Mapping the Institutional and Socio-Political Barriers to Smart Mobility Adoption: A TISM-MICMAC Approach

Highlights

Despite extensive research on barriers to smart mobility, prior studies have largely treated them in isolation, overlooking their systemic interconnections and root causes. This study employs Total Interpretive Structural Modelling (TISM) and Matrice d’Impacts Croisés Multiplication Appliquée à un Classement (MICMAC) analysis, underpinned by Critical Urban Theory, to structurally map and interpret the interrelationships among the barriers.

What are the main findings?

• Legacy paradigms in conventional transport planning, fragmented mandates, and outdated regulations emerge as root causes of resistance to smart mobility.
• User-related issues such as affordability, spatial coverage, and cultural attachment to cars are dependent barriers, shaped by upstream institutional and governance failures.

What is the implication of the main finding?

• Tackling smart mobility challenges requires structural reforms that dismantle entrenched planning logics and improve cross-institutional coordination.
• Policymakers and planners, particularly in the Global South, should prioritize high-driving-power barriers to unlock cascading benefits across the mobility system.

Abstract

Smart mobility is widely promoted as a solution to urban congestion, pollution, and inefficiency. Yet, its adoption remains inconsistent, particularly in developing and small cities. While prior research has examined technological enablers, the structural and systemic barriers that constrain adoption are less understood. This study identifies and analyzes the institutional, political, technological, and socio-cultural barriers that collectively inhibit smart mobility transitions. Using Total Interpretive Structural Modelling (TISM) and MICMAC analysis, the study hierarchically maps 14 interrelated barriers derived from literature and validated through expert consultation. Findings reveal that legacy paradigms in conventional transport planning, fragmented institutional mandates, and regulatory misalignment are the foundational constraints that reinforce downstream challenges such as affordability, limited service coverage, and user resistance. Anchored in Critical Urban Theory, the study reframes smart mobility adoption as a contested and political process shaped by institutional inertia and unequal access to technology. The paper contributes to the literature by offering a theory-informed diagnostic framework for understanding mobility transitions. It also provides practical entry points for policymakers, planners, and mobility innovators seeking to target root cause interventions rather than symptoms, to enable more equitable, scalable, and resilient smart mobility transitions.

1. Introduction

Cities worldwide are under mounting pressure to transform their transport systems in response to rapid urbanization, climate imperatives, and digital innovation. Against this backdrop, smart mobility—the integration of digital technologies, multi-modal transport options, and user-centered services—has emerged as a potential solution to longstanding urban mobility challenges [1,2]. It refers to the intelligent application of technology, data, and connectivity to create transportation systems that are efficient, sustainable, and user-friendly, enhancing accessibility and urban livability [3,4,5,6].

Yet, despite its transformative potential, the adoption of smart mobility is inconsistent, particularly in developing and smaller cities that face persistent structural and institutional constraints alongside limited technical capacity. Existing research has largely focused on enabling factors such as service efficiency, digital infrastructure and perceived usefulness [7,8,9], but offers limited insight into how barriers may impede adoption. Furthermore, studies on barriers have explored them in isolation, overlooking the systemic interactions between institutional inertia, governance fragmentation, regulatory constraints, and social resistance [10,11]. This fragmentation of focus obscures the underlying mechanisms that determine why some cities succeed while others lag in implementing smart mobility.

Informed by the literature, the barriers explored in this study span three broad dimensions: (i) institutional and governance constraints, including fragmented mandates, outdated policies, and legacy planning paradigms; (ii) technological and infrastructural limitations, such as inadequate digital systems, data privacy concerns, and weak multimodal integration; and (iii) behavioral and socio-cultural challenges, including digital literacy gaps, affordability, and attachment to private car ownership [12,13,14].

A central but underexplored barrier lies in the legacy paradigms embedded in conventional transport planning (CTP). These paradigms, rooted in car-centric and rational-comprehensive models, prioritize traffic flow, vehicle throughput and saving time over flexibility, accessibility and multimodal integration [15,16]. Such paradigms are fundamentally misaligned with the adaptive, data-driven, and user-centered logics of smart mobility [17,18,19]. This institutional misfit is further compounded by fragmented mandates across transport agencies, outdated policies, gaps in digital literacy, and uneven infrastructure that create a web of resistance to mobility innovation [19,20]. Collectively, these barriers form a reinforcing system of structural inertia that hinders mobility transitions and undermines the promise of smart mobility in practice.

Moreover, beyond large metropolitan regions, smaller cities and peri-urban communities experience similar but often overlooked challenges, such as insufficient infrastructure, weak institutional coordination, and dependence on traditional mobility modes. These areas also face a shortage of digital and financial resources that make the implementation of smart mobility particularly complex [21,22].

This paper addresses these gaps by moving beyond descriptive or fragmented accounts to provide a systems-level and theory-informed analysis of barriers to smart mobility adoption. Drawing on Critical Urban Theory, the study interrogates the political, economic, and ideological forces shaping mobility transitions, thereby responding to recent calls for more socially attuned mobility research [23]. In doing so, it positions smart mobility adoption not as a neutral technological project, but as a politically contested process embedded within wider struggles over urban governance, equity, and sustainability.

The findings reveal that legacy planning paradigms, fragmented mandates, and regulatory misalignment function as root high-leverage barriers exerting a strong influence across the system. These structural factors shape downstream challenges such as affordability, spatial coverage, and cultural attachment to car ownership. This multi-level relationship underscores that resistance to smart mobility adoption emerges not from isolated technical deficits, but from an interconnected ecosystem of institutional, political, and socio-cultural constraints that define urban governance systems.

By integrating Critical Urban Theory with Total Interpretive Structural Modelling (TISM) and MICMAC analysis, this study contributes to the growing literature on urban mobility transitions in three ways. First, it advances a holistic framework that explains not only what barriers exist but also how they interact and reinforce one another. Second, it reframes adoption challenges as products of institutional and political dynamics, challenging the prevailing technocratic narratives of smart mobility. Third, it offers practical entry points for policymakers to target structural root causes rather than symptoms, offering guidance for cities especially in the Global South to build inclusive, scalable, and resilient smart mobility frameworks.

The remainder of the paper is structured as follows. Section 2 reviews the theoretical and empirical literature on smart mobility barriers. Section 3 outlines the research methodology. Section 4 presents the results of the TISM and MICMAC analyses. Section 5 discusses the implications of the findings, and Section 6 concludes with recommendations for policy, practice, and future research.

2. Theoretical Background: Smart Mobility Adoption Barriers

2.1. Smart Mobility and Urban Sustainability

Smart mobility represents a key dimension of contemporary urban sustainability, promising enhanced efficiency, lower emissions, and improved accessibility through the integration of digital technologies, data analytics, and platform-based transport services. Built on digital platforms, algorithmic optimization, and real-time data exchange, smart mobility solutions such as Mobility-as-a-Service (MaaS), micro-mobility, and autonomous vehicles are framed as transformative tools for sustainable urban transport [24,25].

However, beneath its techno-optimistic promise lie institutional, technological, and social contradictions that complicate adoption at scale. Scholars have increasingly cautioned against “technological solutionism,” where digital systems are treated as ends in themselves, disconnected from underlying political and structural dynamics [26,27,28]. These critiques suggest that the success of smart mobility depends not only on technological readiness but also on how power, access, and equity are configured within urban governance systems [29,30,31].

The transformative potential of smart mobility thus depends on how cities negotiate these multiple layers, namely, technological infrastructure, institutional capacity, governance structures, and citizen participation. Understanding the barriers that hinder this transformation requires situating them within broader socio-technical and political frameworks.

2.2. Barriers to Smart Mobility Adoption

Despite the increasing promotion of smart mobility as a sustainable and efficient alternative to traditional transport systems, its widespread adoption remains uneven across cities, especially in developing urban contexts. The literature has examined various barriers impeding smart mobility adoption, yet most studies tend to address these challenges in isolation. These barriers are summarized and discussed below:

• Interoperability and Integration Deficits

A lack of interoperability among transport modes and poor integration of payment systems, routing tools, and real-time information weakens the usability of smart mobility systems. Moreover, the absence of common standards and interoperability among different transport services undermines integration and seamless multimodal travel [31,32,33]. Empirical cases show that institutional fragmentation prevents integration of public transport with app-based services. Conversely, open data standards and public–private collaboration have enabled greater interoperability in cities [34,35].

• Inadequate Digital Infrastructure

The success of smart mobility is contingent on robust digital infrastructure, including high-speed internet, GPS coverage, and IoT-enabled systems [30]. It is equally important to have an elaborate infrastructure to collect, organize, access and control data. In cities or areas where these are lacking or unreliable, deployment is delayed or limited in scope [36]. Smart mobility is predicated on digital access, literacy, and engagement. Yet, significant segments of the population remain digitally excluded due to lack of smartphone access, limited digital literacy, or affordability issues. This digital divide has been found to disproportionately affect low-income, elderly, and marginalized urban residents, signaling that smart mobility could exacerbate mobility injustice rather than alleviate it [37,38,39]. However, targeted investments in digital infrastructure can act as enablers, particularly when aligned with national digital transformation agendas [40].

• Data Privacy and Security Concerns

Smart mobility systems rely on extensive data collection, processing, and sharing, which raises critical concerns about privacy, surveillance, and data misuse. Weak regulatory frameworks and opaque data practices undermine public trust and discourage adoption, particularly in contexts where governance oversight is limited [41,42,43,44,45]. The increasing involvement of private platform operators has also introduced new forms of data asymmetry, as user data are often monetized or repurposed without informed consent [28,46,47].

From a socio-technical perspective, the perception of surveillance and loss of control over personal information erodes user willingness to engage with digital mobility platforms [48,49]. Building trust thus requires transparent governance, ethical data stewardship, and regulatory frameworks that safeguard user rights while enabling innovation [50]. Evidence from European and Asian cities shows that strong data protection regimes and participatory data governance improve user confidence and platform legitimacy [51].

• Inclusive Design Deficiencies

Smart mobility systems may fail to meet the needs of persons with disabilities, older adults, and those with caregiving responsibilities [52,53,54]. Poorly designed interfaces or exclusionary algorithms can deepen inequality by prioritizing profitable routes or user segments [55]. For instance, ride-hailing algorithms may prioritize profitable zones over underserved areas, or price surges may penalize those with limited transport alternatives [56]. Digital and physical exclusion result whenever universal design principles are not integrated into mobility innovations. However, from European and Asian cities show that incorporating universal design principles early in development promotes equitable access and widens adoption [57,58,59].

• Fragmented Institutional Mandates

Urban mobility governance is typically fragmented, involving multiple agencies with overlapping or conflicting roles. This institutional complexity leads to duplication, weak accountability, and policy incoherence [60,61]. Fragmentation hinders smart mobility by stalling integrated planning and the deployment of multimodal systems. On the other hand, governance integration and inter-agency collaboration have been shown to enhance coordination and streamline innovation diffusion [20,62].

• Unsupportive Regulations and Policies

Outdated or misaligned transport regulations can obstruct smart mobility by failing to accommodate emerging models such as micromobility, shared services, or Mobility-as-a-Service (MaaS) [21]. Likewise, regulatory uncertainty deters private investment and experimentation [63]. Yet, enabling policies such as incentives, open data frameworks, and supportive licensing regimes have demonstrated success in accelerating adoption in cities like Helsinki and Singapore [64].

• Legacy Paradigms in Conventional Transport Planning (CTP)

Conventional transport planning (CTP), rooted in mid-20th-century car-oriented logics, is a major obstacle to smart mobility adoption. It emphasizes traffic flow, road capacity, and vehicle throughput, prioritizing private car use while marginalizing public, non-motorized, and shared modes of transport [65,66]. These practices reinforce spatial inequities and lock cities into auto-dependent development patterns, particularly in peripheral urban areas.

Institutional inertia and sunk investments in legacy infrastructure perpetuate these paradigms and constrain innovation [67]. CTP relies on predictive traffic models and outdated indicators such as congestion levels and travel times, which do not capture contemporary urban challenges like digitalization, changing travel behavior, and decarbonization goals [68,69,70]. Recent research advocates shifting toward accessibility- and equity-based performance measures that evaluate people’s ability to reach key opportunities rather than focusing on vehicle movement [71,72,73,74].

Notwithstanding, the persistence of CTP reflects a paradigmatic misalignment between legacy governance frameworks and emerging smart mobility models that prioritize adaptability, user-centric design, and integrated data-driven planning [18,22]. Unless planning institutions redefine priorities and adopt inclusive, accessibility-based metrics, smart mobility initiatives risk being constrained by outdated systems ill-suited for sustainable and digitally enabled urban transport transitions [75].

• Political Resistance

The politics of urban mobility typically reflects entrenched interests and ideologies. Resistance can stem from political actors reluctant to disrupt existing systems, fear of backlash from powerful car lobbies, or lack of political will to promote transformative policies [76]. However, political leadership and narrative framing can play a transformative role, particularly when smart mobility is positioned within broader goals of equity, climate action, and economic resilience [38].

• Digital Literacy Gaps

Digital literacy determines whether users can engage meaningfully with app-based mobility platforms. Gaps in digital competence especially among older adults and marginalized groups limit participation, even when technological infrastructure exists [77,78]. Certain approaches to address these gaps have been suggested namely user training, multilingual interfaces, and co-design approaches that reflect user diversity [79].

• Safety Concerns for Pedestrians and Cyclists

Safety remains a persistent challenge for active mobility modes. Poorly designed streets, inadequate lighting, and unsafe intersections discourage walking and cycling, especially among vulnerable populations [80]. Safety improvements not only enable greater uptake of active modes but also increase public confidence in multimodal travel options with empirical evidence showing significant improvement in uptake of non-motorized modes due to dedicated cycling infrastructure and improved lighting [59].

• Public Transport Appeal

Smart mobility is most effective when integrated with high-quality public transport. However, chronic issues of unreliability, overcrowding, and poor service perception deter users [81,82]. Integrating high-quality transit services with real-time data and digital ticketing can reposition public transport as the core of multimodal systems, as demonstrated in Singapore and Seoul [83].

• Limited Coverage of Smart Mobility Services

Smart mobility services often concentrate in central urban areas, neglecting peri-urban and low-income zones. This spatial bias perpetuates exclusion and unequal access to opportunities [83]. To bridge this gap, it has been suggested to expand coverage through geographic subsidies, demand-responsive transport and spatial equity policies [16,84].

• Affordability of Smart Mobility Options

Affordability remains a defining determinant of smart mobility adoption. While shared mobility promises cost efficiency, surge pricing and subscription-based models may exclude lower-income users [85]. Flexible fare structures, subsidies, and bundling services for options like MaaS schemes can make smart mobility more affordable [86].

• Car as a Status Symbol and Personal Space

Automobile ownership is deeply embedded in cultural and social norms, which to some degree symbolize success, freedom, and identity [87]. This deeply embedded symbolism poses psychological and social barriers to modal shift. However, generational changes, environmental awareness, and urban lifestyle trends are gradually challenging these norms. Policy measures that reframe car-free living as progressive and aspirational can accelerate behavioral transition [15].

2.3. Critical Urban Theory and Smart Urbanism

The rise of smart mobility cannot be fully understood through technological or functionalist lenses alone. Instead, its evolution and adoption must be situated within broader political, institutional, and ideological contexts [23]. Critical Urban Theory, rooted in neo-Marxist and post-structuralist traditions, provides an essential lens for interrogating how power, inequality, and ideology are embedded within the technological transformation of cities [88]. From this perspective, smart urbanism, and by extension, smart mobility, is not a neutral or universally beneficial innovation but a politically charged project that reflects and reproduces dominant interests [28,89].

At its core, critical urban theory challenges the instrumental rationality that underpins many smart city initiatives, including transport innovations. It critiques the “techno-solutionism” embedded in smart mobility discourse; the idea that digital technologies alone can resolve prevailing urban problems such as congestion, pollution, or inequity. Instead, critical theorists argue that such solutions often obscure underlying structural issues and depoliticize urban governance by shifting attention away from social justice, redistribution, and institutional accountability [90].

In the context of smart mobility, this critique is particularly salient. Technologies such as ride-hailing, Mobility-as-a-Service (MaaS), autonomous vehicles, and AI-based traffic systems have been introduced without sufficient attention to who benefits, who governs, and who is excluded. Therefore, smart mobility platforms frequently reproduce spatial, economic, and digital inequalities, privileging affluent users and well-served urban cores while marginalizing peripheral communities [37,46]. Furthermore, the privatization of data infrastructures and the influence of large tech firms in shaping urban transport agendas raise questions about democratic oversight, transparency and citizen agency.

More importantly, critical urban theory foregrounds the role of institutional path dependencies and systemic inertia, which are key concerns in this study. The adoption of smart mobility does not occur in a vacuum; it is filtered through existing urban governance systems, including Conventional Transport Planning (CTP) frameworks, which have long prioritized automobility, infrastructure expansion, and predictive forecasting models. These legacy systems can subtly constrain or redirect smart mobility in ways that reinforce established priorities and resist transformative change. As such, the conflict between CTP and smart mobility is not just technical or procedural, as it reflects a deeper ideological clash between rational comprehensive planning models and more emergent, data-driven, and user-centric logics [19].

Moreover, critical urban theory allows us to see CTP as a site of epistemic power, where certain forms of knowledge (e.g., traffic modeling, vehicle throughput metrics) are privileged over others (e.g., lived experience, community-based mobility practices and quality of life). This epistemic hierarchy contributes to the marginalization of alternative mobility visions, such as feminist, de-colonial, or sustainability-oriented transport paradigms [38].

Through the critical theory lens, this study interrogates not just the visible barriers to smart mobility adoption, such as digital access or institutional silos, but also the invisible logics, inherited paradigms, and institutional configurations that structure how urban mobility is imagined, governed, and enacted. This perspective helps uncover how legacy planning systems may constrain innovation not by design, but by habit, ideology, or governance inertia. By so doing, critical urban theory reframes smart mobility adoption as a contested and political process, shaped by entrenched institutional logics, uneven power relations, and conflicting urban imaginaries [23]. It offers a valuable framework for analyzing how emerging technologies interact with and are constrained by legacy systems like CTP, thereby enriching our understanding of why smart mobility adoption remains partial, uneven, and fraught with contradictions.

The next section presents the TISM methodology employed to reveal the causal layering among the 14 synthesized barriers as well as a MICMAC analysis to classify them by driving and dependence power.

3. Materials and Methods

This study adopts Total Interpretive Structural Modelling (TISM) to identify and analyze the key barriers that hinder smart mobility adoption in urban settings. Specifically, the research combines a mini literature review to establish a list of potential barriers and then examines their interrelationships using expert consultation. The role of experts was to aid in distilling the list of barriers, analyzing the correlation among barriers, and providing the interpretive logic [91]. This approach is particularly suitable due to the dual nature of the study objectives: (i) exploring the range of obstacles to smart mobility uptake and (ii) establishing structured relationships of how these barriers reinforce or interact with one another.

3.1. Research Approach

The research is situated within an interpretivist paradigm, which supports the co-construction of meaning through the subjective interpretation of expert insights and existing literature. Thus, an inductive approach was used that enabled the emergence of key themes from data and structured theory building. An inductive approach enabled the derivation of the list of barriers through literature synthesis and expert validation. Subsequently, relationships among these barriers were mapped and modeled using TISM.

3.2. The Choice of Total Interpretive Structural Modelling (TISM) in This Study

The core objective of this study is twofold: first, to identify the key barriers that hinder the adoption of smart mobility in urban contexts, and second, to analyze how these barriers interact to reinforce or impede adoption. This dual aim, which is both diagnostic and explanatory, requires a methodological approach that can reveal not only the existence of barriers but also their structural interdependencies and relative influence.

Total Interpretive Structural Modelling (TISM) was selected as the principal analytical technique because it is well-suited for theory-building in complex, underexplored domains. Smart mobility adoption, especially in developing urban contexts, is characterized by multilayered challenges spanning technological, institutional, and behavioral dimensions. While the field has seen growing interest in recent years, mass adoption has been hindered due to persistent barriers that limit system scalability [2]. There is a dearth of research that has systematically analysed the barriers using a structured methodology that moves beyond simple identification toward a layered interpretation of how they interconnect and evolve over time.

TISM is an advanced extension of the Interpretive Structural Modelling (ISM) method [92,93,94] enriched by the incorporation of interpretive logic. Unlike ISM that identifies binary relationships between factors, TISM advances this by requiring justifications for why and how specific relationships exist [91,95,96]. This interpretive depth enables researchers to build more meaningful, actionable models relevant for policy makers who need to understand systemic resistance in order to design effective interventions.

Following a literature review and expert consultations, this study identified a preliminary set of barriers operating across three levels: user-related (e.g., behavioral inertia, low digital literacy), technological (e.g., limited interoperability, inadequate infrastructure), and institutional (e.g., regulatory fragmentation, lack of policy coordination). TISM was then employed to model how these barriers interact, resulting in a structured hierarchy that distinguishes between driving barriers (those that initiate or intensify other challenges) and dependent barriers, (which manifest as downstream symptoms of broader systemic issues). This approach allows the study to address the core dimensions of theory development articulated by [97]. By so doing, the model not only identifies which barriers are most critical, but also explains how they reinforce or amplify one another and why they deserve priority attention in policy or planning.

Recent methodological enhancements in TISM, such as the integration of argumentation-based reasoning [98], were integrated in the interpretative logic to further strengthen the model’s validity. This extension allows the relationships between elements to be grounded not just in observation or consensus, but also in stakeholder justification. This is particularly relevant for smart mobility systems, which evolve and require different policy responses at different stages from initiation, scaling, through to stabilization. TISM’s flexibility makes it possible to identify which barriers are most salient at each of these stages, offering a dynamic and context-sensitive roadmap for intervention.

3.3. Justification for TISM over Alternative Methods

To ensure methodological rigor, TISM was evaluated against alternative multi-criteria decision-making and systems modeling techniques, including Interpretive Structural Modelling (ISM), Analytic Hierarchy Process (AHP), and DEMATEL. While ISM is widely used to identify and structure key elements within complex systems [99,100], providing clarity by establishing relationships among these elements [101], it does not allow for interpretive elaboration, fails to include transitive links, and hence limits its explanatory power [102]. On the other hand, AHP, though useful for prioritizing factors, is more suited to quantitative comparisons and lacks the ability to map causal relationships or generate hierarchical models based on qualitative inputs [103]. Similarly, Decision-Making Trial and Evaluation Laboratory (DEMATEL) is effective at identifying cause-and-effect relationships in complex problems but does not yield structured hierarchies of influence [104].

TISM overcomes the aforementioned limitations and uniquely integrates interpretation, hierarchy, and causal logic [91,105,106]. It captures transitive relationships, enhances theoretical development by explicitly addressing the “why,” and transforms complex systems into accessible, well-structured models [107]. Furthermore, TISM enables the creation of a coherent structural model that reveals the layered influence of various barriers, critical for understanding systemic resistance to smart mobility. The method’s ability to incorporate stakeholder reasoning and build consensus around interdependencies enhances its practical relevance for smart mobility.

4. Data Analysis and Results

4.1. Data Collection and Analysis

The data collection and analysis process for this study followed a structured three-phase approach. In the initial phase, a list of 31 potential barriers to the adoption of smart mobility was compiled. This list was informed by a review of relevant literature and expert consultations. These barriers represent a broad range of institutional, technological, and user-level challenges.

4.2. Identification of Barriers

This study began with a literature review to extract potential barriers to the adoption of smart mobility. Drawing from over 10 peer-reviewed articles, policy reports, and systematic reviews, a total of 31 barriers were identified (see

Table 1. Original List of Barriers to Smart Mobility Adoption obtained from literature review.

No.	Barrier	Source
1	Infrastructure for sustainable mobility	[32,108]
2	Openness to Innovation and new technologies in mobility	[32,109,110]
3	Platform security	[32,39]
4	Public transport attitude	[110,111,112]
5	Car-dependency reduction	[113]
6	Conventional Transport Planning (CTP)	[14,18,113]
7	App-based (on-demand) mobility services	[18]
8	Active mobility	[18]
9	Basic transportation access	[39,114]
10	Access to data and the internet	[39]
11	Digital and banking divide	[32]
12	Low digital literacy and privacy concerns	[39,115]
13	Perceived security	[115,116]
14	Conservative mindsets and customer acceptance	[112,116]
15	Data safety, ethics, and surveillance concerns	[11]
16	Regulatory and legal uncertainty (liability, insurance)	[112,116,117]
17	Urban-centric service coverage	[116]
18	Lobbyism and industry influence	[116]
19	Loss of car as personal space/status symbol	[110,116]
20	Public acceptance of shared mobility	[112]
21	Legislation, taxation, and funding gaps	[117]
22	Weak business models and market uncertainty	[117]
23	Lack of cooperation among stakeholders	[117]
24	Interoperability and multimodal coordination issues	[32,118]
25	Ticketing/payment system fragmentation	[118]
26	Weak integration with public transport in underserved areas	[119]
27	Tradition of private vehicle ownership	[108,110]
28	Limited platform appeal to older adults and conservative users	[108]
29	Fragmented vision, collaboration, and data-sharing	[108,120]
30	Financial/resource constraints	[120]
31	Unsafe walking/cycling infrastructure and vulnerable road user neglect	[121,122]

). The original list drew from various interdisciplinary sources across transport planning, urban studies, technology adoption, and public policy [2,8,9]. These barriers captured a wide range of institutional, technological, behavioral, and infrastructural constraints. A summary of the original barriers and their sources is presented in Table 1.

The list reflected a wide range of conceptual overlaps and varying levels of abstraction; thus, it was necessary to streamline the barriers to facilitate effective modeling and interpretation. To make the subsequent structural modelling tractable and meaningful, it was necessary to refine and consolidate the list. First, duplicate or overlapping barriers were merged. For instance, “access to data and internet,” “digital and banking divide,” and “lack of knowledge about smartphones and internet” were conceptually aggregated under broader categories such as digital literacy gaps or inadequate digital infrastructure [123]. Similarly, multiple entries relating to the socio-cultural role of car ownership (e.g., loss of car as status symbol, as personal space, tradition of private vehicles) were consolidated into a unified barrier representing the psychosocial attachment to private cars.

Following this conceptual distillation, a focus group discussion with the six domain experts drawn from Nairobi, Kenya’s commercial and transportation hub. The experts who comprised urban transport planners, policy analysts, and digital mobility practitioners was convened to further validate and streamline the list. This step was necessary to assess the face validity and contextual relevance [124,125]. Drawing on the Delphi method principles [126], the experts were organized into a focus group that engaged in iterative discussions to ensure consistency, contextual relevance, and non-redundancy of the identified barriers [127]. The experts assessed each item based on its systemic influence, relevance to the study objectives, and potential for policy intervention.

In parallel, a theoretical filtering process ensured that the retained barriers aligned with the aims of the study and were analytically suited for the Total Interpretive Structural Modeling (TISM) framework [91]. Barriers that were too broad to allow precise pairwise comparison or lacked sufficient interpretability (e.g., “smart leadership”) were deemed unsuitable for interpretive modeling. Only those barriers that were specific, actionable, and theoretically grounded yet broad enough to capture systemic dynamics were retained. The expert-driven reduction aligns with prior studies that have emphasized the importance of domain-specific filtering in interpretive modelling exercises [91,128]. Finally, the process narrowed the original list down to 14 barriers, which were judged to be the most salient, structurally relevant, and theoretically significant for the study context. These 14 barriers served as the input variables for the Total Interpretive Structural Modelling (TISM) and MICMAC analysis conducted in subsequent sections.

4.3. Expert Consultations

In the second phase, a panel of domain experts with over a decade of professional experience in transport planning and smart mobility was engaged to evaluate the relevance and significance of the identified barriers. Experts were selected through purposive sampling, a strategy well-suited for studies requiring rich, experience-based input. This approach was chosen because it allows flexibility and ensures that participants possess the contextual and technical knowledge necessary to evaluate complex socio-technical systems such as smart mobility [129,130].

In line with recommendations by [131], a pre-screening process was undertaken to ensure that only well-informed participants were included. All selected respondents had professional experience ranging from seven to thirty years across diverse roles. The panel comprised two academic experts in transport planning and smart urban systems, each with over 20 years of teaching and research experience. In addition, the panel drew from three practitioners from the transport sector (two from government agencies and one from a private mobility firm), each with more than twenty-five years of professional experience in transport policy, planning, or digital mobility services. This composition ensured a balanced representation of policy, practice, and research perspectives, consistent with methodological standards for Total Interpretive Structural Modelling (TISM). The final panel of five experts was considered sufficient given that smart mobility remains an emerging and specialized field. Prior studies in similarly nascent domains have successfully applied TISM using small, homogenous panels to capture interpretive depth and causal reasoning [132,133]. The list of experts is presented in

Table 2. List of Experts.

Expert	Sector	Role/Title	Experience Area
E1	Academia	Professor of Urban Transport	Smart mobility policy, urban governance
E2	Government	Senior Urban Planner (City Transport Dept.)	Multi-modal transport planning
E3	Industry	Head of Operations, Ride-hailing Platform	Platform-based mobility service delivery
E4	Civil Society	Director, Mobility and Access NGO	Inclusive design and transport justice
E5	Regulator	Officer, National ICT Authority	Data governance, privacy, and digital infrastructure

.

Data collection was carried out at mutually convenient times. Each expert received a standardized matrix presenting all 14 identified barriers for pairwise comparison. In each case, the row represented the potential cause and the column represented the effect. Respondents indicated whether a relationship existed between each pair (assigning “1” for a relationship and “0” otherwise) and provided brief rationales for their judgments to enrich interpretive depth.

To enhance the credibility and reliability of expert input, several bias control strategies were implemented:

• Anonymity and independent elicitation: Experts completed their evaluations independently to minimize social desirability or dominance bias that might arise in group settings.
• Triangulation across sectors: The inclusion of participants from academia, government, industry, civil society, and regulation ensured diversity of perspectives and mitigated institutional or disciplinary bias.
• Iterative clarification: Follow-up calls were conducted to verify ambiguous responses and ensure consistent interpretation of causal relationships.
• Consensus validation: The results were retained only when at least 80% consensus was achieved across expert responses, following best practices in TISM–MICMAC studies [134].

The data collection tool is presented in Appendix A.

4.4. Analysis of the Barriers

To ensure methodological rigor in determining consensus among the experts, the Gage Repeatability and Reproducibility (Gage R&R) technique was employed. This method evaluates the level of agreement among raters in qualitative assessments. Following the benchmark established by [135], a consensus score exceeding 80% was considered indicative of acceptable reproducibility and inter-rater reliability. Based on this criterion, 14 out of the initial 31 barriers met the inclusion threshold and were retained for further analysis using the Total Interpretive Structural Modelling (TISM) methodology. The initial reachability matrix was developed based on the cumulative input of five experts. A consensus rule was applied: if three or more respondents rated a particular relationship as “1,” the corresponding matrix entry was marked as “1”; otherwise, it was marked as “0.” This approach ensured both rigor and reliability in capturing expert agreement and laid the foundation for the subsequent TISM analysis.

4.5. Procedure for Conducting Total Interpretive Structural Modelling (TISM)

The Total Interpretive Structural Modelling (TISM) procedure applied in this study followed a structured and iterative set of steps designed to identify, analyze, and interpret the interrelationships among the most critical barriers to smart mobility adoption.

• Step 1: Identification of Key Barriers

The process began with an identification of barriers based on a mini literature review, expert consultations. A final list of 14 validated barriers was established for modeling (see Section 4.2).

• Step 2: Establishing Contextual Relationships

To investigate how the identified barriers interact, a contextual relationship was defined for each pair as follows: “Barrier i influences or enhances Barrier j.” These directional relationships were elicited through a pairwise comparison process, where experts assessed the extent to which one barrier affects another. For each pair (Bi, Bj), experts were asked to indicate their level of agreement with the proposed influence using one of three responses: (1) Agree, (2) Neutral, or (3) Disagree.

Given that 14 barriers were selected for analysis, this yielded a total of 182 pairwise comparisons (14 × 13). To determine whether a relationship exists between two barriers, a consensus threshold was applied: if at least 50% of expert responses indicated agreement, the relationship was considered valid and retained; otherwise, it was discarded as non-significant [96,102,132]. This approach ensured that only those directional links with collective expert support were incorporated into the structural model, thereby enhancing the reliability of the relationship mapping and enabling the identification of causal linkages across the system.

• Step 3: Construction of the Structural Self-Interaction Matrix (SSIM)

Expert input was used to assess the existence and direction of influence for every possible pair of the 14 barriers. The SSIM was developed based on four possible relationships and presented in

Table 3. SSIM.

Barriers	B14	B13	B12	B11	B10	B9	B8	B7	B6	B5	B4	B3	B2	B1
B1: Fragmented institutional mandates	O	V	V	V	O	V	A	V	V	V	O	O	V	X
B2: Lack of inter-operability and information on multimodal journeys and Integration	O	V	O	A	A	A	A	A	V	A	O	A	X
B3: Inadequate digital infrastructure	O	V	O	V	O	O	A	A	V	A	O	X
B4: Data privacy and security concerns	A	V	V	V	O	O	A	V	V	A	X
B5: Existing regulations and policies are not favorable for Smart Mobility	O	V	V	V	V	V	A	A	V	X
B6: Absence of inclusive design	O	V	O	V	A	A	O	A	X
B7: Political resistance	O	V	O	V	O	V	A	X
B8: Legacy paradigms in CTP	O	V	V	V	V	V	X
B9: Lack of appeal of public transport	O	V	V	O	O	X
B10: Lack of safe environment for pedestrians and cyclists	O	V	V	V	X
B11: Lack of coverage -smart mobility not extensively covered	O	V	V	X
B12: Loss of car as status symbol and personal space (e.g., flexibility).	O	A	X
B13: Affordability of smart mobility services	O	X
B14: Digital literacy gaps	X

Source: Authors’ own creation.

.

• V: Barrier i influences Barrier j.
• A: Barrier j influences Barrier i.
• X: Barriers i and j influence each other.
• O: No relationship exists between Barriers i and j.

The SSIM, reflecting expert judgments, is presented in Table 3.

• Step 4: Binary Interpretation of Relationships

Once the valid relationships were established, they were encoded into a binary matrix to construct the initial reachability matrix. This square “n × n” matrix (where n is the number of barriers) captures the direct influence of each barrier on another. Specifically, cell (i, j) represents the influence of Barrier i on Barrier j, and is coded as

• 1, if a direct influence exists (as agreed upon in Step 2);
• 0, if no influence is identified.

This binary representation simplifies the analysis of interrelationships and provides a foundation for subsequent hierarchical structuring. By default, all diagonal elements in the matrix (i.e., where i = j) are assigned a value of 1, indicating that each barrier is assumed to influence itself, in line with conventional practice [91,94,102]. The initial reachability matrix is presented in

Table 4. Initial Reachability Matrix.

Barriers	B1	B2	B3	B4	B5	B6	B7	B8	B9	B10	B11	B12	B13	B14
B1: Fragmented institutional mandates	1	1	0	0	1	1	1	0	1	0	1	1	1	0
B2: Lack of inter-operability and information on multimodal journeys and Integration	0	1	0	0	0	1	0	0	0	0	0	0	1	0
B3: Inadequate digital infrastructure	0	1	1	0	0	1	0	0	0	0	1	0	1	0
B4: Data privacy and security concerns	0	1	0	1	0	1	1	0	0	0	1	1	1	0
B5: Existing regulations and policies are not favorable for Smart Mobility	0	1	1	1	1	1	0	0	1	1	1	1	1	0
B6: Absence of inclusive design	0	0	0	0	0	1	0	0	0	0	1	0	1	0
B7: Political resistance	0	1	1	0	1	1	1	0	1	0	1	0	1	0
B8: Legacy paradigms in CTP	1	1	1	1	1	0	1	1	1	1	1	1	1	0
B9: Lack of appeal of public transport	0	1	0	0	0	1	0	0	1	0	0	1	1	0
B10: Lack of safe environment for pedestrians and cyclists	0	1	0	0	0	1	0	0	0	1	1	1	1	0
B11: Lack of coverage -smart mobility not extensively covered	0	1	0	0	0	0	0	0	0	0	1	1	1	0
B12: Loss of car as status symbol and personal space (e.g., flexibility).	0	0	0	0	0	0	0	0	0	0	0	1	0	0
B13: Affordability of smart mobility services	0	0	0	0	0	0	0	0	0	0	0	1	1	0
B14: Digital literacy gaps	0	0	0	1	0	0	0	0	0	0	0	0	0	1

Source: Authors’ own creation.

.

• Step 5: Development of the Interpretive Logic Knowledge Base

An interpretive knowledge base was created, documenting the logical explanations for each confirmed relationship. Each barrier was compared with all others in a pairwise manner. If a relationship was affirmed (“Yes”), a justification was documented based on expert judgment and literature support.

• Step 6: Reachability Matrix and Transitivity Check

To ensure logical consistency and completeness of the model, the principle of transitivity was applied. This rule posits that if Barrier A influences Barrier B, and Barrier B influences Barrier C, then Barrier A is assumed to indirectly influence Barrier C. All such inferred transitive relationships were systematically checked and incorporated into the matrix.

The resulting matrix thus reflects both the direct and transitive influences among the barriers, enabling the construction of a hierarchical structure in subsequent steps. The finalized reachability matrix, inclusive of transitive links, is presented in

Table 5. Final Reachability Matrix with Transitivities.

Barriers	B1	B2	B3	B4	B5	B6	B7	B8	B9	B10	B11	B12	B13	B14	Driving Power
B1: Fragmented institutional mandates	1	1	0	1	1	1	1	0	1	1	1	1	1	0	11
B2: Lack of inter-operability and information on multimodal journeys and Integration	0	1	0	0	0	1	0	0	0	0	1	1	1	0	5
B3: Inadequate digital infrastructure	0	1	1	0	0	1	0	0	0	0	1	1	1	0	6
B4: Data privacy and security concerns	0	1	1	1	1	1	1	0	1	1	1	1	1	0	11
B5: Existing regulations and policies are not favorable for Smart Mobility	0	1	1	1	1	1	1	0	1	1	1	1	1	0	11
B6: Absence of inclusive design	0	1	0	0	0	1	0	0	0	0	1	1	1	0	5
B7: Political resistance	0	1	1	1	1	1	1	0	1	1	1	1	1	0	11
B8: Legacy paradigms in CTP	1	1	1	1	1	1	1	1	1	1	1	1	1	0	13
B9: Lack of appeal of public transport	0	1	0	0	0	1	0	0	1	0	0	1	1	0	5
B10: Lack of safe environment for pedestrians and cyclists	0	1	0	0	0	1	0	0	0	1	1	1	1	0	6
B11: Lack of coverage -smart mobility not extensively covered	0	1	0	0	0	1	0	0	0	0	1	1	1	0	5
B12: Loss of car as status symbol and personal space (e.g., flexibility).	0	0	0	0	0	0	0	0	0	0	0	1	0	0	1
B13: Affordability of smart mobility services	0	0	0	0	0	0	0	0	0	0	0	1	1	0	2
B14: Digital literacy gaps	0	1	0	1	0	1	1	0	0	0	1	1	1	1	8
Dependency Power	2	12	5	6	5	12	6	1	6	6	11	14	13	1

.

• Step 7: Level Partitioning

Using the reachability matrix, level partitions were identified by determining the reachability and intersection sets for each barrier. Barriers with matching sets were placed at the same level in the hierarchy. This step was repeated iteratively until all barriers were assigned a hierarchical level. The level partitioning process is illustrated in

Table 6. Partition Level Matrix of the Final Reachability Matrix.

Iteration 1
Barrier	Reachability Set	Antecedent Set	Intersection Set	Level
B1	B1, B2, B4, B5, B6, B7, B9, B10, B11, B12, B13	B1, B8	B1
B2	B2, B6, B11, B12, B13	B1, B2, B3, B4, B5, B6, B8, B9, B10, B11, B14	B2, B6, B11
B3	B2, B3, B6, B11, B12, B13	B3, B4, B5, B7, B8	B3
B4	B2, B3, B4, B5, B6, B7, B9, B10, B11, B12, B13	B1, B4, B5, B7, B8, B14	B4, B5, B7
B5	B2, B3, B4, B5, B6, B7, B9, B10, B11, B12, B13	B1, B4, B5, B7, B8	B4, B5, B7
B6	B2, B6, B11, B12, B13	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B14	B2, B6, B11
B7	B2, B3, B4, B5, B6, B7, B9, B10, B11, B12, B13	B1, B4, B5, B7, B8, B14	B4, B5, B7
B8	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B12, B13	B8	B8
B9	B2, B6, B9, B12, B13	B1, B4, B5, B7, B8, B9	B9
B10	B2, B6, B10, B11, B12, B13	B1, B4, B5, B7, B8, B10	B10
B11	B2, B6, B11, B12, B13	B1, B2, B3, B4, B5, B6, B7, B8, B10, B11, B14	B2, B6, B11
B12	B12	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B12, B13, B14	B12	I
B13	B12, B13	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B13, B14	B13
B14	B2, B4, B6, B7, B11, B12, B13, B14	B14	B14
Iteration 2
Barrier	Reachability Set	Antecedent Set	Intersection Set
B1	B1, B2, B4, B5, B6, B7, B9, B10, B11, B13	B1, B8	B1
B2	B2, B6, B11, B13	B1, B2, B3, B4, B5, B6, B8, B9, B10, B11, B14	B2, B6, B11
B3	B2, B3, B6, B11, B13	B3, B4, B5, B7, B8	B3
B4	B2, B3, B4, B5, B6, B7, B9, B10, B11, B13	B1, B4, B5, B7, B8, B14	B4, B5, B7
B5	B2, B3, B4, B5, B6, B7, B9, B10, B11, B13	B1, B4, B5, B7, B8	B4, B5, B7
B6	B2, B6, B11, B13	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B14	B2, B6, B11
B7	B2, B3, B4, B5, B6, B7, B9, B10, B11, B13	B1, B4, B5, B7, B8, B14	B4, B5, B7
B8	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B13	B8	B8
B9	B2, B6, B9, B13	B1, B4, B5, B7, B8, B9	B9
B10	B2, B6, B10, B11, B13	B1, B4, B5, B7, B8, B10	B10
B11	B2, B6, B11, B13	B1, B2, B3, B4, B5, B6, B7, B8, B10, B11, B14	B2, B6, B11
B13	B13	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B13, B14	B13	II
B14	B2, B4, B6, B7, B11, B13, B14	B14	B14
Iteration 3
Barrier	Reachability Set	Antecedent Set	Intersection Set
B1	B1, B2, B4, B5, B6, B7, B9, B10, B11	B1, B8	B1
B2	B2, B6, B11	B1, B2, B3, B4, B5, B6, B8, B9, B10, B11, B14	B2, B6, B11	III
B3	B2, B3, B6, B11	B3, B4, B5, B7, B8	B3
B4	B2, B3, B4, B5, B6, B7, B9, B10, B11	B1, B4, B5, B7, B8, B14	B4, B5, B7
B5	B2, B3, B4, B5, B6, B7, B9, B10, B11	B1, B4, B5, B7, B8	B4, B5, B7
B6	B2, B6, B11	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11, B14	B2, B6, B11	III
B7	B2, B3, B4, B5, B6, B7, B9, B10, B11	B1, B4, B5, B7, B8, B14	B4, B5, B7
B8	B1, B2, B3, B4, B5, B6, B7, B8, B9, B10, B11	B8	B8
B9	B2, B6, B9	B1, B4, B5, B7, B8, B9	B9	III
B10	B2, B6, B10, B11	B1, B4, B5, B7, B8, B10	B10
B11	B2, B6, B11	B1, B2, B3, B4, B5, B6, B7, B8, B10, B11, B14	B2, B6, B11	III
B14	B2, B4, B6, B7, B11, B14	B14	B14
Iteration 4
Barrier	Reachability Set	Antecedent Set	Intersection Set
B1	B1, B4, B5, B7, B10	B1, B8	B1
B3	B3	B3, B4, B5, B7, B8	B3	IV
B4	B3, B4, B5, B7, B10	B1, B4, B5, B7, B8, B14	B4, B5, B7
B5	B3, B4, B5, B7, B10	B1, B4, B5, B7, B8	B4, B5, B7
B7	B3, B4, B5, B7, B10	B1, B4, B5, B7, B8, B14	B4, B5, B7
B8	B1, B3, B4, B5, B7, B8, B10	B8	B8
B10	B10	B1, B4, B5, B7, B8, B10	B10	IV
B14	B4, B7, B14	B14	B14
Iteration 5
Barrier	Reachability Set	Antecedent Set	Intersection Set
B1	B1, B4, B5, B7	B1, B8	B1
B4	B4, B5, B7	B1, B4, B5, B7, B8, B14	B4, B5, B7	V
B5	B4, B5, B7	B1, B4, B5, B7, B8	B4, B5, B7	V
B7	B4, B5, B7	B1, B4, B5, B7, B8, B14	B4, B5, B7	V
B8	B1, B4, B5, B7, B8	B8	B8
B14	B4, B7, B14	B14	B14
Iteration 6
Barrier	Reachability Set	Antecedent Set	Intersection Set
B1	B1	B1, B8	B1	VI
B8	B1, B8	B8	B8
B14	B14	B14	B14	VI
Iteration 7
Barrier	Reachability Set	Antecedent Set	Intersection Set
B8	B8	B8	B8	VII

.

• Step 8: Development of the Digraph

A directed graph (digraph) was constructed to visualize the hierarchical structure and relationships among the barriers. Arrows were used to denote directional influences between elements based on the reachability matrix. This graphical representation is shown in Figure 1.

The level partitioning process arranged the 14 barriers to smart mobility adoption into a seven-level hierarchy based on their driving and dependence relationships (see Table 6). At the base of the hierarchy, B8: Legacy paradigms in conventional transport planning (CTP) was positioned at Level VII, indicating it as the most foundational barrier influencing others but not directly influenced itself. Similarly, B1: Fragmented institutional mandates and B14: Digital literacy gaps occupy Level VI, highlighting their systemic influence across regulatory and user domains.

Mid-level barriers such as B5: Unfavorable regulations, B4: Data privacy concerns, and B7: Political resistance (Level V) act as both influencers and recipients of influence. These barriers typically represent policy-related and governance constraints that both shape and respond to institutional and technological challenges.

The other barriers such as B3: Inadequate digital infrastructure and B10: Lack of safe pedestrian and cyclist environments fall under Level IV, indicating their intermediate structural importance. These are technical and infrastructural issues that are influenced by high-level governance barriers but also shape user-level perceptions.

User-experience-focused barriers like B2: Lack of inter-operability, B6: Absence of inclusive design, B9: Lack of appeal of public transport, and B11: Limited coverage of smart mobility appear in Level III, demonstrating their dependence on upstream technical and regulatory conditions.

At the top of the model, B13: Affordability of smart mobility services (Level II) and B12: Loss of car as a status symbol and personal space (Level I) are positioned as the most dependent barriers. These represent end-user concerns that are largely shaped by upstream system, policy, and infrastructure factors.

• Step 9: Construction of the Final TISM Model

The final TISM model (Figure 1) was developed by translating the digraph into an interpretive structural framework. This model illustrates both direct and transitive relationships among the barriers and reflects the cumulative insights gathered through literature, expert interpretation, and TISM logic.

4.6. Interpretation of the Hierarchical Model

The TISM model (Figure 1) organizes the fourteen barriers across seven hierarchical levels, providing a structured view of causal dependencies.

• Level VII—Foundational Drivers:

(B8: Legacy paradigms in conventional transport planning) sits at the base of the hierarchy. It represents deep-seated institutional inertia and car-centric logics that shape all subsequent barriers. This finding aligns with [18,136], who argue that outdated planning paradigms remain the single most persistent obstacle to mobility innovation.

• Level VI—Structural Governance Barriers:

(B1: Fragmented institutional mandates and B14: Digital literacy gaps) form the next layer. These barriers mediate between institutional structures and user behavior. Fragmented mandates reflect policy incoherence across transport agencies, while literacy gaps shape how users engage with smart platforms. Together, they perpetuate governance fragmentation and inequitable access.

• Level V—Policy and Regulatory Barriers:

(B5: Unsupportive regulations, B4: Data privacy and security, B7: Political resistance) act as pivotal linkages connecting institutional design with implementation. These barriers embody the “politics of adoption” [63], where power asymmetries, outdated laws, and political inertia jointly constrain innovation diffusion.

• Level IV—Technical and Infrastructural Barriers:

(B3: Inadequate digital infrastructure and B10: Lack of safety for pedestrians and cyclists) occupy an intermediate position. These are shaped by upstream governance decisions yet directly influence user confidence and system accessibility. Their position underscores that technological deficits are not root causes but reflections of higher-level governance failures.

• Level III—Service and Design Barriers:

(B2: Interoperability deficits, B6: Absence of inclusive design, B9: Limited public transport appeal, B11: Limited coverage) cluster here, capturing user-facing challenges. They represent the translation of institutional constraints into service design and urban experience, reaffirming that user adoption is structurally mediated.

• Level II and I—User Perception and Behavioral Barriers:

(B13: Affordability constraints and B12: Car as a status symbol) represent the most dependent barriers. They reflect end-user perceptions and cultural norms, which are products—not causes—of the institutional, infrastructural, and policy environment. This hierarchical positioning reinforces Geels’ (2012) [137] multi-level perspective: behavior change at the micro-level depends on macro-structural transformation.

This layered structure illustrates that smart mobility adoption is inhibited less by technology itself and more by institutional, political, and cultural inertia embedded in planning and governance systems. The model thus supports a systemic interpretation consistent with Critical Urban Theory, which views technological adoption as socially embedded and power-mediated.

Further, based on the explanation of the experts, an interpretive logic was added to each relationship as presented in

Table 7. The Interpretive Matrix.

From Barrier	To Barrier	Interpretation of Influence	Supporting Literature
B8: Legacy paradigms in CTP	B1: Fragmented institutional mandates	Outdated planning practices in conventional transport (e.g., car-centric, siloed approaches) continue to shape institutional mandates, leading to fragmented roles among agencies. This weakens coordination and undermines integrated mobility planning.	[20,25]
B8	B14: Digital literacy gaps	Legacy systems often marginalize digital innovation and public ICT education, contributing to inadequate digital literacy among users and decision-makers.	[24]
B1: Fragmented institutional mandates	B4: Data privacy and security concerns	Fragmented institutional structures lack unified standards or oversight mechanisms for managing mobility data securely, increasing user concerns over privacy and data misuse.	[138]
B1	B5: Unfavorable regulations and policies	Disjointed mandates lead to inconsistent or outdated policy frameworks that do not support smart mobility innovations such as e-hailing, shared micromobility, or MaaS platforms.	[139,140]
B1	B7: Political resistance	Institutional fragmentation dilutes accountability and policy leadership, making it easier for entrenched political interests to resist disruptive mobility reforms.	[141]
B14: Digital literacy gaps	B4: Data privacy and security concerns	Users with limited digital literacy are more likely to be wary of data privacy issues, leading to reluctance in using app-based smart mobility services.	[142]
B14	B5: Unfavorable regulations and policies	Poor digital understanding among decision-makers hampers the development of responsive policies that address smart mobility governance and regulation.	[143]
B14	B7: Political resistance	Political leaders with low digital fluency may perceive smart mobility initiatives as risky or unmanageable, leading to resistance or policy inertia.	[20]
B4, B5, B7	B3: Inadequate digital infrastructure	Data privacy concerns (B4), outdated policies (B5), and political reluctance (B7) delay investment in and deployment of essential digital infrastructure for smart mobility (e.g., sensors, real-time data platforms).	[25]
B4, B5, B7	B10: Lack of safe environments for cyclists and pedestrians	Data privacy concerns (B4), outdated policies (B5), and political reluctance (B7) prevent urban redesign that accommodates active mobility and smart infrastructure like connected crossings or bike-tracking apps.	[113]
B3, B10	B2: Lack of interoperability and integration	Poor digital infrastructure (B3) limits the ability to interconnect transport modes, while unsafe streets (B10) reduce the feasibility of integrating active travel into mobility-as-a-service (MaaS) platforms.	[83]
B3, B10	B6: Absence of inclusive design	Limited infrastructure investment and inadequate safety planning lead to systems that do not account for the needs of vulnerable users (e.g., elderly, disabled and marginalised), undermining inclusivity.	[58]
B3, B10	B11: Limited coverage	Poor infrastructure and unsafe environments constrain the geographic and demographic reach of smart mobility services, particularly in peripheral or underserved areas.	[144]
B3, B10	B9: Lack of appeal of public transport	Poor infrastructure and safety concerns discourage the use of public transport and make it less attractive in comparison to private vehicles.	[143]
B2, B6, B9, B11	B13: Affordability of smart mobility services	Fragmentation, exclusivity, poor coverage, and unappealing alternatives reduce economies of scale, increase operational costs, and pass affordability burdens to users.	[145]
B13	B12: Loss of car as a status symbol and personal space	When smart mobility becomes affordable, accessible, and comprehensive, it begins to offer a viable alternative to private car ownership, challenging cultural norms and attachment to personal vehicles.	[22,87]

.

4.7. MICMAC Analysis

To complement the structural insights generated by the TISM model, a MICMAC analysis (Matrice d’Impacts Croisés Multiplication Appliquée à un Classement) was performed to assess the driving and dependence power of each identified barrier [105]. MICMAC provides a systematic means of mapping causal relationships between elements in a system, revealing how strongly each factor influences others (driving power) and how strongly it is influenced by them (dependence power). This dual perspective allows researchers to classify barriers based on their structural position and strategic importance within the overall system [146,147].

4.7.1. Analytical Process

The analysis proceeded in three main steps.

First, the Final Reachability Matrix (FRM) generated through TISM was used as input. The FRM captures the binary influence (1 = influence exists; 0 = no influence) between all pairs of barriers.

Second, the driving power of each barrier was calculated as the total number of barriers it influences (both directly and indirectly), while the dependence power represented the number of barriers influencing it.

Third, the driving and dependence scores were plotted on a two-dimensional graph to classify the barriers into four categories:

• Driving barriers—high driving, low dependence;
• Dependent barriers—low driving, high dependence;
• Linkage barriers—high driving, high dependence;
• Autonomous barriers—low driving, low dependence.

This structure forms the MICMAC map (Figure 2), which enables visualization of systemic leverage points and feedback dynamics across the network of barriers.

4.7.2. Results and Interpretation

The MICMAC results revealed a well-differentiated system with most barriers distributed between the driving and dependent quadrants. This suggests that institutional and policy-related constraints dominate the causal hierarchy, while user-level and infrastructural barriers occupy more reactive positions within the system. The classification of barriers is summarized below.

• Driving Barriers

These are barriers with high driving power but low dependence. These are foundational barriers that exert influence over many others and thus serve as strategic leverage points for intervention. In this study, six driving barriers fall in this category, namely: legacy paradigms in CTP (B8), fragmented institutional mandates (B1), unfavorable regulations and policies (B5), data privacy and security (B4), political resistance (B7) and digital literacy gaps (B14).

Among these, legacy paradigms in CTP (B8) emerged as the most influential root barrier. Consistent with [20,25], entrenched car-centric planning norms, rigid evaluative models, and path dependencies reinforce outdated infrastructure priorities, impeding the transition toward flexible and user-centric mobility systems. These institutional barriers, in turn, exacerbate fragmented mandates (B1) and regulatory misalignments (B5), as observed in studies of governance inertia in European and African cities [148,149].

Similarly, political resistance (B7) amplifies the institutional rigidity by limiting policy experimentation and regulatory adaptation, while data privacy concerns (B4) and digital literacy gaps (B14) represent critical socio-technical foundations upon which user trust and participation depend. Overall, the driving barriers reflect deeply embedded institutional and cognitive lock-ins that determine the trajectory of smart mobility reform.

• Dependent Barriers

Dependent barriers possess low driving power but high dependence, meaning they are largely symptomatic of upstream structural conditions. They include limited coverage of smart mobility (B11), lack of inter-operability and multimodal integration (B2), absence of inclusive design (B6), affordability constraints (B13) and loss of car as status symbols or personal space B12). These barriers manifest as the outcomes of broader systemic issues in governance, policy, and digital access. For example, limited service coverage (B11) and interoperability challenges (B2) stem directly from institutional fragmentation and weak regulatory frameworks, echoing findings by [34,64]. Similarly, the cultural attachment to the car (B12)—a persistent symbolic barrier—represents the final behavioral threshold in transitioning toward sustainable mobility, as also noted by [87,150]. The dependent barriers cannot be addressed effectively without first tackling foundational institutional and political impediments.

• Linkage Barriers

Linkage barriers exhibit both high driving and high dependence power, rendering them dynamically unstable and sensitive to changes elsewhere in the system [95,151]. Such elements often serve as both causes and consequences of systemic transformation, requiring iterative monitoring and adaptive management.

Interestingly, in this study, no barriers were identified within this quadrant. This absence suggests a relatively hierarchical and structured system, where causal relationships are more linear than recursive. This finding may also reflect the expert panel’s consensus, which emphasized institutional over feedback-driven dynamics in the context of emerging smart mobility systems.

• Autonomous Barriers

Autonomous barriers demonstrate low driving and low dependence power, implying a peripheral role in the systemic hierarchy. Although these factors do not exert major structural influence, they remain important for improving direct user experience and enhancing acceptance of smart mobility innovations. This category includes inadequate digital infrastructure (B3), lack of appeal of public transport (B9) and lack of safe environment for cyclists and pedestrians (B10). While these issues are operational rather than systemic, they play a critical role in shaping public perceptions. For instance, persistent safety risks discourage active modes, reinforcing car dependency [80], while poor-quality public transport undermines the attractiveness of multimodal travel [152]. Autonomous barriers should thus be integrated into long-term strategies that enhance user confidence and experience.

The MICMAC analysis enriches the findings from TISM by identifying high-leverage intervention points. Barriers such as legacy planning paradigms (B8), institutional fragmentation (B1), and regulatory resistance (B5) stand out as the most influential and should be addressed first to trigger cascading improvements across the system [141]. Meanwhile, dependent and autonomous barriers, although less influential, should be integrated into long-term strategies that build public trust, increase access, and encourage behavioral shifts toward smart mobility.

5. Discussion, Implications and Limitations

5.1. Discussion

This study set out to identify and map the interrelationships among key barriers to smart mobility adoption through an integrated TISM and MICMAC analysis. Specifically, it sought to (1) determine the structural hierarchy of barriers influencing adoption, (2) identify the most foundational and dependent barriers within the system and (3) reveal leverage points for strategic policy and governance interventions.

Through a combination of expert-based interpretive modeling and multi-criteria analysis, the study provides a systems-level understanding of why smart mobility, despite its technological promise, is unevenly adopted, particularly in emerging urban contexts.

The hierarchical structure generated through TISM demonstrates that barriers to smart mobility adoption are systemic, extending far beyond technical or behavioral challenges. Legacy paradigms in conventional transport planning (CTP) emerged as the most influential root barrier, reinforcing path-dependent investment patterns and institutional inertia. This finding resonates with [20,153], who argue that traditional planning logics, centered on road expansion, congestion reduction, and predictive modeling have entrenched car-dependence while marginalizing multimodal innovation. The persistence of these paradigms reveals a critical disjuncture between smart mobility’s user-centered ethos and the bureaucratic rationalities that continue to guide urban transport policy. As [25,154] observe, transforming these paradigms requires reframing evaluation metrics away from traffic flow and toward accessibility, equity, and environmental performance.

Closely linked to these structural constraints are fragmented institutional mandates and unsupportive regulatory frameworks, which the MICMAC analysis classified as high-driving barriers. These barriers create a fragmented governance landscape that obstructs inter-agency coordination and discourages private-sector innovation. The findings align with recent studies on Mobility-as-a-Service (MaaS) and shared mobility ecosystems, which highlight how regulatory uncertainty and overlapping mandates can stall platform integration and scalability [155,156,157]. In such fragmented settings, policy experimentation tends to occur in isolation, resulting in pilot projects that fail to scale, and in some cases, exacerbate inequalities in access to smart mobility services or even its adoption. For instance, in Johannesburg, publicly funded and subsidized transport systems such as Gautrain, Metrorail, Rea Vaya, and Putco operate under separate governance and fare structures with minimal interoperability, illustrating how institutional fragmentation undermines multimodal cohesion and weakens the broader smart mobility transition [158,159].

At the mid-levels of the structural hierarchy of the barriers, the analysis identified data privacy, political resistance, and inadequate digital infrastructure as barriers that are both influential and dependent, positioning them as pivotal points of systemic tension. These results echo [8,160], who found that data insecurity and weak governance frameworks undermine public trust in digital mobility platforms, constraining adoption even in technologically advanced cities. In developing urban contexts, where institutional trust and enforcement capacity are limited, these challenges are amplified. The interaction between data governance and political influence observed in this study reflects what [161] terms “algorithmic governance”, where private platforms assume quasi-regulatory roles in the absence of robust state oversight, shifting control of urban mobility from public accountability to market logics.

Toward the upper levels of the hierarchy, user-centered barriers including affordability, inclusivity, and the enduring cultural association of the car with social status emerged as the most dependent. These findings suggest that user adoption outcomes are downstream manifestations of structural and institutional conditions rather than isolated individual preferences. Studies of car-sharing systems (e.g., [162,163,164]) have shown that affordability, convenience, and perceived social prestige decisively shape modal choices. The present analysis extends these insights by demonstrating that such behavioral dynamics are structurally conditioned. Affordability challenges, for instance, are not simply a matter of pricing but stem from limited regulatory incentives and fragmented subsidy frameworks that fail to make shared mobility financially viable for low-income users. This pattern has also been observed in cities such as São Paulo, and Delhi, where weak institutional support and inconsistent fare-integration policies constrain equitable access to shared and on-demand mobility services [165,166,167]. Likewise, the persistence of automobility as a status symbol reflects decades of car-oriented planning and cultural reinforcement—what [168] describes as the “emotive infrastructure” of automobility.

Interestingly, the MICMAC results show an absence of “linkage barriers”, the factors with both high driving and high dependence power. This suggests that systemic and user-level barriers are connected primarily through hierarchical, not feedback, relationships. The structural gap highlights the lack of adaptive mechanisms in smart mobility ecosystems, where user experiences rarely inform policy redesign or platform innovation. The absence of such feedback loops may explain why many smart mobility projects remain technically sound but socially fragile. As ref. [169] argued, effective smart governance depends on multi-level integration—where citizen input, institutional capacity, and technological systems co-evolve. The findings from this study thus reinforce the argument that governance innovation is as critical as technological innovation in achieving sustainable adoption of smart mobility.

Synthesizing the insights from the foregoing results, this study bridges the divide between systemic-institutional and user-behavioral perspectives in smart mobility research. It empirically supports the contention of [170,171] that smart mobility must be conceived as a socio-technical assemblage, in which governance frameworks, infrastructural systems, and social norms co-produce the trajectory of urban transformation. From this perspective, the barriers identified, particularly those related to planning paradigms, institutional coordination, and regulatory adaptation represent leverage points for reconfiguring both the governance and lived experience of urban mobility. Addressing them would not only unlock technological efficiency but also align smart mobility with broader goals of social equity, climate action, and urban resilience.

Overall, integrating Critical Urban Theory with structural modelling approaches (TISM–MICMAC) provides a realistic pathway for unpacking the complex, multi-level barriers to smart mobility adoption. Whereas previous studies have tended to examine institutional, technological, and user-related challenges in isolation, this paper demonstrates their deep interdependence and identifies the institutional and socio-political roots of technical and behavioral constraints. By revealing how governance structures, legacy planning paradigms, and cultural norms interact to shape adoption outcomes, the study advances a systems-oriented and theory-informed understanding of urban mobility transitions. This integrated approach is novel in that it bridges the gap between conceptual and empirical scholarship while offering policymakers a structured roadmap for prioritizing high-leverage interventions capable of catalyzing sustainable, inclusive, and resilient smart mobility systems.

5.2. Implications for Theory and Policy

This study contributes to both theoretical advancement and practical understanding of smart mobility adoption by unpacking how institutional, political, and socio-technical barriers interact to shape urban mobility transitions. The integration of TISM and MICMAC analysis advances existing frameworks by revealing not just what the key barriers are, but how they hierarchically reinforce or constrain one another within complex urban systems.

From a theoretical perspective, the study makes three main contributions.

First, it extends Critical Urban Theory (CUT) by empirically demonstrating how entrenched transport planning paradigms and fragmented governance structures act as root constraints that mediate technology adoption. While much of the literature conceptualizes smart mobility barriers as discrete challenges, namely technological, behavioral, or infrastructural, this study shows that these barriers are systemically coupled through institutional inertia and policy incoherence. In doing so, it positions smart mobility not merely as a technological innovation but as a contested urban transformation process, structured by historical path dependencies and governance asymmetries.

Second, by employing TISM within a critical theoretical frame, this study deepens socio-technical transition theory. Existing frameworks such as the Multi-Level Perspective (MLP) emphasize regime resistance but often under-specify causal hierarchies within the regime itself. The present analysis fills that gap by identifying structural leverage points, notably legacy planning paradigms, regulatory misalignment, and institutional fragmentation, that determine whether smart mobility transitions move beyond experimental enclaves into mainstream urban systems. This theoretical refinement enhances our understanding of how systemic change can be sequenced in complex urban environments.

Third, the study’s hierarchical model contributes to the emerging discourse on data governance and digital trust in smart cities. The positioning of data privacy, digital literacy, and inclusivity barriers within the mid to upper levels of the hierarchy highlights how user trust and equity outcomes are contingent upon upstream institutional reforms. This theoretical linkage bridges the persistent divide between system-level governance and user-level experience, offering a multidimensional understanding of technology adoption in urban mobility.

From a practical perspective, the findings offer actionable insights for policymakers, urban planners, and mobility service providers.

At the governance level, the identification of legacy transport paradigms and institutional fragmentation as foundational barriers underscores the urgency of regulatory and institutional reform. Cities seeking to scale smart mobility must prioritize cross-sector coordination, develop interoperable data standards, and embed mobility integration mandates into statutory planning frameworks. Practical examples including integrated planning of residential with transport and smart mobility solutions might help overcome the barriers in the future. Without such reforms, digital mobility innovations risk being absorbed into outdated, car-centric systems rather than transforming them.

At the operational level, the findings suggest that smart mobility implementation must be sequenced strategically. High-driving barriers such as governance reform, regulatory coherence, and digital infrastructure should be addressed first, as improvements here create enabling conditions for downstream adoption. Only after these systemic enablers are in place can policies targeting affordability, inclusivity, and user engagement achieve sustained impact.

Furthermore, the persistent role of car ownership as a cultural status symbol implies that behavioral interventions must go beyond information campaigns. Instead, policymakers should link pricing, branding, and infrastructure policies to reshape mobility culture, e.g., through incentives for shared mobility, redesign of public space for active modes, and visibility of sustainable transport as a premium urban lifestyle.

For practitioners and mobility platform providers, the results reinforce the importance of inclusive and ethical design. Digital mobility services must account for the realities of low digital literacy and limited smartphone penetration in many developing contexts. Practical steps include simplified user interfaces and multi-channel access models. For instance, Kenya’s Swvl and Little Shuttle initially integrated USSD and call-based booking options alongside mobile apps to reach users without smartphones. Similarly, co-designing solutions with marginalized communities, as seen in Cape Town’s GoMetro initiative, can ensure that service planning reflects users’ lived mobility needs. Transparent data governance frameworks such as those adopted by the EU’s Mobility Data Space or India’s National Data Sharing and Accessibility Policy can also help build public trust and promote responsible data sharing. By embedding such inclusive and context-aware design practices, mobility providers can enhance adoption, equity, and the long-term legitimacy of smart mobility ecosystems.

5.3. Study Limitations and Future Research Directions

While this study advances understanding of the structural barriers to smart mobility adoption, several limitations provide valuable entry points for future research.

First, the scope and composition of the expert panel, though methodologically rigorous, necessarily limited the diversity of perspectives. The purposive sampling approach ensured that all participants possessed deep domain expertise in transport planning, governance, and digital mobility systems. However, it did not directly capture the lived experiences of end-users such as commuters, cyclists, and low-income residents—groups most affected by urban mobility transitions. Future studies could therefore integrate multi-stakeholder perspectives, combining expert elicitation with participatory approaches such as focus groups, citizen juries, or community-based mapping. This would enrich the interpretive depth of the Total Interpretive Structural Modelling (TISM) framework and strengthen the connection between system-level analysis and user-centered realities.

Second, the geographical and contextual focus of the study is anchored in the Global South urban context. While offering novel insights into institutional and governance dynamics, the choice of context may constrain generalizability to other urban regions. Mobility systems in rapidly urbanizing cities often exhibit institutional fluidity, informal transport networks, and infrastructural gaps not typically found in mature economies. Future comparative studies across different urban typologies could test the transferability and robustness of the structural model, identifying which barriers are context-specific and which reflect universal governance constraints in smart mobility adoption.

Third, the study’s reliance on qualitative expert judgment introduces an element of subjectivity inherent in interpretive modeling approaches. Although consensus measures such as the Gage R&R test were used to ensure inter-rater reliability, expert biases rooted in institutional affiliation or disciplinary orientation may still influence the structure of relationships derived. Future research could integrate quantitative validation techniques, such as Structural Equation Modelling (SEM), Bayesian network analysis, or hybrid multi-criteria decision-making frameworks, to test and refine the causal pathways identified through TISM-MICMAC.

Fourth, while the hierarchical model captures the static relationships among barriers, it does not fully reflect the temporal or adaptive dynamics of smart mobility transitions. Urban transport systems evolve through feedback loops, political shifts, and technological disruptions that reshape the relevance and strength of specific barriers over time. Longitudinal or scenario-based modeling could complement this approach, exploring how interventions at high-driving levels such as regulatory reform or institutional realignment cascade through the system in practice.

Finally, although this study was grounded in Critical Urban Theory (CUT), future research could extend the theoretical lens to include complementary perspectives such as transition management, institutional work theory, or co-evolutionary governance frameworks. Such approaches could further illuminate how actors strategically navigate, reproduce, or resist change within entrenched urban regimes, thereby deepening our understanding of the political economy of smart mobility adoption.

6. Conclusions

This study set out to identify and structurally map the interrelationships among barriers to smart mobility adoption using TISM and MICMAC analysis, guided by the lens of Critical Urban Theory. The results show that adoption challenges are systemic rather than isolated, shaped by institutional inertia, fragmented mandates, outdated planning paradigms, and socio-political resistance. Technological and user-level barriers—such as affordability, safety, and digital literacy—are downstream manifestations of these deeper structural issues.

A key contribution of this study lies in demonstrating that legacy transport planning paradigms, institutional fragmentation, and regulatory misalignment function as root barriers with the strongest driving power. In contrast, user-related constraints such as car-dependence, service affordability, and safety perceptions emerge as dependent barriers that can only be resolved when structural reforms are enacted. This systems perspective underscores that achieving sustainable and inclusive smart mobility transitions requires addressing governance and institutional bottlenecks before expecting large-scale behavioral change.

From a policy standpoint, the findings offer several actionable insights. First, governments should embed digital literacy and inclusion programs into transport policy to ensure that marginalized populations are not excluded from emerging smart mobility systems. Second, cross-agency governance frameworks and joint planning units should be established to reduce fragmentation between transport, ICT, and urban development institutions. Third, cities should introduce equity-based performance indicators—such as accessibility gains for low-income users or gender-sensitive safety measures—to monitor progress beyond traditional metrics like congestion reduction. Finally, data governance standards must be strengthened to ensure transparency and user trust in digital platforms.

Looking ahead, these barriers are not static but will evolve alongside emerging technologies. The expansion of mobility-as-a-service (MaaS) platforms, autonomous vehicles, and AI-driven transport management will likely reshape the institutional and ethical landscape of mobility governance. New regulatory and equity challenges may arise as data ownership, algorithmic bias, and automation transform traditional models of mobility provision. Consequently, future research should track how these evolving technologies intersect with political and social dynamics, shaping both new opportunities and new forms of exclusion in smart mobility ecosystems.

In conclusion, this study contributes to the growing body of work that views smart mobility not merely as a technological innovation but as a socio-political transformation project. By identifying root causes and mapping their interdependencies, it provides a foundation for policymakers, planners, and researchers to design interventions that are structurally grounded, socially inclusive, and future-ready.