Determinants of smart factory adoption in emerging economies: empirical evidence from the automotive industry in Morocco and Nigeria

In this era of rapid technological advancement, smart factory adoption has gained significant attention in various industries around the world. Smart factory technology as described by Sjödin, Parida, Leksell, & Petrovic (2018), is the emergence of a connected and adaptable manufacturing system that effectively utilizes a continuous stream of data from interconnected operations and production systems to learn and swiftly adapt to evolving demands. The advent of these smart factory technologies has the potential to trigger an industrial revolution and therefore disrupt established companies that are not swift in adopting these emerging technologies (Oztemel & Gursev, 2020). Proactive manufacturers are embracing and adeptly adapting these technologies to maintain a competitive edge (Sjödin et al., 2018).

The success of organizations is being driven by the advent of smart factories, representing a new dawn where technology assumes the role of a primary catalyst for achievement (Grabowska, 2020). By leveraging the latent potentials of the Internet of Things (IoT) (Rezaei, Shirazi, & Karimi, 2017), machine learning (Hofmann & Rüsch, 2017), predictive analytics and advanced robotics (Soori, Arezoo, & Dastres, 2023), organizations can unlock an unprecedented level of efficiency, enhance quality control, reduce costs and achieve unprecedented agility (Sousa, Cruz, Rocha, & Sousa, 2019). As this revolution continues to reshape the global industrial landscape, manufacturing enterprises, especially those in the automotive industry, are facing increasing pressure to embrace smart factory technology to enhance productivity, efficiency and competitiveness.

The smart factory adoption has been historically associated with developed countries, namely Germany, the United States, France, Japan and China (Ojubanire, Sebti, & Berbain, 2022). However, developing countries need to catch up with this fourth revolution and overcome technological, social, legal and economic challenges (Tortorella & Fettermann, 2018). Morocco, which is considered a privileged destination for investors in aeronautic and automotive industries, is trying to take advantage of the dynamics of its industry and is also keen on the smart factory concept (International Monetary Fund [IMF], 2023; Oxford Business Group, 2020).

Nigeria, which is the largest and one of the fastest-growing economies in Africa, aims to diversify its economy into other sectors, including the automotive industry and thus is willing to invest in the industry's digital transformation (Abioye et al., 2018). Moreover, research on smart factory in Morocco and Nigeria is in its embryonic stage. Therefore, this study explores the determinants of smart factory adoption in the context of the automotive industry in Morocco and Nigeria. It aims to provide an enlightening investigation of smart factory technology, current level of adoption and the critical determinants of successful adoption in African emerging economies using Morocco and Nigeria as case studies.

This study extends the technology, organization, environment (TOE) framework beyond its conventional application in developed economies (Europe, North America) (Baker, Gaspard, & Zhu, 2022; García-Moreno, 2025) and Asian contexts (Nguyen, Le, & Vu, 2022; Silva et al., 2023) by examining smart factory adoption in the African automotive sector, where institutional environments, infrastructure constraints and digital maturity levels differ significantly from previously studied regions. While existing studies have explored Industry 4.0 (I4.0) adoption in Latin America (Brixner et al., 2020; Tortorella & Fettermann, 2018; Treviño-Elizondo & García-Reyes, 2020) and Asia (Prause, 2019; Sung, 2018), the African context remains underexplored despite its unique challenges of limited technological infrastructure, nascent innovation ecosystems, and distinct regulatory frameworks. Our contribution is threefold: (1) we provide empirical evidence on smart factory adoption patterns in two African countries with contrasting industrial development trajectories, (2) we identify context-specific determinants that may not be captured in studies from other emerging economies and (3) we demonstrate how the TOE framework requires adaptation when applied to economies with lower digital maturity and less developed innovation ecosystems.

The rest of the article is structured as follows. Section two presents a literature review on smart factories, the automotive industry and the theoretical positioning of the study. Section three explains the adopted methodology and the validity/reliability of our study. Section four highlights the main results, comparative analysis of results from the two use-case countries, and a test of the hypothesis of the proposed framework. The last section presents the study conclusions and research implications.

The smart factory, considered one of the key elements of I4.0, addresses a digital transformation period in the manufacturing landscape, requiring the integration of digital technologies into production systems (Kagermann, Wahlster, & Helbig, 2013). The I4.0 concept depends on the use of technologies that are more communicative, open, astute and organized than those utilized in conventional manufacturing (Ojubanire et al., 2022).

The smart factory, as an evolving concept, has garnered considerable scholarly attention, with diverse definitions offered by authors (e.g. Lucke, Constantinescu, & Westkamper, 2008; Osterrieder, Budde, & Friedli, 2020). In this study, a smart factory is characterized as an automated manufacturing system capable of collecting and sharing data across the manufacturing system while autonomously making intelligent manufacturing-related decisions. That is, a smart factory is “a hi-tech manufacturing system, generating, aggregating, analyzing, storing and sharing data among the production units and components, and able to independently make intelligent decisions based on available data and initiate actions within the production system.” While automation is a familiar concept in industrial and manufacturing enterprises, the novelty of smart manufacturing lies in system autonomy through digitalization across organizational dimensions – vertical and horizontal. What is new is “(1) real-time data collection (sensors), (2) connectivity and real-time data exchanges (internet), (3) adaptive decision and control of the physical system” (powerful computers and use of IA, operations research algorithms, etc.) (Ojubanire et al., 2022).

At the core of a smart factory is the development, application and distribution of I4.0 technologies for operational efficiency in manufacturing enterprises and industrial plants (Vrchota, Volek, & Novotna, 2019). The rapid evolution of smart factories has significantly impacted the global value chain and ultimately the global economy. The effectiveness of a smart factory is, however, dependent on the appropriate integration of human and material resources – objects, technologies and systems, as an autonomous and fully optimized value-adding structure (Benotsmane, Kovacs, & Dudas, 2019).

Furthermore, Lu, Morris, and Frechette (2016) assert that I4.0 innovations leverage ICT and intelligent technologies to provide a comprehensive view of the production environment, enabling data-driven decision-making and responsiveness to market changes. These facilities retrieve data from various resources, enhancing production, maintenance, stock monitoring and functional digitization (Shi et al., 2020). Smart factory technologies, as highlighted by Stock and Seliger (2016), improve efficiency, agility and user-centrism in operations and production. This involves internet-connected smart products, services and the collection/analysis of data through smart applications (Benotsmane et al., 2019).

The automotive industry, a significant global source of employment, holds paramount importance due to its intricate value chain and extensive linkages within both domestic and international economies. Manufacturing new vehicles relies heavily on a sophisticated and sequential production system characterized by refined processes (Howell & Hsu, 2002). Recognized as a capital-intensive and knowledge-driven sector, the automotive industry plays a pivotal role in fostering economic prosperity and generating employment across various sectors and nations (Saberi, 2018). This industry serves as the backbone for both developed and developing economies. In developed economies, it constitutes a vital segment of industrial manufacturing, with virtually every major economy hosting a substantial automotive presence (Saberi, 2018).

Amidst the rising influence and complexity of factors affecting the automotive industry (Wells, Wang, Wang, Liu, & Orsato, 2020), smart manufacturing technologies introduce new business potential, necessitating organizational adaptation (Llopis-Albert, Rubio, & Valero, 2021). In response, the automotive sector globally is investing in enhancing the skills of its workforce through collaborations involving industry associations, vocational and research institutions, and government agencies (Redman, Friman, Garling, & Hartig, 2013). The industry faces a crucial need to focus on attracting, retaining and providing comprehensive training, reskilling and upskilling opportunities for its workforce as it undergoes significant digital transformation (Opazo-Basáez, Vendrell-Herrero, & Bustinza, 2022).

The integration of novel programming and hardware technologies in vehicles has expanded their functionality and complexity, with connected vehicles altering business strategies toward delivering a targeted customer-centric experience (Llopis-Albert et al., 2021). Looking ahead, the future of automotive manufacturing will witness significant changes, incorporating autonomous driving systems, satellite navigation and increased utilization of vehicle dynamic control systems. The automotive industry, therefore, stands as one of the fastest-growing sectors for IoT development (Rahim et al., 2021), emphasizing its ongoing revolution and adaptation to emerging technologies for smart production systems.

This study builds on the TOE Framework (Tornatzky & Fleischer, 1990) in order to study the determinants of smart factory adoption in the automotive industry. The TOE Framework is one of the most reliable frameworks to study technology adoption from an organizational perspective (Nguyen et al., 2022). The framework is composed of three main constructs, namely technology, organization and environment. The technology construct is related to the technological levers that could have an impact on smart factory adoption. The first two sub-constructs of this construct, which are relative advantage and complexity, are adopted from Diffusion of Innovation (DOI) theory (Rogers, 2003). Relative advantage describes the degree to which the smart factory concept is considered better than traditional or conventional manufacturing, while complexity is linked to the difficulty of understanding or using the smart factory concept (Rogers, 2003).

Although the TOE framework provides a robust lens for technology adoption, its application to African emerging-economy manufacturing requires explicit contextualization. In many African automotive contexts, firms face uneven digital maturity, infrastructure constraints and thinner innovation ecosystems, which affect the feasibility and sequencing of I4.0 investments. As a result, adoption decisions may be shaped less by “perceived benefits” alone and more by whether minimum feasibility conditions and ecosystem enabling resources exist to reduce implementation uncertainty (Ojubanire et al., 2025, 2026). TOE is appropriate because it explains adoption as a joint outcome of (1) technology-related attributes and feasibility conditions, (2) organizational capabilities and internal commitment, and (3) environmental pressures and support structures.

More precisely, this study constitutes a contextual refinement of the TOE framework rather than a wholesale extension or structural reconceptualization (Amin et al., 2024; Satyro et al., 2024). The original TOE constructs, technology, organization and environment, are retained in their entirety; what changes is the relative salience and the underlying mechanisms of each factor when the framework is transposed to a context of lower digital maturity and weaker institutional support (Bhuiyan, 2024). Specifically, in constrained environments, the Technology construct operates primarily through feasibility conditions (readiness, compatibility, cybersecurity baseline) rather than through perceived advantage alone (Amin et al., 2024; Bhuiyan, 2024); the organization construct is shaped by capability gaps and change management constraints rather than purely by managerial volition (Hussain Shahadat, Nekmahmud, Ebrahimi, & Fekete-Farkas, 2023; Ojubanire et al., 2025) and the environment construct is amplified by ecosystem enablement and regulatory scaffolding that substitute for internal organizational resources (Hussain Shahadat et al., 2023; Nguyen et al., 2022). These contextual shifts in factor mechanisms distinguish the present contribution from earlier TOE applications in developed or more mature emerging economies (Baker et al., 2022; Silva et al., 2023).

Further, in emerging-economy automotive contexts (such as Morocco and Nigeria), the influence of TOE factors is shaped by contextual conditions such as uneven digital infrastructure, resource constraints, skills gaps and less mature innovation ecosystems. These conditions can amplify the role of feasibility and enablement factors (e.g. technological readiness, ecosystem support, regulatory support) and can weaken the extent to which perceived benefits alone translate directly into adoption decisions. Consistent with Figure 1, the study therefore models smart factory adoption as influenced by technology, organization and environment determinants that reflect both adoption attractiveness and adoption feasibility.

The technology readiness sub-construct describes the capacity to pave the way for using state-of-the-art information technologies and solutions, namely IoT, big data and additive manufacturing, among others (Moldabekova, Philipp, Satybaldin, & Prause, 2021). The fourth technological element or sub-construct is related to adopting cybersecurity and personal data protection standards (Culot, Fattori, Podrecca, & Sartor, 2019). Organizational construct has three major sub-constructs – management support, organizational readiness and level of integration. Studies have shown that smart factory adoption could be influenced by the support of top leaders through visionary leadership, support and sponsorship of digitalization initiatives (Schumacher, Erol, & Sihn, 2016). Similarly, organizational competence, which, on one side relates to the organization's process management and on the other side employees' skills for mastering and interacting with complex solutions and machines (Shevyakova, Munsh, Arystan, & Petrenko, 2020), is a driver for technology adoption. Lastly, the integration of digital equipment into a single system and the level of the company's integration with partners across the value chain through systems' interoperability constitute a major determinant for smart factory adoption (Senna, Ferreira, Barros, Bonnin Roca, & Magalhaes, 2022).

Lastly, the environmental construct encompasses environmental pressure, ecosystem support and Government & Regulatory support. The environmental pressure can be either from company's stakeholders for achieving performance and sustainability, or from competition, as companies not adopting the smart factory concept may not achieve a competitive advantage and desired performance. Moreover, ecosystems of innovation can have an influence on value co-creation and help companies to collaborate and build together higher value solutions (Benitez, Ayala, & Frank, 2020). These ecosystems can encompass innovative firms, educational/training organizations and policy makers (Matt, Molinaro, Orzes, & Pedrini, 2021). Figure 1 summarizes the study's TOE-based theoretical framework and the hypothesized relationships between technology, organization and environmental determinants and smart factory adoption.

As part of a mixed-methods investigation of smart factory adoption in the automotive industry from emerging economies, this study employed a cross-sectional research design to examine the adoption of smart factory technologies. The data collection process involved surveying respondents from selected companies in Nigeria and Morocco. The study gathered data from 167 respondents across 6 automotive companies. The sample included three automotive manufacturing assemblies and two tier-I suppliers. The participants, selected using purposive sampling, were employees occupying senior to mid-level management positions within these companies.

The study's adopted purposive sampling was to ensure that participating firms had sufficient organizational relevance to smart factory adoption decisions and were positioned within the automotive value chain where I4.0 pressures are salient. The five participating companies were selected based on three core criteria: (1) they operate directly in automotive production networks (vehicle assembly or Tier-1 supply), (2) they have established production operations where digitalization initiatives are feasible and (3) they have accessible decision-informants (senior to mid-level managers) with visibility over technology, operations and strategy.

To support representativeness, the sample intentionally included both assemblers and Tier-1 suppliers across two African emerging economies with different industrial trajectories. This design captures variation in institutional pressures, ecosystem maturity and infrastructure conditions, thereby strengthening the analytical usefulness of the Morocco–Nigeria comparison. While the firm sample is not statistically representative of all automotive firms in both countries, it is theoretically representative of the segment most exposed to smart manufacturing pressures and most likely to face adoption decisions in the near term.

The primary data were collected through a structured questionnaire administered to the selected respondents. The questionnaire consisted of items related to respondents' educational background, work experience, company size, level of adoption of smart factory technologies, future plans for adopting a smart factory model, perceived barriers and benefits of smart factory adoption. This study utilized a self-administered questionnaire completed by managerial and professional staff. While survey methods are appropriate for capturing organizational perceptions, readiness conditions and adoption status, self-reported data have inherent limitations. Responses may be influenced by social desirability (the tendency to portray organization in a favorable light), perceptual and recall bias, and respondent interpretation differences. In addition, because the study is cross-sectional, the findings should be interpreted as associations rather than definitive evidence of causality.

To minimize potential bias and improve response accuracy, several procedural steps were incorporated into instrument design and administration. First, participation was voluntary and respondents were informed that responses would be treated confidentially and reported only in aggregate form, reducing evaluation apprehension and socially desirable responding. Second, the survey was structured into distinct sections (e.g. respondent profile, adoption measures and determinant measures) to improve cognitive focus and reduce automatic consistency responding. Further, the instrument was pre-tested and piloted to confirm clarity and relevance of items in the study context, thereby reducing measurement error. Finally, respondents were drawn from roles with direct visibility of operations, technology initiatives and organizational decision-making, which improves the credibility of organizational-level responses.

The data collected were analyzed using two software tools: SPSS and SmartPLS4. SPSS was utilized for descriptive statistics, such as frequencies and means, to examine respondents' characteristics and the level of adoption of smart factory technologies. SmartPLS4, a structural equation modeling (SEM) technique, was employed to analyze the relationships among the constructs and test the proposed model (Rigdon, 2014).

The study underwent meticulous pre-test and pilot test phases to ensure the robustness of the questionnaire design and reliability of the research instruments (Hair, Risher, Sarstedt, & Ringle, 2019). Both the pre-test and pilot test were conducted in a systematic manner, leveraging input from expert evaluations and respondent feedback to refine the questionnaire's constructs (Diamantopoulos, Sarstedt, Fuchs, Wilczynski, & Kaiser, 2012; Dijkstra, 2014) and to guarantee linguistic consistency across multiple language translations.

This initial phase involved a panel of seven experts, comprising four individuals from academia and three professionals from the automotive manufacturing industry. These experts were carefully chosen based on their specialized knowledge within the domains of digital transformation and industrial manufacturing. Their valuable input and expertise assisted in the refinement of the questionnaire's items to ensure their contextual relevance within the specified domains.

The questionnaire, originally scripted in English, underwent a rigorous translation process into French. To ensure linguistic consistency, a back-translation to English was executed by two independent bilingual experts. This process was implemented to detect any potential inconsistencies or discrepancies between the original and translated versions. Comparative analysis between the original and back-translated versions was undertaken to further refine and improve the questionnaire's linguistic consistency and translation equivalence.

Subsequently, a pilot test was conducted with a cohort of 26 respondents. The pilot test aimed to assess the questionnaire's reliability and gather valuable feedback from respondents regarding any difficulties they encountered when responding to the questions (Diamantopoulos et al., 2012).

Upon completion of the survey, respondents were prompted to provide feedback on any perceived difficulties encountered while responding to the questionnaire. Notably, significant attention was directed toward items related to Relative Advantage, Environmental Pressure, Level of Integration and Ecosystem constructs. This emphasis was due to observed instances of perceived repetitions, misunderstanding and lack of clarity within these specific questionnaire sections. The affected questions were either refined for clarity or dropped in the case of repetition (Henseler, Ringle, & Sarstedt, 2015).

Analysis of the pilot test results revealed that Cronbach's alpha values for all the constructs exceeded the 0.70 threshold, indicating strong internal consistency and confirming the reliability of the constructs. Further, the composite reliability (rho_c) shows values between 0.60 and 0.90, which are considered within the range of “acceptable to satisfactory to good” (Hair et al., 2019). The findings reinforced the robustness of the questionnaire design, ensuring the validity of the study's research instruments and paving the way for subsequent phases of the research (Table 1).

First, the result of the country-wise comparison is presented in Table 2. The result revealed significant differences in the level of adoption for the majority of the smart factory technologies between the two countries, except for the IoT (χ² = 1.197, p = 0.550), cloud computing (χ² = 4.227, p = 0.121) and simulation (χ² = 2.746, p = 0.253). Regarding the areas of differences in adoption, 33% of enterprises in Morocco reported full adoption of additive manufacturing/3D printing when compared with none from Nigeria (χ² = 97.09, p = 0.001). On Big Data, 17% of enterprises from Morocco reported full adoption relative to none from Nigeria (χ² = 21.09, p = 0.001). Furthermore, regarding robotics, 27% of enterprises from Morocco reported full adoption, while none of the enterprises from Nigeria reported adoption (χ² = 95.24, p = 0.001). In addition, with respect to the level of adoption of artificial intelligence, 5% of enterprises from Morocco reported full adoption compared with none from Nigeria (χ² = 13.72, p = 0.001). More so, 47% of enterprises from Morocco reported full adoption of cybersecurity relative to 43% from Nigeria (χ² = 27.86, p = 0.001).

The Morocco–Nigeria adoption gap is consistent with TOE's environment and organizational readiness logic: export exposure, stronger supply chain integration requirements and ecosystem maturity typically increase both competitive/institutional pressures and the availability of enabling resources. Morocco's comparatively higher adoption across additive manufacturing, robotics and big data suggests a context where firms face stronger external performance expectations and have greater access to ecosystem actors (solution providers, training capacity and industrial support structures) that reduce implementation uncertainty (Ojubanire et al., 2026). In contrast, Nigeria's lower adoption in advanced technologies is consistent with an environment where infrastructure constraints and limited innovation ecosystem depth increase perceived implementation risk and raise the threshold level of technological readiness required before adoption becomes feasible (Ojubanire et al., 2026). This implies that in African emerging economy contexts, adoption is shaped not only by perceived benefits but by whether environmental and capability conditions reduce implementation uncertainty enough to justify investment.

Overall, the level of adoption regardless of country is shown in Figure 2. The result showed full adoption was found to be highest for cybersecurity (45%), followed by cloud computing (39%) and IoT (23%). The findings highlight the potential for increased adoption efforts in technologies such as additive manufacturing, big data, IoT, augmented reality (AR), robotics, artificial intelligence (AI) and simulation. According to Frank, Dalenogare, and Ayala (2019), there are 3 stages characterizing the complexity level of implementing I4.0 technologies: stage 1 for vertical integration and traceability; stage 2 for automation, and stage 3 for flexibilization. The results obtained are mainly in line with stage 1, as the first leveraged technologies are related to implementing cloud computing and IoT technologies. Big data and artificial intelligence are still not widely adopted across companies.

The stage-1 concentration (IoT, cloud, cybersecurity) can be interpreted as a “readiness-first” pattern: firms prioritize technologies that strengthen visibility, traceability and basic digital control before investing in advanced automation and intelligence. This sequencing aligns with a threshold view of technological readiness in constrained environments, where firms must build a stable digital foundation (connectivity, data capture, security, integration capability) before robotics, additive manufacturing and AI can deliver reliable operational value (Ojubanire et al., 2025).

Furthermore, to determine the extent of adoption across countries, i.e. whether it is low, moderate or high, principal component analysis (PCA) was performed, and the result is presented in Figure 4. To achieve this, composite scores of all the smart factory technologies were generated, while PCA was used to categorize the level of adoption into low, moderate and high. The Chi-square test was used to determine the existence of any significant difference in the level of adoption across the two countries. The result showed significant variation (χ² = 31.998, p = 0.001) in the level of adoption across the two countries examined. Slightly more than half (51%) of enterprises in Morocco, relative to 11% in Nigeria, demonstrated a high level of adoption of smart factory technologies.

From a TOE perspective, this “high adoption” concentration in Morocco suggests the coexistence of enabling technological conditions (readiness and integration capability) and stronger external pressures (market/export requirements), which jointly reduce uncertainty and legitimize investment. The low proportion of high adopters in Nigeria suggests that many firms remain below the readiness threshold where adoption becomes economically and operationally defensible, reinforcing the argument that ecosystem enablement and foundational digital capability are central levers for accelerating I4.0 adoption trajectories in sub-Saharan contexts.

Figure 3 summarizes the country comparison and the overall level of smart factory adoption, highlighting the distribution of adoption intensity across firms and the observed differences between Morocco and Nigeria.

Following the assessment of the level of adoption of smart factory (Section 4.1), hypotheses were tested using bootstrapped partial least squares structural equation modeling (PLS-SEM) estimates. Table 3 reports path coefficients and significance levels. Results show that complexity is negatively associated with adoption, consistent with the expectation that perceived implementation difficulty reduces adoption likelihood. In addition, other studies support this negative relationship (Kumar & Agrawal, 2023; Prause, 2019). In contrast, ecosystem support, environmental pressure, government and regulatory support, level of integration, management support and security concerns show positive relationships with adoption, indicating that enabling conditions and external pressures play a central role in motivating adoption decisions in African automotive firms. Technological readiness and compatibility emerge as one of the strongest positive predictors, suggesting that foundational digital capability is a prerequisite for meaningful adoption. These results confirm those of other studies. Indeed, Silva et al. (2023) confirm a positive impact of environmental pressure (especially the competitive pressure) on smart factory adoption. Kim, Jeong, and Park (2023) have compared in their meta-analysis the existing empirical analysis studies. The results confirm the positive effect of network/ecosystem and government support on smart factory adoption. Moreover, management support is a key enabler for smart factory adoption and can, according to Zhou and Zheng (2023), be driven by competitive pressure and government initiatives.

The study performed by Wagire, Joshi, Rathore, and Jain (2021) on Indian companies confirms the importance of technological competence in terms of readiness and compatibility. In addition, the study highlights the negative effect of organizational resistance. Organizational competence has a weak effect, but it may differ depending on the required skills, which can be technological/digital, social or personal skills. Finally, relative advantage shows a negative relationship, suggesting that a lower perceived relative advantage may impede adoption.

Figure 4 presents the estimated structural model with standardized path coefficients, visually summarizing the direction and relative strength of the TOE determinants influencing smart factory adoption.

The negative relationship between relative advantage and smart factory adoption appears counterintuitive but aligns with recent findings in emerging economy contexts. Senna et al. (2022) argue that in resource-constrained environments, the perception of relative advantage may paradoxically inhibit adoption when organizations recognize the benefits but simultaneously acknowledge their inability to realize them due to financial, infrastructural or skill limitations. In the context of Morocco and Nigeria, firms may perceive smart factory technologies as advantageous yet feel overwhelmed by the gap between current capabilities and required investments, leading to adoption paralysis rather than action (Kumar & Agrawal, 2023).

Similarly, the weak negative relationship of organizational competence requires nuanced interpretation. While organizational competence typically facilitates technology adoption (Wagire et al., 2021), our findings, even though they appear counterintuitive, suggest that in the African automotive context, awareness of competence gaps may create caution rather than confidence. Organizations with higher self-awareness of their current competencies may more accurately assess the substantial skill development required for I4.0 implementation, leading to a delayed adoption until competence-building initiatives are in place (Shevyakova et al., 2020). This finding underscores the importance of distinguishing between perceived organizational readiness and actual capability development in emerging economies.

Conversely, technological readiness demonstrates a strong positive relationship with adoption, confirming that when basic digital infrastructure and ICT systems are in place, organizations are significantly more likely to pursue smart factory implementation (Moldabekova et al., 2021). This highlights that foundational technological capabilities serve as essential prerequisites, whereas advanced financial, infrastructural and organizational competencies may develop alongside adoption rather than preceding it.

Our comparative analysis reveals patterns that both align with and diverge from findings in other emerging economies. The primacy of cost barriers in both Morocco and Nigeria echoes studies from India (Wagire et al., 2021) and Brazil (Tortorella & Fettermann, 2018), suggesting that financial constraints represent a universal challenge across emerging markets regardless of regional differences. However, the magnitude of adoption disparities between Morocco (51% high adoption) and Nigeria (11% high adoption) exceeds differences typically observed between emerging economies in Asia or Latin America.

Morocco's higher adoption levels may reflect its strategic positioning as a manufacturing hub (especially in the automotive sector) for European markets, creating stronger institutional pressures and ecosystem support similar to patterns observed in Eastern European automotive clusters (Matt et al., 2021). In contrast, Nigeria's adoption trajectory resembles patterns in sub-Saharan African economies where limited ICT infrastructure and nascent innovation ecosystems constrain I4.0 implementation (Ojubanire et al., 2022). This suggests that within the emerging economy category, there exists substantial heterogeneity that warrants region-specific analysis rather than generalized approaches.

The positive influence of government and regulatory support found in our study aligns with findings from China and South Korea (Zhou & Zheng, 2023), where state-led industrial policies catalyze technology adoption. However, the relatively stronger effect of ecosystem support in our African context suggests that in environments with weaker formal institutions, informal networks and collaborative arrangements play a more critical role than in Asian emerging economies, where government coordination mechanisms are more developed (Benitez et al., 2020).

The study provides valuable insights into the adoption of smart factory technologies in the automotive industry in Morocco and Nigeria. The analysis of respondent characteristics reveals that a significant proportion of participants hold advanced degrees, indicating a high level of education in both countries. Work experience among participants varies, with different distribution patterns in Morocco and Nigeria. Large companies dominate the industry in both countries, highlighting the importance of addressing smart factory adoption within this context.

This study makes three main contributions. First, it provides rare empirical evidence on smart factory adoption patterns in the African automotive sector and demonstrates substantial cross-country divergence between Morocco and Nigeria. Second, it advances TOE-based adoption research by clarifying that in African emerging economies, adoption tends to be threshold- and enablement-driven, where technological readiness and ecosystem support play central roles under constrained infrastructure and skills availability. Third, it refines practitioner understanding by showing that some commonly assumed drivers (e.g. perceived relative advantage and competence-related perceptions) may not translate straightforwardly into adoption under conditions of capability gaps and implementation risk, highlighting the need for sequenced readiness-building and ecosystem-oriented interventions.

The findings suggest that several factors play a significant role in the adoption of smart factory technologies. Technological readiness, compatibility and management support emerge as crucial factors influencing adoption. Organizations should focus on enhancing their technological capabilities and ensuring compatibility with smart factory technologies. Additionally, strong management support is essential for driving successful adoption initiatives.

Managers seeking to accelerate adoption should treat smart factory implementation as a staged capability-building program rather than a one-off technology purchase. First, firms should establish a minimum readiness baseline (connectivity, data capture, cybersecurity hygiene and system interoperability) and quantify gaps through a simple readiness audit. Second, top management should formalize sponsorship by assigning cross-functional adoption lead, ring-fencing pilot budgets and setting up short-cycle key performance indicators (KPIs) (e.g. downtime reduction, scrap reduction, traceability compliance). Third, firms should prioritize low-regret pilots that build the digital foundation (IoT-enabled monitoring, cloud-based dashboards and cybersecurity controls) before advanced automation and AI deployments. This sequencing aligns with the strong role of technological readiness and management support observed in the model and reduces the “implementation gap” where perceived benefits do not translate into adoption. Since the correlations between relative advantage and organizational competence are weak and negative, managers should complement technology investments with organized change management initiatives that can cope with resistance, re-establish expectations and re-align available process capabilities with data-driven operations. Enhancing management support with explicit digitalization plans, performance indicators and special budgets will be important in transitioning from awareness to implementation.

Policymakers can accelerate adoption by shifting from generic I4.0 aspirations to targeted enabling instruments. Priorities include: (1) co-funded testbeds and demonstration lines in industrial clusters to reduce adoption uncertainty; (2) tax credits or accelerated depreciation for digital equipment and cybersecurity investments; (3) workforce development programs designed with industry (short micro-credentials in automation, data and cybersecurity) and (4) procurement and standards initiatives that reward traceability, quality digitization and data interoperability. These tools strengthen ecosystem support and reduce readiness barriers, which our results indicate are central constraints in African automotive contexts.

Universities, technical institutions, industry organizations and technology providers play a pivotal role in strengthening ecosystem support by co-designing specific curricula, joint training courses, applied research projects, vendor-neutral advisory services and joint pilot programs that allow firms to learn at lower cost. Cluster-based “shared services” (e.g. cybersecurity support, data integration expertise and maintenance analytics capability) can substitute for internal competence deficits and reduce barriers to scaling beyond foundational technologies. Knowledge sharing between original equipment manufacturer (OEMs), Tier I suppliers and digital solution providers can be expedited with the help of collaboration platforms, which can lead to co-creation of context-specific I4.0 applications and ecosystems. These measures are likely to close the skills gap and lessen reliance on imported skills, which were often mentioned as an obstacle in our survey.

To translate the findings into implementable actions, we propose a sequenced roadmap (Table 4) that aligns interventions with the strongest adoption determinants identified in the model. This approach recognizes that firms in constrained environments often require readiness-building steps before advanced automation and intelligence investments become viable.

In the longer term, the observed divergence between Morocco and Nigeria highlights the risk of creating a dual-speed I4.0 landscape in Africa, where a few export-oriented hubs advance rapidly while other economies lag. Strategic I4.0 roadmaps must thus center the adoption of smart factories within wider industrial and skills development strategies, so that digital manufacturing potential does not remain concentrated in a limited number of large companies. With proper coordination, technological preparedness, management support, and ecosystem building can transform African automotive clusters into competitive, resilient value chain nodes in global value chains, leading to structural change and higher-quality jobs within the continent.

This study's cross-sectional design captures adoption patterns at a single point in time; longitudinal research could illuminate adoption trajectories and identify factors that accelerate or impede progress over time. Future studies should expand to other African countries and industrial sectors to assess the generalizability of our findings. Additionally, qualitative case studies could provide deeper insights into the organizational change processes, implementation challenges and success factors that quantitative analysis cannot fully capture. A further limitation is reliance on self-reported survey responses, which may be subject to social desirability and perceptual bias. Although the study incorporated instrument design and administration steps to improve response accuracy, future research should triangulate findings using objective adoption indicators and/or multi-informant designs to further strengthen inference.

In terms of boundary conditions, it is useful to distinguish findings that are likely to be transferable from those that are more context-specific. Findings with broader transferability include: (1) the primacy of technological readiness and foundational digital infrastructure as a prerequisite for I4.0 adoption, which is consistent with evidence from other low-to-middle-income manufacturing contexts (e.g. India, Vietnam) and is likely to apply to any setting combining resource constraints with nascent innovation ecosystems; (2) the enabling role of ecosystem support in compensating for weak internal capabilities, which is theoretically generalizable to other sub-Saharan African economies with similarly thin innovation systems and (3) the counterintuitive suppressor effects of perceived relative advantage and organizational competence under conditions of implementation uncertainty, which align with capability-gap arguments developed in the broader emerging economy I4.0 literature.

In contrast, several findings are likely to be context-specific: (1) the magnitude of the Morocco–Nigeria adoption gap reflects the particular combination of Morocco's export-platform integration and Nigeria's infrastructure deficit, and is unlikely to replicate in pairs of countries with more similar industrial trajectories; (2) the specific threshold levels at which readiness enables adoption will vary with sector-level automation norms and national infrastructure quality and (3) the relative strength of government and regulatory support as a positive predictor may reflect the particular policy environments of the two countries examined. Researchers applying these findings in other contexts should assess the degree of ecosystem maturity and digital infrastructure that characterizes their setting before assuming transferability.