured that the decision-making process comprehensively reflected multiple ecological dimensions and provided a systematic basis for comparing the sustainability performance of each scenario.

3.3.3. Commuting Alternatives

To explore sustainable mobility pathways at the University of Coimbra, a set of prospective alternatives for 2030 was developed. This year aligns with both the United Nations’ Sustainable Development Goals and the University’s carbon neutrality target. Future transportation trends were projected using linear regression on PKM data for each mode of transport (see Appendix C.2). These projections established the baseline activity through 2030. Additionally, the university population was forecast using linear extrapolation of recent annual records, ensuring demographic consistency across all scenarios. The development of these scenarios was informed by survey data showing willingness to adopt different transportation modes, as detailed in Appendix C.2 

Table A10. Willingness to adopt sustainable transportation alternatives among survey respondents.

Transportation Mode	Preference (%)
Electric Scooter	8.74%
Public Transport	29.85%
Motorcycle	1.47%
Bicycle	11.99%
Electric Car	21.36%
Conventional Car	0.00%
Walking	18.69%
Carpooling	7.91%

. The designed alternatives are described below:

• Business-as-Usual (BAU): Assumes the continuation of current modal split patterns through 2030, without significant interventions or policy changes. Projections were based on the full 2019–2023 dataset, including pandemic years, to capture recent variability in mobility behavior. In this scenario, conventional vehicles and walking remain the dominant transportation modes, with smaller proportions of electric/hybrid vehicles, public transport, bicycles, electric scooters, and motorcycles.
• Business-as-Usual Pre-Pandemic Trend (BAUp): Similar to BAU but excludes data from 2020 and 2021 to avoid distortions caused by pandemic-related disruptions. This alternative reflects a trajectory based solely on typical operational conditions.
• Optimistic Shift (OS): Reflects the stated willingness of survey respondents to adopt more sustainable transport modes. This scenario models a complete transition away from conventional vehicles, with their usage being redistributed across alternative modes according to the preferences indicated in the mobility survey. It assumes that all users of conventional fossil fuel vehicles would switch to their preferred alternative mode, creating an ambitious yet preference-based vision of future mobility.
• Public Transport (PT): Models a targeted shift from private vehicles to conventional diesel buses, aligned with survey responses indicating willingness to adopt public transport by 2030. This scenario isolates and implements only the public transport preference segment from the survey data, focusing on the substantial group of respondents who expressed interest in shifting to bus transportation specifically, while other modal preferences remain unchanged from the BAU scenario.
• Public Transport with Electric BRT (PT-BRT): Builds on the previous scenario by incorporating the electric bus rapid transit (BRT) system set for implementation in Coimbra by 2026. The same modal shift assumptions apply as in the PT scenario, but trips are allocated to electric buses instead of diesel buses. This represents a technological enhancement of the public transport strategy without changing behavioral assumptions.
• Active Mobility (AM): Focuses on respondents willing to replace private vehicle use with walking or cycling. This scenario selectively implements the active transportation preferences from the survey, reallocating conventional vehicle trips to walking and cycling based on appropriate distance ranges. Walking is prioritized for shorter trips and cycling for medium-distance commutes, reflecting the practical limitations of these modes.
• Full Electrification (FE): Envisions complete electrification of all motorized transport by 2030, assuming 100% replacement of conventional vehicles (private cars and public buses) with electric counterparts, while maintaining existing modal split proportions. Unlike the preference-based scenarios, this alternative represents a purely technological transition, without behavioral changes in mobility patterns, focusing on the impact of powertrain substitution rather than modal shifts.

Figure 5 shows the 2030 modal share projections for each one of the analyzed alternatives (in PKM%). In both the BAU and BAUp alternatives, walking and using conventional vehicles remained dominant. The PT and PT-BRT alternatives achieved slight reductions in private car usage, accompanied by modest increases in public transport adoption. The OS and AM alternatives introduced more substantial changes, with higher shares of cycling, walking, and the use of electric or hybrid vehicles, highlighting the potential for behavioral shifts to reshape mobility patterns. In contrast, the FE alternative retained modal shares similar to BAU but replaced internal combustion engine vehicles with electric counterparts. Subsequently, all alternatives were evaluated using the OEF 3.1 impact assessment methodology to quantify environmental performance across 16 impact categories.

4. Results and Discussion

This section presents the key findings from the integrated LCA and MCDA. It explores the environmental impacts of current commuting patterns and evaluates prospective mobility alternatives across multiple sustainability criteria. The discussion highlights critical trade-offs and offers insights to guide institutional strategies towards more sustainable mobility solutions.

4.1. Life Cycle Assessment Outcomes

Table 2. Annual per-person environmental impacts (2019–2023) based on the OEF 3.1 categories.

Criteria	Unit (per Person)	2019	2020	2021	2022	2023	$\mathbf{\Delta }$23/19 (%)
AP	mol H eq	1.24	0.30	0.81	1.20	1.19	−4.0
GWP	kg CO2 eq	399	97	261	386	382	−4.3
FETP	CTUe	3050	741	1990	2950	2920	−4.4
PM	Disease incidence	1.90 × 10−5	4.63 × 10−6	1.24 × 10−5	1.84 × 10−5	1.82 × 10−5	−4.1
MEP	kg N eq	0.25	0.06	0.16	0.24	0.24	−3.7
FEP	kg P eq	0.06	0.01	0.04	0.06	0.06	−4.1
TEP	mol N eq	2.57	0.63	1.68	2.49	2.47	−3.9
HTPc	CTUh	3.01 × 10−7	7.33 × 10−8	1.97 × 10−7	2.91 × 10−7	2.88 × 10−7	−4.3
HTPnc	CTUh	3.94 × 10−6	9.59 × 10−7	2.57 × 10−6	3.81 × 10−6	3.77 × 10−6	−4.3
IRP	kBq U-235 eq	11.0	2.69	7.21	10.7	10.6	−4.1
LU	Pt	1900	462	1240	1830	1820	−4.3
ODP	kg CFC-11 eq	9.02 × 10−6	2.19 × 10−6	5.88 × 10−6	8.71 × 10−6	8.62 × 10−6	−4.4
POFP	kg NMVOC eq	1.62	0.40	1.06	1.57	1.56	−3.7
RUF	MJ	5130	1250	3350	4960	4910	−4.3
RUM	kg Sb eq	4.34 × 10−3	1.06 × 10−3	2.83 × 10−3	4.20 × 10−3	4.15 × 10−3	−4.4
WU	m3 water depriv.	35.1	8.54	22.9	34.0	33.6	−4.3

presents the environmental impact intensity metrics per person from 2019 to 2023, demonstrating significant variations in both magnitude and temporal trends.

The final column of Table 2 presents the percentage variation between the values observed in 2023 compared to 2019. These data reveal an overall improvement ranging from 3.7% to 4.4% between 2019 and 2023, indicating a modest but consistent decrease in per-capita environmental impacts related to transportation. Notably, the COVID-19 pandemic created a significant disruption in 2020, with all impact categories showing their lowest values during this period, followed by a gradual rebound through 2021–2023. Although 2023 levels did not exceed those of 2019, they were very close, suggesting a near-complete return to previous mobility patterns.

Within the human-health-related impacts, the indicators demonstrated uniform improvement patterns, with toxicity and radiation metrics showing synchronized decreases. The ecosystem quality indicators exhibited slightly more variability in their reduction profiles, with freshwater ecotoxicity and ozone depletion achieving the most substantial improvements. Similarly, the resource-use metrics showed noteworthy enhancements, particularly in mineral resource depletion. These coherent reductions across diverse impact categories suggest systematic efficiency gains in university commuting patterns, rather than merely shifting environmental burdens between domains. The parallel improvements across distinctly different environmental mechanisms indicated integrated approaches to sustainability within the transportation system.

Monte Carlo simulation identified distinct reliability patterns, with ecosystem-quality-related indicators showing the most significant statistical validity—climate change (CV 11.9%), acidification (CV 15.3%), and particulate matter (CV 17.4%) all displaying coefficients of variation below 20%. On the other hand, impact categories related to human health had the highest CVs. The WU also showed high variability, due to its modeling, which included the treatment and return of water to the environment. Further details are available in Appendix B.

4.2. Multi-Criteria Decision Analysis

4.2.1. Performance Matrix

Table 3. Performance matrix of commuting alternatives across environmental impact criteria.

Criteria	Units	BAU	BAUp	OS	PT	PT	AM	FE
	(per Person)					BRT
AP	mol H eq	1.21	1.09	0.79	0.80	0.78	0.77	1.52
GWP	kg CO2 eq	388	347	132	231	230	229	258
FETP	CTUe	2960	2650	836	1750	1730	1720	1560
PM	Disease incidence	1.85 × 10−5	1.66 × 10−5	9.77 × 10−6	1.19 × 10−5	1.16 × 10−5	1.16 × 10−5	1.84 × 10−5
MEP	kg N eq	0.24	0.22	0.18	0.18	0.16	0.16	0.29
FEP	kg P eq	0.06	0.05	0.06	0.04	0.04	0.04	0.13
TEP	mol N eq	2.50	2.28	1.77	1.83	1.69	1.67	2.82
HTPc	CTUh	2.93 × 10−7	2.63 × 10−7	1.39 × 10−7	1.80 × 10−7	1.77 × 10−7	1.81 × 10−7	2.73 × 10−7
HTPnc	CTUh	3.83 × 10−6	3.43 × 10−6	2.39 × 10−6	2.36 × 10−6	2.38 × 10−6	2.39 × 10−6	5.20 × 10−6
IRP	kBq U-235 eq	10.7	9.67	12.7	7.31	7.81	7.35	29.5
LU	Pt	1840	1650	714	1110	1120	1100	1440
ODP	kg CFC-11 eq	8.75 × 10−6	7.82 × 10−6	2.05 × 10−6	5.06 × 10−6	5.01 × 10−6	5.00 × 10−6	3.97 × 10−6
POFP	kg NMVOC eq	1.58	1.43	0.98	1.10	1.05	1.05	1.57
RUF	MJ	4980	4460	1730	2980	2970	2940	3400
RUM	kg Sb eq	4.21 × 10−3	3.77 × 10−3	3.07 × 10−3	2.61 × 10−3	2.61 × 10−3	2.62 × 10−3	6.99 × 10−3
WU	m3 water depriv.	34.1	30.6	23.3	21.3	21.8	21.5	51.3

presents the performance matrix used to perform the MCDA, considering the seven alternatives over the 16 criteria. Unlike previous commuting research that focused primarily on CO₂ emissions [6,11,12], this comprehensive approach prevented the problem shifting identified in single-indicator studies and revealed trade-offs between impact categories that would otherwise remain hidden. For instance, our analysis showed that electrification reduced GHG emissions but increased mineral resource depletion and water use, environmental consequences that carbon-only assessments systematically overlook.

The comparative assessment of the 2030 projected scenarios against the 2023 baseline revealed distinct environmental performance patterns. The Optimistic Shift (OS) scenario demonstrated the most substantial improvements across multiple impact categories, exhibiting remarkable reductions of 65.4% in GWP (from 382 to 132 kg CO₂ eq) and 71.4% in FETP (from 2.92 × 10³ to 8.36 × 10² CTUe). Conversely, the Full Electrification (FE) scenario presented contradictory outcomes, with modest improvements in atmospheric emissions indicators juxtaposed against significant deteriorations in resource depletion metrics. Particularly concerning are the increases of 38.0% in HTP_nc (from 3.77 × 10⁻⁶ to 5.20 × 10⁻⁶ CTUh), 52.7% in WU (from 33.6 to 51.3 m³ world eq), and 68.4% in RUM (from 4.15 × 10⁻³ to 6.99 × 10⁻³ kg Sb eq) compared to the 2023 values. The active mobility (AM) and public transport scenarios (PT and PT-BRT) occupied intermediate positions, with consistently favorable profiles, demonstrating reductions of approximately 40% in most impact categories relative to 2023 baseline measurements. This evidence substantiated the hypothesis that mobility strategies predicated on collective and non-motorized transportation modalities offer a superior environmental equilibrium compared to mere technological substitution of conventional vehicles with electric alternatives.

The performance matrix highlights how each alternative addresses multiple environmental criteria, revealing that specific options offer more balanced and sustainable outcomes. This reinforces the value of applying MCDA in mobility planning, as it enables a holistic assessment beyond single-focus metrics. Without such a comprehensive approach, critical trade-offs, such as shifting environmental burdens across impact areas, may be overlooked.

4.2.2. Outranking Results

Table 4. PROMETHEE ranking results for commuting scenarios.

Rank	Scenario	$\mathsf{\Phi }$	${\mathsf{\Phi }}^{\mathbf{+}}$	${\mathsf{\Phi }}^{\mathbf{-}}$
1	Optimistic Shift (OS)	0.389	0.414	0.025
2	Active Mobility (AM)	0.262	0.292	0.030
3	Public Transportation with BRT (PT-BRT)	0.255	0.286	0.031
4	Public Transportation (PT)	0.231	0.271	0.041
5	Business-as-Usual Pre-Pandemic Trend (BAUp)	−0.230	0.087	0.317
6	Business-as-Usual (BAU)	−0.399	0.046	0.446
7	Full Electrification (FE)	−0.508	0.059	0.566

presents the results of the PROMETHEE outranking analysis, showing the complete ranking of scenarios, along with their net flow ($\mathsf{\Phi }$), positive flow (${\mathsf{\Phi }}^{+}$), and negative flow (${\mathsf{\Phi }}^{-}$) values.

The MCDA analysis revealed a clear preference structure among the commuting alternatives. OS emerged as the most environmentally preferable option with the highest net flow ($\mathsf{\Phi }$ = 0.3892), followed by AM ($\mathsf{\Phi }$ = 0.2623). These two leading alternatives both demonstrated strong positive flows (${\mathsf{\Phi }}^{+}$ > 0.29) and remarkably low negative flows (${\mathsf{\Phi }}^{-}$ < 0.03), indicating consistent outranking of other alternatives across multiple impact categories. Public transportation alternatives occupied the middle positions in the ranking, with PT-BRT ($\mathsf{\Phi }$ = 0.2550) slightly outperforming the conventional PT scenario ($\mathsf{\Phi }$ = 0.2305). Both showed respectable positive flows but marginally higher negative flows than the top-ranked scenarios, reflecting their mixed performance across different environmental impact categories.

Notably, FE ranked last, with the most negative net flow ($\mathsf{\Phi }$ = −0.5078), positioning it below even the BAU scenarios. This unexpected result contrasts with its strong performance in specific individual impact categories, such as climate change. This low ranking in the multi-criteria evaluation stemmed from substantial environmental trade-offs identified in the performance matrix, particularly in categories related to resource depletion and ecosystem impacts. This finding underscores the importance of comprehensive environmental assessment beyond carbon-centric approaches when evaluating transportation strategies. Both BAU alternatives occupied intermediate-to-low positions in the ranking, with BAUp performing marginally better ($\mathsf{\Phi }$ = −0.2298) than BAU ($\mathsf{\Phi }$ = −0.3994). These results confirm that most intervention scenarios represent environmental improvements over business-as-usual conditions, although to varying degrees across the different impact categories.

Figure 6 highlights the strengths and weaknesses of each alternative across the environmental impact categories. In this graph, the horizontal axis lists the alternatives, while the vertical axis shows how well they performed for each criterion.

The OS alternative displays the most significant $\mathsf{\Phi }$, reflecting its superior performance in the criteria such as GWP, LU, and FETP. However, it presented minor weaknesses in IRP, represented in blue, indicating upstream energy-related impacts. The AM and PT-BRT alternatives show well-balanced environmental profiles, demonstrating strong results across most impact categories, particularly in reducing emissions and ecosystem impacts. This highlights the benefits of promoting active mobility and electrified public transport solutions. In contrast, the PT alternative has a more heterogeneous pattern, combining positive outcomes in specific categories and negative impacts in others. This suggests that while shifting to public transport offers environmental benefits, conventional diesel buses still have notable drawbacks in particular areas. The most pronounced trade-offs were evident in the FE scenario, which displays a strong polarization. Characterized by significant negative impacts in resource-use criteria, particularly RUM and WU, represented by deep brown segments, it still achieved positive outcomes in several metrics, including reductions in GWP. Both the BAU and BAUp scenarios exhibited consistently poor environmental performance across nearly all impact categories, confirming their lack of sustainability in the absence of intervention. However, unlike FE, they did not present extreme negative peaks in specific areas. This multidimensional visual analysis underscores the importance of comprehensive sustainability assessments. It reveals how technology-driven transitions, such as electrification, while beneficial for climate-related goals, can inadvertently introduce substantial environmental burdens in other areas.

The results presented in Figure 7 illustrate the performance of each alternative individually, highlighting the evaluated criteria. The grey shading is the $\mathsf{\Phi }$ alternative value.

4.3. Discussion: Implications for Institutional Mobility Planning

The findings of this study highlight the pressing need to adopt holistic, system-oriented approaches rather than relying solely on technology-driven solutions for sustainable mobility in institutional settings. In this context, the following discussion outlines the key insights derived from the integrated LCA–MCDA analysis:

Electrification: While electrification remains a cornerstone of many climate policies due to its capacity to reduce tailpipe emissions, our analysis revealed the hidden complexities and trade-offs associated with this pathway. Echoing concerns raised by Hawkins et al. [7] and Messagie et al. [8], the FE alternative demonstrated how shifting from internal combustion engines to electric vehicles can displace environmental burdens upstream, particularly in the form of increased ionizing radiation, mineral resource depletion, and water consumption due to battery production and electricity generation. The FE scenario demonstrated a 66% increase in RUM (6.99 × 10⁻³ vs. 4.21 × 10⁻³ kg Sb eq) and 50% increase in WU (51.3 vs. 34.1 m³) compared to BAU, whilst achieving only a 33% reduction in GWP (258 vs. 388 kg CO₂ eq). These systemic impacts align with the warnings of Manzolli et al. [5] regarding the broader consequences of electrification for energy grids and resource flows. This evidence suggests that electrification, while necessary, is insufficient as a standalone solution. Institutions must adopt complementary strategies that address supply chain decarbonization, promote circular economy practices in battery life cycle management, and ensure that clean energy sources underpin increased electricity demand. Without such measures, electrification risks becoming a form of ‘green shifting’, where emissions reductions mask more profound ecological costs.

Active Mobility: In contrast, active mobility emerged as an ecologically robust strategy. The consistently superior performance of the AM alternative across all environmental impact categories reinforced the principle that the most sustainable mode of transport is often the one that avoids motorization altogether. The AM scenario achieved the lowest environmental impacts across 14 of 16 categories, including 41% lower GWP (229 vs. 388 kg CO₂ eq), 37% reduction in PM impacts (1.16 × 10⁻⁵ vs. 1.85 × 10⁻⁵ disease incidence), and substantial reductions in RUM, ranking second in PROMETHEE analysis ($\mathsf{\Phi }$ = 0.262). This supports prior research by Guido et al. [3] and Aytekin et al. [4], emphasizing the environmental and social co-benefits of walking and cycling in campus environments. Beyond reducing emissions, active mobility promotes healthier lifestyles, reduces infrastructure strain, and requires minimal resource input compared to other transportation alternatives. Despite demonstrating superior environmental performance across all impact categories, active mobility faces adoption challenges, with only 30.68% of survey respondents willing to switch to walking or cycling compared to 51.21% preferring motorized alternatives. This gap reflects institutional barriers including campus sprawl, inadequate cycling infrastructure, weather dependency, and safety concerns. Moreover, the topographical characteristics of Coimbra city, marked by steep slopes and elevation changes, create significant barriers to pedestrian and bicycle mobility. Thus, to fully unlock its potential, targeted policies are essential to make active mobility more appealing, safe, and practical. These include the development of continuous and protected cycling lanes, improved pedestrian pathways with adequate lighting and shading, secure bicycle parking facilities, and integrating bike-sharing systems tailored to campus needs. Additionally, incentive programs such as subsidies for bicycle purchases, rewards for active commuting, and awareness campaigns highlighting health and environmental benefits can further encourage adoption. Importantly, our findings show that this low-impact pathway aligns with user preferences, as reflected in the mobility survey, making it environmentally sound and socially acceptable when supported by proper infrastructure and policy frameworks.

Public Transportation: In particular, its electrified form (PT-BRT) offers a pragmatic middle ground. While its environmental performance did not match AM’s, it significantly outperformed private motorized transport in key areas. The PT-BRT scenario reduced GWP by 41% compared to BAU (230 vs. 388 kg CO₂ eq) and PM impacts by 37%, ranking third in the PROMETHEE analysis ($\mathsf{\Phi }$ = 0.255), whilst demonstrating mixed performance in the resource depletion categories. The PT-BRT scenario demonstrated how modernizing collective transport systems can yield meaningful reductions in eutrophication and emissions, positioning it as a viable transitional solution for institutions seeking scalable, low-carbon commuting options. However, as highlighted by Saxe et al. [11], the true sustainability of public transport hinges on life cycle considerations, including vehicle manufacturing and infrastructure development. Moreover, policy effectiveness in promoting public transport depends not only on environmental performance but also on public perception and service quality. Chronic issues, such as delays, infrequent service, and outdated infrastructure, undermine efforts to encourage a modal shift. Therefore, institutions and municipalities must implement policies that enhance user experience, such as fare incentives (e.g., discounted rates for students, seniors, or off-peak travel), improved service frequency, and infrastructure upgrades, to make public transport a more attractive alternative to private car use.

A key takeaway from this study is the importance of grounding mobility strategies in empirical user data and systemic thinking. By integrating behavioral insights from the 2022 mobility survey, scenario modeling captured not only technical possibilities but also social realities, bridging the gap between aspirational planning and feasible outcomes. This analysis revealed critical disparities between environmental performance and user acceptance: whilst OS ranked highest ($\mathsf{\Phi }$ = 0.389) and AM second ($\mathsf{\Phi }$ = 0.262) environmentally, the behavioral data indicated only 30.68% of respondents would adopt active mobility compared to 51.21% preferring motorized alternatives. This user-centric approach is essential as institutions face growing pressure to report on sustainability transparently and comprehensively under frameworks such as the European Green Deal and the United Nations’ Sustainable Development Goals. Universities, in particular, are uniquely positioned as catalysts for sustainable mobility transitions, not merely as transport infrastructure providers, but as influential actors shaping cultural norms and mobility behaviors. Leveraging integrated assessment tools like the LCA–MCDA framework enables institutions to craft policies that are environmentally sound, socially responsive, and aligned with broader urban sustainability objectives. This study highlights that true sustainability goes beyond technological optimism, requiring context-sensitive strategies that minimize environmental burdens across all impact dimensions, not just carbon emissions.

5. Conclusions

This study demonstrates that promoting sustainable commuting transitions within institutional settings requires a holistic multi-criteria approach that goes beyond traditional carbon-focused assessments. By integrating LCA with MCDA, this study provided a comprehensive evaluation of seven commuting alternatives across 16 environmental impact categories. The findings clearly demonstrated that active mobility, such as walking and cycling, consistently achieved the lowest environmental burdens, offering ecological, social, and health co-benefits when reinforced by supportive infrastructure, safety measures, and incentive policies. The pronounced elevation differences and hilly terrain of Coimbra significantly hinder the feasibility of pedestrian and cycling mobility. Therefore, additional efforts will be needed to encourage a change in mobility habits among the academic community. In contrast, while electrification remains a key pillar in climate strategies, our analysis highlighted critical upstream trade-offs, including increased mineral resource use, water consumption, and ionizing radiation impacts, often overlooked in simplified assessments. Electrified public transport, particularly in its bus rapid transit form, emerged as a balanced and scalable solution, though its success hinges on improvements in service quality, infrastructure, and user perception. Overall, the study advocates mobility planning that embraces modal diversity and avoids one-size-fits-all approaches. By anchoring strategies in empirical user data and system-level evaluation, institutions can implement mobility transitions that are environmentally sound, socially acceptable, and operationally feasible. In this context, sustainable mobility planning must consider the full spectrum of environmental impacts, to avoid unintended trade-offs that can undermine long-term sustainability goals.

This work has some limitations that should be acknowledged. It relied on static assumptions about behavioral shifts from a single-year mobility survey, which may not accurately capture evolving user preferences or external influences, such as policy changes. The use of fixed environmental loads and a static energy mix did not account for the potential decarbonization of the electricity grid, affecting the environmental profiles of electrification scenarios. Additionally, LCI data may not reflect local specifics in transport and energy production. While the MCDA process was comprehensive, socio-economic and equity dimensions were excluded, limiting the analysis to environmental impacts and overlooking important aspects for sustainable mobility planning. Future research should address these limitations by incorporating dynamic modeling of energy systems, policy scenarios, and real-time mobility data. Expanding the framework to include socio-economic, equity, and accessibility considerations would provide a more comprehensive assessment of sustainable mobility transitions. Moreover, spatial analysis could offer deeper insights into infrastructure needs and urban integration.

In conclusion, while cleaner technologies like electrification can play a role in decarbonizing transportation, this study highlights that the most resilient and sustainable pathways involve fostering behavioral change, reducing reliance on motorized transportation, and designing mobility systems that are context-sensitive, resource-efficient, and socially inclusive. As microcosms of larger urban systems, universities have both the opportunity and responsibility to lead this shift by fostering mobility ecosystems that are low-emission, low-impact, equitable, and resilient.