Abstract
The application of Industry 4.0 technologies specifically the internet of things (IoT) and big data (BD) have changed the global industrial system. Many emerging economies are increasingly investing in Industry 4.0 technologies but these economies have not achieved the expected benefits due to insufficient digital infrastructure, financial liquidity and manpower inadequacy. This study reveals the impact of BD and IoT on operational performance (OP) in manufacturing context, with the mediating role of employee training (ET). Drawing upon the resource-based view (RBV) theory, the research examines how the strategic incorporation of technological abilities and human capital leads to improve productivity, quality, flexibility, and cost efficiency in manufacturing sector. A quantitative method was used to analyse survey data collected from 320 managers and executives representing manufacturing firms engaged in Industry 4.0 implementation. To analyse the relationship among variables statistical analyses were performed using SPSS 26 and Smart PLS 4. The findings indicate that BD and IoT have a positive and significant impact on OP, while ET further strengthens these effects by establishing the connection between technological aspects and performance outcomes. The results underscore the importance of ongoing employee training and digital innovation endeavours. This study recommends valuable information for practitioners and policymakers to enhance sustainable competitiveness through advanced technology and human resource abilities. Empirically, this research is among the first to study the impact of Industry 4.0 adoption with employee training in developing economies, providing valuable insights for both researchers and practitioners.
Plain Language Summary
This study explores the influence of transformative Industry 4.0 technologies, namely big data (BD) and the Internet of Things (IoT), on enhancing operational performance (OP) in the manufacturing sector, while highlighting the mediating role of employee training (ET). This research work examines the drivers of digital transformation in developing economies. Drawing on a sample of 320 managers and higher-level executives who have already integrated Industry 4.0, research employs quantitative survey method to measure quality improvement, manufacturing flexibility, productivity, and cost reduction. Analysis is performed through SPSS 26 and SmartPLS 4. Results showed that both big data and IoT significantly improve performance indicators. Employee training strengthens these relationships, serving as a critical mediating factor. From a managerial perspective, these insights highlight the imperative of aligning technological investment with human capital development to maximize long-term benefits. Equally, policymakers could capitalise on this knowledge to develop policies that encourage digital adoption, alongside structured training programmes, to address the lag in industrial innovation. Ultimately, this study contributes by highlighting its strategic significance and clarifying how organisations in developing economies can optimise performance through a balanced interplay of technology and people.
Keywords
Introduction
Industry 4.0 represents highly interconnected and intelligent production systems for accomplishing autonomous decision making and communication (Piccarozzi et al., 2018). In the information technology (IT) driven revolution, manufacturing organisations’ workflow changes include the digitalisation of vertical and horizontal integration (Haverkort & Zimmermann, 2017). Automation of organisational processes, incorporation with e-commerce, and decentralised decision-making are a few of the substantial business opportunities (Haverkort & Zimmermann, 2017; Hitpass & Astudillo, 2019). While digital technologies are not novel, they are progressively advancing in complexity and integration with innovative information and communication, assuming a critical function in restructuring societies and the worldwide economy (Schwab, 2017).
Amid global digitalisation, developing economies must urgently adapt these technologies to maintain competitiveness. Organisations are continuously adapting to the emerging technologies, including artificial intelligence and IoT devices, which are integral components of a global shift known as Industry 4.0. South Asia have accelerated swiftly with contemporary technologies, especially cloud-based services and mobile applications (Imran et al., 2018). However, the manufacturing sector in Pakistan has not fully leveraged the rise in its production capacity (Abbas, 2023). Tasneem and Khan (2024) state that, unlike global trends, the manufacturing sector in Pakistan is adopting innovations more slowly in the mining, construction, and textiles sectors, even when the development of information technologies is considered a systems opportunity. Automation and digitalisation on a larger scale have the potential to alleviate inefficiencies concerning traditional operational approaches. Due to its prevalence, decision-making based on data usage has been recognised as an essential element to modern life. Economically, systems for autonomous decision-making are now feasible (Khaleel et al., 2015). Within the context of the Industry 4.0 framework, this is important to lessen obstacles while improving performance, especially in developing economies (Imran et al., 2018). However, there are significant barriers to take advantage of Industry 4.0. In the literature, the most significant are barriers to implementation. The literature refers to the most serious barrier as inadequate training in digital and technical skills as well as the least serious compared to other barriers that could be inclusive of having to make a financial investment, cybersecurity, company size, business model or cultural change, and lack of infrastructure (Cohen, 2017; Franzoni & Zanardini, 2017). Comprehensive training of staff while situated in a low-tech human-centred context utilising first principles of management theories is fundamental for a sustained continuous improvement effort (Marodin et al., 2019; Seppälä & Klemola, 2004). The expanding body of literature highlights the increasing importance of technology, while the literature also reveals some major gaps in the research agenda. There is still a noteworthy gap in the literature despite the added focus for digital transfer and transformation focused on around the world. First, the majority of research on Industry 4.0 has taken place in contexts within advanced economies. Second, there is lacking empirical research that focuses on the use of IoT and big data, especially in developing economy contexts such as Pakistan (Sumbal et al., 2025). While a consensus does exist about the disruption that digital technology could bring to organisations, there has not been much academic historical work on training employees as a mediating variable. The human factor is an important driver for organisations to realise the performance benefits associated with adopting Industry 4.0 (Fahim et al., 2023). Previous research has predominantly proceeded with discrete or isolated measures of performance benefit. There are very few studies that have actually addressed the multifaceted measures of performance including operation cost efficiencies, production flexibility, and operational efficiency (Kalsoom et al., 2021). The present study aims to address these gaps by exploring the influence of IoT and big data on operational performance in Pakistan's manufacturing industry, and the mediating role of employee training. The proposed study aims to:
Investigate the influence of IoT and big data on various operational performance aspects.
Investigate whether Industry 4.0 technologies and operational performance is mediated by employee training.
Investigate if employee training has a direct and unique effect on operational performance.
In order to achieve the aims mentioned above the research study will address the following important research questions (RQ) in sequence:
Literature Review
This section proposes to explore the elements of Industry 4.0, what they can offer, and how these elements may be adopted by manufacturing industries in Industry 4.0.The section then identifies gaps in the literature around the manufacturing industry.
Industry 4.0
The concept of Industry 4.0 has become a trending topic in academia and has provided a foundation for developing studies in many areas, including transportation, manufacturing, mining, and health (Habraken et al., 2019; Jabbour et al., 2020). The idea of Industry 4.0 arose during industrial demonstrations in Germany (Bader et al., 2020; Drath & Horch, 2014). The term is even credited to the German federal government (Frank et al., 2019). Operating at a high level, Industry 4.0 is regularly perceived as a future development, rather than even being called a groundbreaking phenomenon (Drath & Horch, 2014). In a related note, while the technology itself is not new, its techniques, or relationships, have rendered technology far more complex and have disrupted mankind, society, and culture (Schwab, 2017). The use of real-time wholesale data collection by businesses, allows for decentralised decision making with artificial intelligence and machine learning algorithms (Piccarozzi et al., 2018). These technologies can change business models and provide important improvements (Haverkort & Zimmermann, 2017). Industry 4.0 consists of the use of automation, control systems, and information technology to improve production in factories of the future (Wamba et al., 2020).
Industrial Internet of Things
The industrial Internet of Things (IoT) allows decentralisation of decision-making and automation which can tremendously improve production processes, and in particular, Cyber Physical Systems (CPS). With heightened connectivity of devices, performance may also improve because integrated processing in CPS systems has improved opportunities for coalitions (Drath & Horch, 2014). According to Brettel et al. (2014), German companies have successfully built large and complex supply chains using extensive end-to-end CPS systems. Implementing Industry 4.0 has a number of potential benefits for developing countries like Brazil, including digitising industrial plants and converting business models to accommodate innovations to come (Frank et al., 2019). Certain dimensions of Industry 4.0 show applicability in a variety of areas such as mining, food, textile, leather and related products, chemicals, and soap and detergents. Thus, it creates various possibilities to reap rewards from its extensive use. For example, building an IoT-based Smart Agriculture system, where IoT devices and sensors give information about the crops and their health, could greatly increase crop yields (Ayaz et al., 2019).
Big Data
Big data technology, which deals with the management of large and complex datasets which could be structured datasets or unstructured datasets has enabled the effective use of traditional information systems. These datasets, which are frequently perceived as information waste, can be converted into manageable and digestible information (Shabbir & Gardezi, 2020; Ying et al., 2021). Big data technologies could provide businesses with limitless data collection and aggregation, real-time insights, and current awareness of trends (Gupta et al., 2020; A. Kumar & Chauhan, 2024; Su et al., 2022), the ability to evaluate their position relative to competitors, access to comparative data, and prompt action (S. Ali et al., 2020; Sekli & De La Vega, 2021). In addition, the growth of big data has led to a technological solution that is also emerging called cloud computing. (Bag et al., 2023; Zhu et al., 2022). It is anticipated that the effective use of big data will affect value creation and create a competitive advantage. Two instances of such an influence are new ways of engaging customers (Verma et al., 2023) and enabling the development of new goods, services, and initiatives that enhance productive efficiency (Ahmad et al., 2023); a considerable big data architecture has been constructed to support the data requirements of the manufacturing organisations (Santos et al., 2017). This architecture incorporates various layers and components to improve data collection, storage, processing, analysis, and sharing (Aljumah et al., 2021). Therefore, it can be argued that the resulting integrated context aids decision-making at the different managerial tiers based on the preceding evidence.
Industry 4.0 and Operational Performance
The overarching goal of Industry 4.0 is to enable improved operational performance and productivity combined with greater automation (Črešnar et al., 2023; Mubarak & Petraite, 2020). The role of Industry 4.0 in the manufacturing and service sectors is significant as there is an evident link between performance and Industry 4.0 (Abdullah et al., 2025). Reyes et al. (2023) noted that some of the characteristics of Industry 4.0 relates strongly to internet-based technologies and even more advanced computational procedures. However, it is also noted that Industry 4.0 is a collection of technological processes that improve value addition using suitable information management systems to improve performance (Rahman et al., 2022; Sharma et al., 2022). The current expectation of Industry 4.0 states it can solve multiple problems by utilising advanced technologies. The research literature identifies unified data systems attributed to automation, collaboration and integration, are key facilitators of productivity and performance within Industry 4.0 (Pozzi et al., 2023; Sony & Naik, 2020). The presence of these facilitators is of paramount significance for the manufacturing and service industries, regardless of the type of economic system. Moreover, the assistance and encouragement of Industry 4.0 would automatically increase the general efficacy and performance of the manufacturing industries (Yadav et al., 2020).
With Pakistan being a predominantly agricultural society, it can also have social and political changes. However, it is essential to note that most of these technological capabilities are emerging and require additional time before achieving broad adoption (Haverkort & Zimmermann, 2017; Marodin et al., 2019). ICT (Information and Communication Technology) can reduce costs and improve production capabilities in manufacturing companies (Haverkort & Zimmermann, 2017). Industry 4.0 has also sparked interest in investigating the potential benefits of sustainable manufacturing processes in manufacturing companies (Seppälä & Klemola, 2004). Increasingly, larger firms and corporations have started utilising big data analytics to process large real-time data swathes (Akter et al., 2016). Technology sectors in developing countries have begun to gain benefits from IoT devices. One example is the Careem or Uber applications that utilise consumer mobile devices as a decentralised IoT platform where multiple devices communicate and interact.
Industry 4.0 promotes an innovation and learning paradigm reliant on human resources which puts an additional emphasis on human beings performing creative and communicative tasks (Erol et al., 2016; Shamim et al., 2016). Skilled workers are important in Industry 4.0 as they were in the previous industrial revolutions (Balasingham, 2016; Sirotek & Firlus, 2016). It is important to develop employees to facilitate a seamless transition to digitalisation, thus, training employees to use digital systems is essential. Employee technical training supports the adaptation of new technologies and skill development ensures digitalisation will be seamless (Agostini & Filippini, 2019).
Theoretical Framework and Hypotheses Development
This research paper draws upon two foundational theories; the Sociotechnical System (STS) theory and resource-based view (RBV) theory. The STS theoretical model (Pasmore et al., 1982; Trist & Bamforth, 1951) states that organisational performance occurs as a function of technological systems and social (human) systems, and in order to achieve maximum operational effectiveness, both parts must be collectively transformed. Research suggests that acceptance of Industry 4.0 requires current Human-centric approaches that are synergistic with technical capabilities. There is recent evidence of technology substantiating improvements in operational performance (Roth & Farahmand, 2023; Tortorella et al., 2023). These studies show improvements in technology integration contributing to employee job satisfaction, quality of products/services, and overall organisational performance. According to empirical evidence based on resource-based view (RBV) theory (Bag et al., 2023), employee training is essential to enhance operational performance through the integration of digital technologies. Thus, through enhancing organisational learning, flexibility, and performance outcomes, employee training is a core capability that transforms baseline technology (big data, IoT) into firm-specific advantages.
Consequently, instead of being redundant, the proposed theoretical framework is theoretically supported through these two complementarity frameworks. Based on the framework's theory proposed in the model -
Big data and IoT directly influence operational performance.
This relationship is mediated through employee training, enabling the employee to incorporate and use digital technology prudently.
This positivist approach is based on established theories that interconnect technology, human capital, and organisational performance, and offers a means to test given hypotheses based on empirical data. Industry 4.0 entails interconnecting and incorporating the real and virtual worlds by means of the Internet of Things (IoT) and big data. It accomplishes this interconnectivity by integrating intelligent devices that communicate and interconnected consistently with one another (Öberg & Graham, 2016). This inter-connection of devices is intended to accomplish certain strategic objectives. Industry 4.0 technologies alter organisations' performance, their implementation must be at the level of a strategic decision. Therefore, organisations need to evaluate their readiness to implement Industry 4.0 before making this critical decision. A strategic transformation of the company's entire organisation is required to progress through digitisation, which also is another important strategic choice that requires consideration. This study examines the various facets of Industry 4.0 that affect an organisation's operational performance. These arguments suggest there is an important need for alternative lenses and frameworks and provide empirical proof of the merit in investigating this new paradigm. Furthermore, it substantiates that possibility of improvements and challenges associated with Industry 4.0 provides a new outlet for value creation of operational management practice. It emphasises the significance of effective operational management, in a transformative global industrial context. Manufacturers need to prove some agility, efficiency, reactiveness, and a cost-effective approach to continue reducing operational expenses, and respond to shifting client demands in a highly volatile environment. The increased usage of automation and digitalisation in the company's external operations and distribution network is proof of concept. A company survey involving over 200 employees working in Pakistan's textile and services sector, revealed Industry 4.0 could help offset productivity challenges faced by the manufacturing sectors of the country (Imran et al., 2018). Developing an operational strategy is crucial for the implementation of Industry 4.0, to improve operational performance, and to establish competitive advantage. Thus, it is important for organisations to clearly delineate their transformation process and how they will manage operations to develop business objectives that will help direct them towards their overarching goal. These strategic objectives create competitive priorities that are then used to execute the operational plan for performance improvement.
As shown in Figure 1, big data and IoT are viewed as the independent variables, employee training is viewed as the mediating variable (Shamim et al., 2016) and operational performance is the dependent variable. This study positions employee training as a mediator to explore how big data and IoT impact performance. The mediating role of employee training is significant as it points to the ability of the digital technologies to translate into operational performance outcomes.

Theoretical framework.
Big Data and Operational Performance
The concept of “big data” encompasses the adoption of new and complex data sources to obtain extensive understanding of an organisation’s operational activities (Hussain et al., 2023; Raj et al., 2024). Using big data to extract business-related insights, patterns, or trends is imperative for supporting sustainable innovation within an organisation, in alignment with Industry 4.0 (Khang et al., 2023; Tabesh et al., 2019). Assessing current developments in analytics is important. Industry 4.0 involves the adoption of appropriate big data tools and technologies to successfully integrate the data collection, storage, processing analysis needs (Awan et al., 2022; Niu et al., 2021). In the context of big data systems, taking on denormalised structures allows quick processing and easier methods of obtaining data, due diligence, and scaling, which reduced time lag between the initial stage of obtaining data and later data analysis stages (Mamo et al., 2022). Collaboration helps to enhance decision-making through the intimate connection of big data and analytics. In a previous study, the researchers suggested that the academic community investigate holistic data management capabilities for enhanced creativity in public policy formulation (Sadiq et al., 2021; Shams & Solima, 2019). Moreover, there is limited prior research to elaborate on the relationship between big data and quality improvement, productivity, and manufacturing flexibility to help optimise operations (H. Saleem et al., 2021). Businesses must continue to employ manufacturing flexibility to improve operational efficiency and effectiveness, while the manufacturing sector inspires their management personnel to facilitate the innovation process (Latif et al., 2023). The RBV theory perspective observes big data technologies as valuable assets that facilitate the achievement of greater productivity and efficiency. When successfully developed and integrated into several manufacturing processes, big data provides real-time insights, supports predictive analytics and the optimisation of work-in-progress inventory, processes and operations. These capabilities are critical for improving productivity, reducing costs, enabling operational flexibility, and enhancing product quality (A. Kumar and Chauhan, 2024). Several empirical studies suggest operational advantages of using big data analytics. For example, Wamba et al. (2020) and Imran et al. (2018) both demonstrate that big data analytics contributes to flexible production and continuous quality improvement. Kazmi and Abbas (2020) found, in industrial setups, data driven operations were positively associated with overall productivity and cost effectiveness.
IoT and Operational Performance
The transition to the fourth industrial revolution involves digitalisation and the incorporation of Internet of Things (IoT) into industrial processes create further opportunities, and examines the new opportunities and challenges of digitalisation (Bhatti et al., 2021). In addition, there is increased market competition, the importance of IoT and efficient business processes is crucial for organisations to facilitate operational efficiency. In order to facilitate these objectives, the introduction of advanced systems of digital information and communication technologies can highly facilitate productivity and performance (Mostafa et al., 2020). The Internet of Things is greatly shaping how organisations engage with innovation and add value to their daily operations. The Internet of Things is emerging as a prominent area of interest for many organisations in several studies (Antouz et al., 2023; M. U. Saleem et al., 2023). Prior research was unsuccessful in considering the efficient performance implications of IoT technologies. This research suggests that previous studies exhibited limited interest in studying the impact of IoT technology on performance in the manufacturing sector within the context of Pakistan. The integration of a technological infrastructure with human and organisational factors is highlighted in the use of IoT technologies in manufacturing. Fundamentally derived from Socio-Technical Systems Theory (Bostrom & Heinen, 1977), IoT produces a production environment that is more responsive and agile, in a process of continuous operational performance and adaptability through machine-based communication and monitoring, and automating tasks. New research demonstrates how IoT impacts various aspects of performance, and further explanations are now available. Kortmann (2020) demonstrated how IoT improves industrial responsiveness and adaptation. Similarly, Nawanir (2016) and Iqbal and Rahim (2021) reported that IoT also leads to better quality control during the production process.
Integration of digital technologies creates an optimistic prospect as well as significant challenges for manufacturing firms to remain competitive and grow their businesses (Nwagwu et al., 2023). The integrations have dramatically shifted from traditional production and operation management practices and methods (Khan et al., 2024). Integration of digital technologies in an organisation does involve the ability to manage information (Waheed et al., 2023) which supports decision-making contexts mostly related to operation management (ul zia et al., 2023).
Employee Training, IoT, Big Data, and Operational Performance
Industry 4.0 implies the impact of the fourth industrial revolution on existing practices of using technology in manufacturing, which pertains to information sharing, conducts transactions, and automates operations. The industrial revolution significantly defines the long-term success of an organisation from an economic and social perspective, which appears in the ideas of big data and the IOT (S. Ali & Xie, 2021; Mubarak et al., 2019). Past studies have indicated the importance of these connections and have shown that an organisation that wants to promote organisational development needs to consider how Industry 4.0 fundamentally influences operational outcomes. Understanding these influences will be important, as Industry 4.0 is required to sustain organisational productivity and successful operational strategy (Imran et al., 2018; Nasir et al., 2022). The assertion pertains to the necessity for organisations to align their ongoing enhancements with emerging industrial advancements in automation and data exchange in order to achieve objectives related to organisation’s efficacy and performance (K. Ali & Kausar, 2022; Umar et al., 2022). Industry 4.0 is forecasted to incorporate digital technologies, to facilitate the link between the Internet of Things (IoT), the collection and analysis of big data, and the enhancement of operational performance organisations. Nevertheless, the current body of research on operational advancement and productivity lacks to explain the mediating effect of staff training on big data, quality improvement, manufacturing flexibility, increased productivity, and cost reduction in the manufacturing industry in developing countries. Training incorporates to equip or to develop specific knowledge, skills or behaviour relating to specific task, and provides the knowledge to employees to perform the tasks accurately (Cohen, 2017). Employee training is necessary for any organisation to implement any new technology. In this digitalisation era where new technology comes quickly, the knowledge and skills of employees become vital sources for all stakeholders (Presbitero, 2016). RBV theory (Barney, 1991) established that organisational capabilities namely, highly skilled human resources are critical to convert technical resources to improvements in sustainable performance. Simply, obtaining technologies is not enough to support an implementation of big data. Employees need to possess the necessary technical and analytical skills to extract meaningful insights from the data. Hence, employee training is a key facilitator in converting investments in technology into tangible operational improvements. Earlier studies have identified the importance of employee preparedness in optimising the benefits of technology driven data. Imran et al. (2018) and Maroufkhani et al. (2020) stated, for example, that improvements in efficiency and effectiveness with big data analytics, occurred mostly in organisations that invested in the development of their workforce. It was found that training programmes improved analytical capabilities, led to better decision-making, and underpinned improvements in productivity and efficiency. Therefore, according the study hypotheses developed the third hypothesis
As posited in STS theory (Bostrom & Heinen, 1977), the effectiveness of technological systems such as the Internet of Things depends on the human workforce's ability to adapt and interact with those systems. The deployment of IoT in the manufacturing sector requires systems, infrastructure, and skilled workforce. Without proper employee training relating to the IoT value propositions, the full utilisation of IoT technological advancements including automation, data-responsive capabilities, and system integrations in a manufacturing setting will be limited. Thus, its impact as an overall contributor to performance outcomes may also be limited. The importance of workforce training to realise the potential of IoT devices is supported by recent investigations. For instance, Nimawat and Gidwani (2023), and Iqbal and Rahim (2021) argue that workers who have received targeted training have better capabilities in managing networked systems, responding to real-time data, and responding to operational signals. Workforce training has been linked to improvements in productivity, costs, operational flexibility, and quality. The study responds to this by proposing the fourth hypothesis.
According to the Resource-Based View (RBV) (Barney, 1991), the capabilities of employees especially those developed through formal training are indispensable organisational resources in the ultimate pursuit of sustainable competitive advantage. Employee capabilities enable the effective operation of best-in-class technologies, mitigate variances in processes, and facilitate an ongoing cycle of operational improvement because of training that emphasises the development of technical skills and problem-solving. The unique benefits of these contributions ultimately drive improvements in productivity, quality, robustness, and cost containment.
Moreover, Human Capital Theory (Becker, 1964) maintains that improved investment in employee training and development enhances their contributions to organisational success, especially in complex and technologically rich contexts, such as Industry 4.0 transitions.
Sung and Choi (2014) support the importance of workforce training in providing a way to strengthen productivity, enhance product quality, and reduce operating costs, as works shown in both manufacturing environments.
Nimawat and Gidwani (2023) have also suggested that employee training improves overall performance outcomes through employee acquisition of the key skills to effectively work with intelligent manufacturing technology, while Maroufkhani et al. (2020) highlighted that trained and capable employees provide operational flexibility and efficiency, especially when undergoing the digital transformation. Based on the discussion above, the following hypothesis is presented:
Methodology
A cross-sectional, quantitative, and explanatory design to study the perspectives was applied. Quantitative research approach was used in order to determine the influence of independent and dependent variables while assessing the mediating effect. This approach creates a respondent base, which is essential for engaging in any statistical techniques, and provides an orderly procedure for collecting data (Creswell, 2018). The information was collected using an online survey questionnaire. Between May 1, 2024, and October 30, 2024, the member received the online questionnaire link via email based on acquired participants' informed consent, voluntary participation in the survey, and understanding of all relevant information regarding the study. Feedback from 320 respondents was included in the study for additional analysis.
There are additional limitations of the sampling method. The external validity of the results are limited if a non-probability sampling approach (such as convenience sampling or purposive sampling) was conducted. The reliability of estimates and significance tests might be affected by a smaller sample size, which also reduces statistical power particularly important for structural equation modelling. The purpose of the study utilised a purposive sample strategy, targeting individuals with real-world experience hosting operational roles, implementing technology, or overseeing workforce development programmes in order to establish the relevance and depth of data. Individuals working within operational, information technology, and human resources roles were among the participants; they were all qualified to provide meaningful insights pertaining to the research variables. While adopting this sampling method may facilitate access to knowledgeable individuals, there are limitations to generalizability. By using this approach, every individual has equal access to being selected to participate (Paul et al., 2014). The rule of thumb method from PLS-SEM literature was used for specifying sample size estimating, by proposing at least 10 comments for each indicator associated with the most complex construct in the study (Chin, 1998). It was determined that 150 responses were appropriate, since the maximum number of indicators for any one latent variable was five. The final data collection comprised 320 usable responses, exceeded any recommended guidelines for sufficient analytical power to address the research questions to enhance statistical reliability, and generalisability (Hair et al., 2022). The survey instrument was derived from previous research by Imran et al. (2018), Agostini and Filippini (2019), Nawanir (2016), and Dalenogare et al. (2018). This study is based in industrial context of Pakistan, which is adopting digital transformation technologies.
Pakistan is a transitional economy and faces structural and technological challenges associated with digital transformation, including unpredictable levels of digital integration among various sectors, limited advanced infrastructure access, and skills shortages in the workforce. As such, it is a strategically relevant context for examining the outcomes of key technologies associated with Industry 4.0 on operational performance, specifically big data and the internet of things (IoT) while also identifying employee training as a mediation variable. The research population consisted of senior executives and managers working in the operations departments of the industrial companies in Pakistan. The secondary data consisted of newspaper reports, conference papers, and academic journal articles. Utilising questionnaires is a frequent method for confirming the validity and reliability of previous studies (Rehman, 2023). Creswell (2018) explains that survey research is also beneficial for determining how variables are interrelated. Researchers utilised online research as a useful and viable method (Sue & Ritter, 2007). All the questions and statements were measured using Likert scales, which reflect attitudes toward a topic and allow for the statistical measurement of relationships between variables (A. A. Kumar et al., 2019). The instrument consisted of two parts, the first part of the instrument included general question and demographics data and the second part included 20 questions related to attitudes and behaviours toward big data, internet of things, employee training, and operational performance. The study used SPSS 26 and SmartPLS 4.0 to explore relationships among latent constructs, specifically when a mediation model was applied with multiple indicators. This procedure is useful for predictive and exploratory models with complex relationships (Hair et al., 2022).
SPSS 26 was used to perform regression, correlation, reliability, and descriptive statistics. The analytical process for the SEM model followed a two-step approach, starting with the assessment of the measurement model (indicator loading, reliability, and average variance extracted) and followed by structural model assessment (path coefficients, determination of coefficient (R2), effect sizes (f2), and model fit).
The SEM model consisted of two steps method, the first step was assessing the measurement model, including indicator loading, reliability, and average variance extracted. The second step was evaluating the structural model and reporting in terms of path coefficients, determination of coefficient (R2), effect sizes (f2), as well as assessing model fit. Tables, graphs, and charts were used to facilitate the understanding of the analysed data.
Analysis and Results
Reliability Test
Cronbach’s alpha measures the reliability of a scale by assessing the internal consistency of a set of items, or how well items in a scale are measured.
Table 1 elaborates the reliability statistics by showing all values are above 0.7, except for productivity growth (.603). Table 1 provides the reliability statistics for each construct. In social science research work, a reliability value of at least 0.70 is acceptable, representing adequate internal consistency (Black, 2010). The main constraint is the measurement reliability of some constructs, specifically, the Cronbach’s alpha for the construct is < .70, the standard cutoff (Nunnally & Bernstein, 1994).
Reliability Statistics.
Note. Listwise deletion based on all variables in the procedure.
Descriptive Statistics for Control Variables
The initial section of the questionnaire consisted of control variables related to gender, age, education, experience, and organisation type, as shown in Table 2. The second part consisted of questions about industry 4.0 components.
Demographic Data of Respondents.
Table 2 represents the demographic values collected from different manufacturing organisations according to gender, age, education, experience, and organisation type. It elaborates that the frequency of male respondents is higher than females. The frequency of the age limit from 36 to 45 is higher than other age limits. The experience limit from 3 to 5 is higher than all other experience limits. In education, most of the respondents are Master’s. The frequency of organisations with above 1000+ employees provides the highest responses in the size of organisation.
The descriptive statistics of primary constructs of big data (MBD), internet of things (MIOT), employee training (MTR), quality improvement (MQI), manufacturing flexibility (MMF), increase in productivity (MP), and cost reduction (MCR) are shown in Table 3. The assumption of data normality was met as the standard deviation of the sample mean is between +3 and -3, and the values of skewness lie between +1 and −1 (as shown in Table 7), demonstrating that the data distribution is normal (Haverkort & Zimmermann, 2017). The standard deviation ranges between 1.082 and 1.249. The mean value of the construct MBD (big data) is 3.59, MIOT (internet of things) is 3.94, MTR (employee’s training) is 3.34, MQI (quality improvement) is 3.90, MMF (Manufacturing Flexibility) is 3.74, MP (increase in productivity) is 3.74, and MCR (cost reduction) is 3.64.
Descriptive Analysis.
Correlation Analysis
The Pearson correlation coefficient can be utilised to ascertain whether a linear relationship exists between the variables and to measure the strength of this relationship. The linear relationship exists and can be easily identified if the value lies between ±1.0; the value <0 means that there is a negative linear relationship and vice versa. The results in Table 4 elaborate that the correlation values lie in the range of 0.372 and 0.832.
Correlation Analysis.
Correlation is significant at the 0.01 level (two-tailed).
Regression Analysis
Regression analysis is conducted to examine the association between the independent variable (IV) and the dependent variable (DV) and to analyse the impact of IV on DV. Table 5 shows the values for the impact of the independent variable on the dependent variable.
Regression Analysis (Big Data and Operational Performance).
Table 5 represents the impact of Industry 4.0 factors on operational performance in the manufacturing industry. For Industry 4.0, big data (MBD) is taken as an independent variable, while quality improvement (MQI), manufacturing flexibility (MMF), increase in productivity (MP), and cost reduction (MCR) are taken as dependent variables. According to Table 5, independent variable big data (MBD) is significantly predicting quality improvement (MQI) with values (t = 23.8, p < .05, b = 0.885), manufacturing flexibility (MMF) with values (t = 16.94, p < .05, b = = 0.803), increase in productivity (MP) with values (t = 9.59, p < .05, b = = 0.607) and cost reduction with values (t = 11.83, p < .05, b = = 0.686). A significant variation is observed in operational performance factors (quality improvement, manufacturing flexibility, productivity increase, and cost reduction) due to industry 4.0 factors (big data and the Internet of Things). As a result of the regression and correlation analysis, Table 5 concludes that hypothesis H1 is accepted.
Table 6 represents the Internet of Things is also significantly predicting dependent variable quality improvement (MQI) with values (t = 30.43, p < 0.05, b = 0.924), manufacturing flexibility (MMF) with (t = 25.56, p < .05, b = 0.897), increase in productivity (MP) with (t = 17.12, p < .05, b = 0. 0.806) and cost reduction with (t = 14.59, p < .05, b = 0.758). Similarly, significant variation is observed in factors of operational performance (quality improvement, manufacturing flexibility, increase in productivity, and cost reduction) due to factors of Industry 4.0 (big data and the internet of things), concluding that hypothesis H2 is accepted.
Regression Analysis (IoT and Operational Performance).
Table 7 reveals a significant positive impact of employee training on key dimensions of operational performance. The strongest effect is observed on quality improvement (β = .808, t = 21.155, p < .001), indicating that employee training enhances technical skills and reduces errors. Similarly, employee training strongly influences productivity (β = .771, t = 20.608, p < .001) and manufacturing flexibility (β = .701, t = 11.675, p < .001), suggesting that a skilled workforce can increase efficiency and respond better to changing demands. Despite being relatively smaller (β = .564, t = 6.359, p < .001), the effect on cost reduction is still significant, indicating better resource usage and less operational waste. These findings demonstrate employee training is vital to optimise the benefits of Industry 4.0 technologies. H5 is approved since employee training significantly improves operational performance components.
Regression Analysis (Employee Training and Operational Performance).
Model Fit Using SEM
The structural equation modelling results (SEM) offer a holistic view of the interrelations among big data, internet of things (IoT), staff training, and operational performance in manufacturing sector. Based on the results, big data has a significant and strong effect on employee training (β = .898, p < .001), showing that firms that successfully leverage big data are more likely to invest in skills development. Furthermore, big data contributes to operational performance (β = .299, p < .001), highlighting its role in supporting better decision-making and efficiency of processes (Figure 2).

Structural equation model (self created).
In contrast, the IoT is not significantly affecting employee training (β = .014, p = .889), showing that adoption of the IoT alone will not produce any training unless it is complemented with other organisational initiatives. The analysis shows that training significantly improves quality (β = .808, p < .001), productivity (β = .771, p < .001), flexibility (β = .701, p < .001), and cost efficiency (β = .564, p < .001). The outcomes highlight the importance of a trained workforce in managing contemporary production processes, reducing errors, and adapting to changing requirements. However, there was a negative effect (β = −.165, p = .017) between training and overall operational performance, indicating that training could potentially have more of an effect on specific performance outcomes. It is also important to note the explanatory capability of the model. While the model explains a high level of operational performance outcomes (R2 = .916), big data and the IoT together explain a large amount of the variance in staff training (R2 = .829). The model also provides a solid explanation of other performance based outcomes such as quality improvements (R2 = .653), productivity (R2 = .594), flexibility (R2 = .491), and cost savings (R2 = 0.319) in relation to staff training. Overall, the model suggests that staff training is an important pathway through which digital technologies affect operational effectiveness, and that big data is a key driver for both training and performance outcomes. These findings underscore the importance of developing the workforce to align with digital transformation to take advantage of developments in Industry 4.0.
In the Model Fit table (Table 8), both the Saturated Model and the Estimated Model evidenced a Standardised Root Mean Square Residual (SRMR) value of 0.099. SRMR is an important statistic in evaluating how closely the predicted correlations line up with the actual data. Commonly, in most applications, SRMR values below 0.08 demonstrates an acceptable model fit (Hu & Bentler, 1999). The model can also be considered to have a reasonable fit instead of good fit because the value of 0.099 was just above this unacceptable cut-off value (Henseler et al., 2014). Although the general framework of the model is rational, there may be possible areas for improvement. In any case, modifying or refining the indicators’ loadings, or adjusting the connection between variables, or correcting any excess correlations with large residuals will make the model more precise.
Model Fit.
Mediation Analysis
Mediation analysis uses mediator variables to decompose overall exposure-outcome effects into direct and indirect effects (Nguyen et al., 2021).
Table 9 represents the mediating role of employee training (MTR) of Industry 4.0 between independent and dependent variables. The values indicate that employee training (MTR) has a partial mediating role between big data (MBD) and quality improvement (MQI) with values (b = 1.1025, p < .05, S.E. = 0.0992, t = 12.4964). The values for the lower limit confidence interval (LLCI) and upper limit confidence interval (ULCI) are also positive, indicating that the model is significant. It is also observed that the effect of IV on DV is positive and in the range of 0-1, suggesting the model's significance. So, hypothesis 3 is accepted that employee training (MTR) plays a mediating role between big data (MBD) and quality improvement (MQI).
Mediation Analysis.
Moreover, according to Table 9, it is also observed that employee training (MTR) plays a mediating role between big data (MBD) and manufacturing flexibility (MMF) with values (b = 0.3430, p < .05, S.E. = .1519, t = 2.2587). Additionally, the values for LLCI and ULCI are positive, as is the effect of IV on DV, resulting in hypothesis acceptance. The values for employee training (MTR) with values (b = 1.5631, p 0.05, S.E. = 0.1635, t = 9.5599 ) recommend that (MTR) plays a mediating role between big data (MBD) and an increase in productivity (MP), with positive values for LLCI, ULCI, and the effect of IV on DV. Employee training (MTR) also plays a mediating role between big data (MBD) and cost reduction (MCR), with positive values for the effect of IV on DV, LLCI, and ULCI, (b = 1.4951, p .05, S.E. = 0.1383, t = 10.8081 ), indicating the significance of hypothesis 3.
Hypothesis 4 represents the mediating role of employee training (MTR) between the Internet of Things (MIOT) and operational performance (comprised of quality improvement (MQI), manufacturing flexibility (MMF), increase in productivity (MP), and cost reduction (MCR).
According to Table 10, hypothesis 4 is accepted because employee training (MTR) is playing a mediating role between Internet of Things (MIoT) and quality improvement (MQI) with values (b = .6811, p < .05, S.E. = 0. 0747, t = 9.1211 ); internet of things (MIoT) and manufacturing flexibility (MMF) with values (b = .3880, p < .05, S.E. = 0. 1210, t = 3.205 ); internet of things (MIoT) and increased production with values (b = .7719, p < .05, S.E = 0. 1306, t = 5.9108 ); internet of things (MIoT) and cost reduction (MCR) with values (b = 1.0285, p < .05, S.E. = 0.1351, t = 7.6121). Similarly, positive values for LLCI, ULCI, and the effect of IV on DV have been observed, indicating the acceptance of hypothesis 4.
Mediation Analysis.
Discussion
This research study aimed to analyse the impact of Industry 4.0 technologies, particularly big data (BD) and the internet of things (IoT) on operational performance (OP) within the manufacturing sector of developing economies, with an emphasis on the mediating role of employee training (ET). The findings suggested that IoT and BD improved productivity, flexibility, cost reduction, and product quality, and the improvements were noticeably larger when applied together, with an effective training programme. There was also some evidence that the relationship between technology adoption and performance was mediated by employee training, suggesting that workers needed to be prepared and digitally competent for organisations to realise the technological improvements.
The main significance of the study is that it addresses a current and urgent need for manufacturing firms to adapt and stay competitive in the new paradigm of digital transformation. Understanding the impact of Industry 4.0 technologies on performance will enable policy-making, prudent investment decisions, and workforce development in emerging economies for digital transformation.
The results suggest that big data improves efficiencies, predictive maintenance, and decision making. The business firms utilising big data are able to use real-time data driven insights to improve performance and have a better understanding of production metrics. This finding confirms the work of Wamba et al. (2020) and Akter et al. (2017), who explained the operational flexibility and proactive management instigated by the use of big data. Current study elaborates that big data is pertinent as a performance-metric tool within digital transformation as it provides a good basis for industry processes to become smarter and more flexible.
Similarly, this research shows how IoT enables an interconnected system that provides real-time visibility and collaboration during the production process. This supports Tortorella et al. (2023) and Ayaz et al.’s (2019) research which identified IoT adoption enables companies to improve operational control, response, and resource utilisation. The concept of IoT as a driver of operational improvement has been addressed as well, especially in terms of cost and quality management (Kortmann, 2020).
The research recognises tech-savvy and technology's role in performance improvements in practice with training employees. Employees with adequate technical and analytical skills are able to deploy digital technology and interpret the analysis of data. These findings further support the observation of Agostini and Filippini (2019), as well as Sony and Naik (2020) that suggested human preparedness and learning capacity were key facilitators within digital transformation. The sequential relationship of technology, training, and performance improvements is more specific in that the combination of human resources and technology leads to operational excellence.
This study corroborates previous studies conducted by Imran et al. (2018) and Kazmi and Abbas (2020), which focused on performance outcomes of IoT and big data implementation in the industrial sector. Nevertheless, this study now adds a new aspect of employee development role as a mediator, unlike the previous studies which were limited in exploring the influence of digital technologies on operational performance. It also builds upon the findings of Bag et al. (2023) and Tortorella et al. (2023), who suggested the need to explore the human element that drives digital revolution. This study now takes a human centred view on the relationship of technology on performance through introduction of employee development, illustrating that developing the capabilities of the workforce is critical to optimising the value of adopting Industry 4.0.
Additionally, the study also extends the RBV theory (Barney, 1991) in demonstrating that information technologies such as big data and IoT can become strategically valuable resources when complemented with an educated and well-trained workforce. Moreover, it is in line with Human Capital Theory (Becker, 1964), which posits that investing in people results in enhancements to human capabilities and allows organisations to capture even larger returns from advances in technology.
Additionally, drawing from the Socio Technical Systems Theory (Bostrom & Heinen, 1977) shows that attention to the human aspects of the system, at the same time, will allow successful technology implementations. Therefore, the conceptual model presented in this paper contributes a richer understanding of digital transformation, particularly as it pertains to emergent industrial contexts.
According to this study, it is necessary to integrate workforce development with technology deployment upon implementation of Industry 4.0 technologies. The supervisors are tasked to invest in training programmes that will develop the competencies of employees to effectively deploy digital tools in their respective workplaces and to positively impact overall operational effectiveness. The implications of the findings point to the need for programmes that support the effects of employee experiences on developing digital skills in manufacturing industries; it sets calls to action for policymakers. The support of these programmes could enhance productivity in industry and increase the rate of adoption of newer technologies.
The study also provides evidence asking to adjust our education systems, to leverage the advancement associated with Industry 4.0. Curriculums in Universities and training organisations should engage big data analytics, IoT technologies, and data-oriented management models. Interdisciplinary programmes, with a focus on engineering, computer science and information science programmes alongside management programmes, would position professional to undertake leadership roles in advancing digital initiatives. Practical decisions such as on job training, internship and certification should strengthen learning to bridge the gap between theory and industry practice.
The findings suggest the need to balance human capital development and technology improvements in developing economies. This balance is often exacerbated by the lack of inadequate infrastructure as well as lack of digital skills. Combination of programmes to train workers and initiatives to utilise big data and IoT would expedite advancements in industry and its competitiveness. Collaborative efforts between policymakers and industry leaders are essential to advance digital innovation while strengthening workforce readiness.
In conclusion, facilitating enhanced digital transformation will entail a balance between technological progress and worker preparedness. Policymakers, educators, and managers, can promote digital transformation in industrial sectors in developing economies by utilising the empirical explanation from reputable theories and empirical evidence from research presented in this study.
Conclusion and Future Directions
This research investigated the impact of Industry 4.0 elements on operational performance in the manufacturing sector. The analysis suggests that the collective use of big data and IoT devices with mediating role of employee training can increase operational performance via quality, flexibility, productivity and reducing costs. The results support the resource-based view (RBV) concept that sustainable competitive advantage is developed through the successful integration of technical resources and human capabilities.
This study supports both theoretical and practical visions of emerging economies within the context of the digitalisation initiative. It reaffirms that technology alone does not ensure sustainable improvement in performance; the development of human capital through explicit training and investment is imperative. As time progresses, advanced technologies and skilled people are promoting innovation, operational efficiencies, and organisational resilience. These implications are especially relevant for manufacturers operating in resource-constrained environments, where operational skills and digital literacy are two important dimensions of competitiveness in many industrial sectors.
The proposed model offers industry practitioners a practical foundation upon which they can assess the performance benefits associated with various Industry 4.0 technologies. It can support managers in making decision about investment priorities and developing training programmes to be effective in enhancing the digital transformation results. Even the findings point to the need to align industrial policy development with educational and vocational training reforms. National policies could promote technology use, skills development and public-private collaboration to accelerate industrial productivity and sustainable economic development.
In addition, this study has contributed significantly to research, it is also important to note that it has limitations. The data collection strategies were impeded by contextual factors that created limitations such as time constraints, funding constraints and limited access to respondents. The analysis focused entirely on the manufacturing sector, which may confine the applicability of results to other industries. The cross-sectional design limits causal explanation, a longitudinal study is required to examine performance variations over time following Industry 4.0 adoption. The research was conducted in Pakistan, a developing economy with distinct regional, institutional and infrastructure conditions, so the findings may not fully generalise other workforce dynamics and maturity levels.
Future research should include Artificial Intelligence (AI), Augmented Reality (AR), or Blockchain to expand the scope of Industry 4.0 technologies. Investigating mediating or moderating variables such as leadership, organisational culture, and digital readiness may enrich a more comprehensive conceptualisation of the technology performance relationship, as well as its social and organisational consequences.
This study provides theoretical and practical contributions by offering a human-centred framework to Industry 4.0 adoption showing the mutual role of technology and employee training in improving operational performance, and that technology improvements alone are insufficient without simultaneous human capital enhancements. The results suggest a need for balance in the digital transformation strategy to align technology innovation with human capital enhancement, to support the organisation's culture, while also considering the institutional basis.
This study sets the foundation for future research which continues to develop more holistic investigations of digital transformation. This study also encourages future researchers and practitioners to consider multi-focused approaches to combine technology systems, people, and organisational systems for industrial advancement in Industry 4.0 context.
Footnotes
Ethical Considerations
This study was approved by the Ethics Review Committee for Human Participants at the Department of Engineering Management, National University of Sciences and Technology (NUST), Islamabad, Pakistan.
Ethics Committee Reference Number: NUST/CEME/DEM/154
Consent to Participate
Prior to participation, all respondents were provided with an informed consent form detailing the study's purpose, its voluntary nature, the assurance of confidentiality, and their right to withdraw at any time.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
