The literature search strategy was designed using comprehensive Boolean logic, targeting studies focused explicitly on how MaaS influences sustainable travel mode choices in urban contexts. The search string combined multiple synonymous and alternative terms to ensure an inclusive and extensive search across titles, abstracts, and keywords:

TITLE-ABS-KEY[(“Sustainable Transport^*” OR “Eco-friendly Travel” OR “Sustainable Mobility” OR “Low-carbon Transport” OR “Green Transport”) AND (“Travel Behavior” OR “Travel Pattern” OR “Mobility Behavior” OR “Transport Behavior” OR “Mode Choice” OR “Modal Shift”) AND (“Shared Car^*” OR “Car sharing” OR “Shared Bike^*” OR “Bike Sharing” OR “Shared electronic Bike^*” OR “E-bike Sharing” OR “Shared Scooter^*” OR “Shared Mobility” OR “MaaS” OR “Mobility-as-a-Service” or “Mobility as a Service” OR “MaaS trial^*”) AND (“City^*” OR “Urban” OR “Metropolitan”)]

The search covered peer-reviewed journal articles and conferences papers published in venues with editorial or peer-reviewed standards, as well as relevant gray literature published up to May 2025. Initially, 2,402 records were initially retrieved. Articles were first screened to remove duplicates, non-English publications, and those lacking full-text availability (e.g., abstracts-only conference proceedings), thus ensuring the comprehensiveness and quality of the analysis (Song et al., 2010). Review articles were excluded to prevent double-counting of primary data (Edinger and Cohen, 2013). This initial screening process reduced the dataset to eight hundred and fifteen papers.

Following the initial screening, a second stage was conducted to assess the relevance of the remaining articles through a review of their titles and abstracts. This step focused on determining alignment with the study's aims, particularly the relationship between MaaS, sustainable urban mobility, and travel behavior. Articles clearly unrelated to these key themes were excluded at this stage. For example, studies focused exclusively on rural or tourist contexts were excluded, as their transportation dynamics, such as lower population densities and seasonal travel patterns, which differ markedly from urban settings where MaaS is most applicable (Li and Voege, 2017; Maas, 2022; Martinčević et al., 2022; Mitropoulos et al., 2023). This stage resulted in a refined set of 197 articles for full-text review.

The final stage involved a detailed, full-text review using predefined inclusion and exclusion criteria (Table 1), which were developed to ensure alignment with the research objectives and support a consistent and transparent evaluation process. This review was independently conducted by two authors, with disagreements resolved through discussion to enhance reliability and minimize selection bias. Additionally, backward and forward citation tracing methods were utilized to identify relevant studies potentially missed through database queries (Hirt et al., 2021). Following this comprehensive screening process, eighty five studies were included in the final synthesis: sixty four from full-text review and twenty one additional records (twelve from citation tracing, nine gray literature sources). The PRISMA flow diagram (Figure 1) provides a visual summary of the review process and exclusion rationale at each stage. A complete summary of all included studies is provided in Supplementary Table S1.

To facilitate a consistent and rigorous synthesis process, two authors independently extracted key information from each included study using a standardized data extraction template. After initial extraction, the results were then cross-checked and consolidated into a shared spreadsheet following discussion and consensus. The recorded variables included the article title, authorship, year of publication, publisher, country or region studied, research methodology, participant demographic characteristics, relevant MaaS modes, and key findings. This process helped reduce the risk of omission and improved the transparency and reliability of the review.

Dues to the diversity of methodological approaches and research contexts among the included studies, such as surveys, interviews, and case studies, a quantitative meta-analysis was not feasible. Instead, a mixed-methods synthesis was employed, integrating descriptive analysis to map research characteristics and systematic analysis organized around the three guiding research questions to examine MaaS adoption and sustainable travel behavior patterns. This synthesis approach included systematic comparison across studies to examine variations in findings, assess evidence quality across different contexts, and identify areas where evidence converges or diverges in relation to each research question. Data extraction and coding were conducted using NVivo software, which enabled the systematic categorization of findings according to the research question framework: (1) user group differences in MaaS adoption and sustainable behavior, (2) transport mode contributions to sustainable travel behaviors, and (3) adoption factors and their interactions. This iterative coding process organized findings into research question-focused categories. Coding was performed by one author and verified by the second, with iterative discussions to resolve discrepancies, and cross-checking with the raw data to ensure rigor and comprehensiveness of the analysis (Houghton et al., 2017).

Academic research on MaaS has gradually evolved from early conceptual discussions to a more practice-oriented and interdisciplinary field. As shown in Figure 2, the period from 2015 to 2017 saw limited publication activity, and with a primary focus on technical and conceptual work within Transportation Engineering and Information Technology. These initial contributions laid theoretical foundations but remained largely untested in practical contexts. After 2018, publication volume increased notably, accompanied by a broadening of disciplinary engagement. In particular, contributions from Psychology, Behavioral Sciences, and Urban Planning became more prominent. This shift reflects growing academic interest in understanding how MaaS functions in real-world settings—emphasizing user behavior, governance, and policy integration, in addition to technical feasibility. Research during this period increasingly examined implementation processes and challenges, reflecting a shift toward more empirical and applied investigations. By 2022, studies grounded in behavioral and psychological approaches accounted for a significant portion of the literature, highlighting the field's growing interest in connecting system design with user engagement and policy support.

At the same time, this evolving focus raises important questions about how well current research aligns with MaaS's sustainability goals. Although MaaS is frequently positioned as a strategy to promote sustainable mobility, Figure 2 shows comparatively limited engagement from disciplines such as Sustainable Development and Environmental Studies during the same period that practical implementations expanded globally. While research into user behavior and system operations has gained prominence, there appears to be less systematic attention to evaluating whether MaaS delivers on its environmental and equity-related promises. This trend invites further reflection on research priorities and whether sufficient evidence is being generated to support sustainability-oriented decision-making.

Co-citation network analysis further examines the intellectual foundations of MaaS sustainability research. As visualized in Figure 3, three primary research clusters have emerged: a core transportation research cluster centered around Transportation Research journals (Parts A, C, and D), a sustainability-focused cluster including environmental and energy journals, and a behavioral sciences cluster encompassing psychology and urban planning perspectives. These clusters demonstrate the field's interdisciplinary nature but also suggest limited intellectual integration across disciplinary boundaries.

Transportation Research journals occupy central positions, reflecting their role as primary venues for MaaS empirical studies and methodological approaches. In contrast, journals with a focus on sustainability are included in the network but tend to be positioned at the margins, particularly when compared to the dense connections among transportation-focused publications. This distribution suggests that, although sustainability is often cited as a guiding objective of MaaS, environmental evaluation has not yet become a central concern within the core empirical literature, which remains more strongly oriented toward technical, policy, and behavioral aspects of implementation.

While the presence of multiple clusters reflects the field's interdisciplinary scope, the network also points to limited cross-referencing between domains whose integration is critical to advancing MaaS as a sustainable solution. Connections between behavioral science and sustainability journals remain relatively weak, despite the clear relevance of both perspectives to understanding how MaaS can support environmental goals through changes in individual travel behavior. Likewise, journals focused on urban planning and information technology are positioned at the margins of the network, suggesting that important spatial and digital aspects of MaaS have not been fully integrated into the core research dialogue. Policy journals tend to draw more heavily on technical transportation studies than on environmental or social science literature, which may limit the development of more holistic or innovative policy approaches. Overall, while the field has attracted contributions from multiple disciplines, closer intellectual exchange is still needed to address the complex challenges of implementing MaaS as a truly sustainable mobility solution.

This section provides a critical evaluation of methodological approaches across the 85 reviewed studies, examining how different research designs affect findings reliability and contribute to inconsistent results in the literature. Survey-based approaches dominate the MaaS literature, with quantitative cross-sectional studies representing the largest category (n = 49), complemented by stated choice experiments (n = 5) that explore hypothetical adoption scenarios. These studies typically apply statistical techniques such as structural equation modeling, logistic regression, and factor analysis, often drawing on the Theory of Planned Behavior and the Technology Acceptance Model. While these approaches provide rigorous evidence on attitudes, preferences, and adoption intentions, their reliance on cross-sectional designs and self-reported data means that they primarily capture perceptions and short-term intentions rather than sustained behavioral change. Empirical findings from the reviewed studies show that environmental attitudes often fail to translate into sustained behavior change, with convenience being a stronger motivator than environmental concerns (Acheampong et al., 2021). However, the predominant use of cross-sectional designs limits our confidence in understanding these relationships over time. Longitudinal field experiments (n = 8) address this limitation by tracking actual travel behavior over time through multi-source data, thereby moving beyond one-time assessments or self-reported intentions. (Sochor et al. 2016) UbiGo trial in Sweden illustrates this approach through mixed-method design combining questionnaires, interviews, travel diaries, and usage data to reveal both behavioral changes and underlying user motivations. Similarly, (Hensher et al. 2021) Sydney MaaS field trial demonstrates how integrating multiple data sources with joint discrete-continuous modeling can examine the relationship between subscription bundle choices and travel behavior patterns.

Mixed-methods studies (n=6) further strengthen the evidence base by triangulating quantitative behavioral data with qualitative insights. For example, (Henderson et al. 2022) combined secondary car club membership data with semi-structured interviews to understand behavior change, while (Hartl et al. 2018) integrated a cross-sectional survey, an experimental design, and exploratory interviews. These designs provide detailed causal accounts of how adoption connects to everyday travel choices and allow for more accurate calibration of models that might otherwise overstate adoption effects. Similarly, qualitative studies (n = 5) using interviews and focus groups provide deep contextual insights understanding of how travelers interpret MaaS and adapt it within their daily mobility routines. Policy and guidance documents (n = 8) complement empirical research by synthesizing implementation experiences across diverse contexts, helping identify institutional and infrastructural factors that enable or constrain MaaS success. These analyses prove particularly valuable given MaaS's dependence on complex stakeholder coordination and regulatory frameworks that vary significantly across different urban contexts. Table 2 presents the methodological diversity in studies investigating the impact of MaaS on sustainable travel behavior, highlighting the growing development of approaches over the past decade. However, the field would benefit from more longitudinal studies and real-world behavioral tracking to better understand how initial adoption translates into sustained travel behavior change.

The results of the eighty five review studies revolve around three guiding research questions, examining how MaaS adoption relates to sustainable travel outcomes across diverse user groups, transport modes, and and contextual conditions. Across these dimensions, the review draws on convergent and divergent evidence, highlighting the consistency and variability of results across different contexts, study designs, or user characteristics. The analysis begins with demographic influences on adoption and sustainability, highlighting contrasting patterns across age, gender, and income groups that challenge common assumptions about technology-led sustainable mobility. It then examines how different transport modes within MaaS systems can both advance and undermine environmental goals. Finally, the synthesis identifies service design, cost, and policy incentives as the most consistently reported drivers of MaaS adoption, while also noting differences in how these drivers operate across settings and interact to shape both uptake and sustainability outcomes.

This systematic review of eighty five studies provides a comprehensive examination of MaaS research development and its relationship with sustainable travel behaviors. The temporal analysis reveals that MaaS research has evolved from early technical and conceptual work toward more practice-oriented investigations, with increasing contributions from psychology and behavioral sciences after 2018. However, co-citation network analysis indicates limited integration between sustainability-focused journals and the core transportation research literature, suggesting that environmental evaluation has not become central to empirical. Methodologically, the field remains dominated by quantitative, survey-based research that offers valuable evidence on adoption intentions, but the predominance of cross-sectional designs over longitudinal approaches limits insight into how these intentions translate into long-term behavioral change. Findings on MaaS adoption and sustainable travel behavior indicate that platform uptake is largely driven by convenience and technological appeal, with environmental motivations playing a secondary role for most users. The sustainability contribution of specific modes varies: public transport integration generally supports positive environmental outcomes, active mobility services can displace other sustainable options, and ride-hailing often diverts users from public transport. Adoption factors reveal complex interactions between convenience features, economic incentives, and policy frameworks that drive platform uptake but may work against sustainability objectives. While MaaS platforms successfully engage users through these combined approaches, sustained engagement does not guarantee more sustainable travel choices.

This section presents a conceptual framework (Figure 4) examining how multiple factors interact to influence MaaS adoption and its capacity to support sustainable travel behavior. The analysis reveals that adoption and sustainability are shaped by different processes and do not necessarily occur together, with their relationship shaped by complex interactions within broader social, technical, and institutional environments. MaaS platforms can improve accessibility and reduce private car dependence, but they also carry forward features of existing transport systems that can limit broader change. Current implementation approaches that prioritize convenience, technological appeal, and market expansion over environmental effectiveness create situations where successful user acquisition may not translate into sustainable travel behaviors. Understanding this dynamic requires examining how demographic characteristics, service design choices, economic incentives, and policy frameworks function as interconnected elements that can either reinforce existing mobility patterns or facilitate transitions toward sustainable behaviors, depending on how their interactions are coordinated within specific contexts. The framework developed here provides a systematic approach for analyzing these interdependencies and identifying pathways for aligning MaaS development with sustainability objectives.

At the systemic level, governance emerges not merely as external regulation but as a component that determines the operational framework within which MaaS platforms function. As explored by (Marsden and Reardon 2017), governance is integrated within and responsive to evolving transportation system requirements, yet evidence indicates that current governance approaches often conceptualize MaaS as a technological rather than socio-political phenomenon. Policy decisions guided by regulatory structures, public-private partnerships, and funding mechanisms significantly influence enabling conditions for MaaS systems, establishing how infrastructure, pricing models, service delivery, and user participation correspond with sustainability objectives. However, effective governance needs to recognize that MaaS platforms are active participants in the development of mobile culture, rather than passive service providers, with the ability to perpetuate or change existing unsustainable patterns of travel behavior. The variation in policy effectiveness across institutional contexts demonstrates that successful governance necessitates intervention addressing cultural and economic transformation rather than relying solely on regulatory management.

At the implementation level, service design and economic incentives act as the primary mechanisms through which governance objectives are translated into user decision-making. Their influence extends beyond providing functional choices, as they structure how different modes are perceived in terms of convenience, value, and social appropriateness. Pricing strategies, interface layouts, and default journey suggestions not only determine the relative attractiveness of available modes but also convey implicit priorities regarding which options should be used. When these mechanisms are explicitly aligned with sustainability goals—for example, by designing pay-as-you-go (PAYG) options or monthly bundles that make shared mobility options financially attractive relative to private car use—they can encourage behavioral patterns consistent with environmental objectives (Hensher et al., 2021). In contrast, when design and incentive structures treat all modes as equally desirable, decisions are more likely to be shaped by established habits and short-term convenience, which often favor less sustainable options. This presents that effective implementation requires the deliberate integration of service design and pricing structures with policy aims, ensuring that sustainable modes are positioned as the most accessible and attractive options within the platform environment.

At the user engagement level, the interaction between demographic characteristics and platform implementation influences which sustainability outcomes MaaS can realistically achieve. Targeting strategies and design choices act as a selection process, shaping both the types of users who adopt the platform and the travel behaviors that become dominant. Platforms designed for broad market appeal, with a focus on convenience, wide service coverage, and technological features, often attract users motivated by personal efficiency and flexibility rather than environmental goals. This can result in travel patterns that diverge from sustainability objectives, reinforcing the behaviors MaaS seeks to change. Evidence indicates that engaging users whose values and practices already align with sustainable mobility can produce stronger and more durable environmental outcomes, as voluntary shifts away from car-centered habits prove difficult to achieve (Haustein and Kroesen, 2022). This approach involves emphasizing reliability, safety, and low environmental impact in platform features, structuring services so that public transport forms the core offer while limiting high-emission modes, and using targeted incentives to make sustainable choices the most attractive. While such an approach may initially lead to a smaller but more committed user base, it can establish behavioral norms that, once embedded, provide a foundation for gradually engaging wider audiences without eroding environmental benefits.