Boosting Innovation Performance through Big Data Analytics Powered by Artificial Intelligence Use: An Empirical Exploration of the Role of Strategic Agility and Market Turbulence

Abstract

Scholars and practitioners have paid attention to the critical role of big data analytics driven by AI in enhancing business performance. However, firms investing in big data analytics often fail to achieve those advantages. Our research explores the critical role played by strategic agility and market turbulence on the link between big data analytics driven by AI and innovation performance. Based on dynamic capabilities view, we developed an integrated model to examine the relationship between our study variables. We utilized a quantitative approach to collect data from manufacturing companies in Saudi Arabia. We employed structural equation modelling (SEM) through AMOS 26.0 to analyze and test the study hypotheses. Our findings indicated that big data analytics driven by AI have a significant impact on strategic agility and innovation performance. It also revealed that strategic agility mediates the relationship between big data analytics driven by AI and innovation performance. The results also showed that higher levels of market turbulence are associated with more robust connections between big data analytics, strategic agility, and innovation performance. Our work provides managers with meaningful implications on the value that organizations can create through the use of big data analytics and strategic agility.

1. Introduction

Data are among the most prized possessions in today’s business world [1,2]. Moreover, when businesses go digital, they generate a great deal of data in their supply chains [3]. In contrast to capital, however, large amounts of data are useless without the means to mine them for actionable insights [4,5]. Managers who are intimately familiar with their data are in the best position to use them to establish company-wide standards [6,7,8]. Costs can be cut [9,10], production times shortened [11], and new products or services developed to satisfy consumers’ evolving wants and needs [12]. The ability to conduct predictive analytics on large amounts of data will in the future be a key factor in the continued digitization of the supply chain [13,14]. Companies are investing in new technologies associated with management applications like big data analytics (BDA), machine learning (ML), and artificial intelligence (AI) to gain a competitive edge [15].

Prior research argues that big data analytics represents a new frontier for fostering innovation, competition, and productivity [2]. This has resulted in a lot of focus, from both researchers and businesses, on the benefits that big data analytics may provide to businesses [3,6]. Literature generally agrees that “big data analytics” refers to “a new generation of technologies and architectures, designed to economically extract value from very large volumes of a wide variety of data by enabling high velocity capture, discovery, and/or analysis” [4]. Despite the fact that most claims on the value of big data analytics are anecdotal, the few empirical research studies in the area have documented a positive relationship between the decision to invest in the firm-wide deployment of big data analytics and performance [2]. Firms can make sense of massive amounts of data, obtain crucial knowledge, and adjust their strategy in response to changes in the market thanks to the use of big data analytics [5]. Therefore, the most important contribution of big data analytics is that it allows for less biased and more evidence-based decision making [6].

Due to its unique characteristics and ontology, artificial intelligence (AI) is causing significant shifts in the methodology and results of digital innovation [16,17]. AI’s creations are engendered as fundamentally distinct from those engendered by more conventional information technologies [18]. Competitors are finding new ways to create value and differentiate themselves because of the fluidity and complexity of digital innovation processes and outcomes [19]. This motivates a reconsideration of the roles that actors, organizations, AI, and potential courses of action can play in the pursuit of innovation. There is a growing body of research suggesting that AI can aid and speed up labor-intensive information processes in HRM [20], evaluate candidates using the same criteria consistently [17], make fairer and less biased decisions than human intuition [16], and promote diversity in organizations [21,22].

In order to maintain a competitive edge in the face of rising customer demands, stiff global rivalry, and a constantly evolving technological landscape, businesses must boost their entrepreneurial image [9]. In addition, market-based policies are quickly being adopted by emerging economies in an effort to boost economic growth and reduce poverty [15]. Businesses in such economies must adapt quickly to new market conditions, heightened environmental unpredictability, and uneven growth [19]. Prior research revealed that “supply chains are multi-structural semantics and often have interrelated structures” (i.e., organizational, functional, informational, financial, topological, technological, product, and energy structures), all of which are in constant flux due to both predetermined and unforeseen factors [23]. Businesses in underdeveloped economies are still dubious about the applicability and potential benefits of big data analytics powered by artificial intelligence (BDA-AI), despite the excitement produced by the potential of BDA-AI among organizations and academia [24]. This pessimism may be due to a number of issues, including a lack of top-level buy-in, an inaccurate assessment of the competitive landscape, a failure to meet customers’ urgent needs, a failure to differentiate the product, and inefficient marketing [11,25].

Companies of varied sizes and across a wide range of industries, such as Honda, Walmart, Samsung, and many more, engage in substantial activities that are receptive to big data technologies. These describe a new class of technologies and modes of construction that are organized to extract monetary value from using massive quantities of diverse data [9,17]. This is similar to the argument made by a prior study: although data are certainly valuable, obtaining this value requires a thorough and in-depth comprehension of the digital environment [13]. Several transactions and activities occur between businesses; thus, it stands to reason that this topic would be of immense use to many different types of organization. This area of study is clearly gaining more and more scholarly attention in the current literature [3,16]. This strongly supports the claim made by a previous examination, which associates it with the sales of novel products [26]. Previous exploration argues that the increased quality attained by using big data in retailing is more important than the sheer increase in data volume for improved outcomes [27]. Unfortunately, insufficient research has examined the role played by these big data analytics powered by artificial intelligence in creating strategic agility and innovation performance.

As a contribution to the previous studies, we drew key insights from the resource-based view (RBV) [28], connecting it with the RBV of big data and DC theories [29,30], to explain how increasing the use of big data analytics powered by AI can boost strategic agility and innovation performance. The extent to which strategic agility and innovation performance are influenced by market turbulence was also investigated. Thus, we developed and empirically tested an integrated model to address the following questions: (1) what is the impact of big data analytics powered by artificial intelligence on strategic agility and innovation performance?; (2) does strategic agility mediate the relationship between big data analytics powered by artificial intelligence use and innovation performance?; and (3) does market turbulence moderate the relationships between big data analytics powered by artificial intelligence use, strategic agility, and innovation performance?

High-quality strategic, operational, and tactical decisions are essential for businesses to maintain a competitive edge in today’s fast-paced market [15]. Professionals in the business world have recently become interested in using big data for decision making in fuzzier settings [5]. However, because data-driven manufacturing is so reliant on the arrangement of material resources and the cultivation of labor capabilities for sustainability [15], knowledge of big data analytics (BDA) and artificial intelligence (AI) has become a tacit resource in the modern day because of its successful implementation is dependent on the capabilities of the workforce. Programming and data analytics know-how are two examples of such abilities. Because they require a team of workers to work together with a small number of experts who have thorough expertise to explain the complete process, these resources can be classified as socially complex. A prior study revealed that the execution of a BDA-AI project requires both basic resources (funds) and tangible resources (big data management infrastructure, technology resources, Hadoop Pfor data processing, data visualization tools, and cloud-based services) [15]. We argue that big data analytics enabled by AI is a critical factor in enhancing innovation performance. Further, we argue that strategic agility plays a critical role on the influence of big data analytics powered by artificial intelligence on innovation performance.

Our paper offers implications for theory and practice. First, this examination seeks to answer calls for more examinations on big data analytics and its effects on firm performance [15]. Therefore, our study explores the impact of BDA on innovation performance. Second, our examination seeks to answer the call for more research on the mechanism on how big data analytics affects business performance [8]. Our paper explores the influence of big data analytics on innovation performance through strategic agility. Third, our study contributes to big data analytics literature by exploring the direct effects of BDA and indirect effects through strategic agility.

2. Theoretical Underpinning and Hypotheses Development

2.1. Research Context

Big data analytics powered by artificial intelligence (AI) has been gaining momentum in Saudi Arabia in recent years, driven by the country’s Vision 2030 initiative, which aims to diversify the economy and promote innovation and technology adoption. The Saudi government has been actively promoting the adoption of AI and big data analytics as part of its Vision 2030 program. The National Data Management Office (NDMO) was established to oversee data-related activities and initiatives. The goal is to leverage data analytics and AI to improve various sectors, including healthcare, finance, education, and public services.

The rapid transformation of Saudi Arabia’s industries is being driven by the increasing availability of data-driven AI solutions and the benefits of machine learning. High-speed internet and 5G deployment, data access, and security are all part of Saudi Arabia’s efforts to create an AI-friendly ecosystem. Plans for rolling out 5G services in Saudi Arabia were advanced in March 2022 by the country’s Communications and Information Technology Commission (CITC). The High-Altitude Platform System, which provides 5G network coverage from the stratosphere, was tested successfully by the CITC in conjunction with the UK company Stratospheric Platforms Limited. In order to fulfil the goals outlined in Vision 2030, the municipalities of the KSA (Kingdom of Saudi Arabia) will have to use municipal data as a national asset. It also encourages the early adoption of artificial intelligence (AI) and big data (Big Data) concepts and solutions via several smart city programs like NEOM, which helps to catalyze innovative solutions. In addition, Saudi Arabia’s Vision 2030 includes an artificial intelligence policy. The government has ambitious plans to modernize metropolitan areas with cutting-edge infrastructure and environmentally-friendly public services. The government of Saudi Arabia also hoped that AI would help it automate and streamline its operations. There are, however, potential threats to the market’s expansion from issues, such as difficulties in adoption and operations and a lack of available AI developers and data scientists. The limits put in place because of the epidemic have also inspired new ideas, such as the Tawakkalna app, which is meant to aid government attempts to combat COVID-19. It makes it easier for government and commercial sector workers to obtain electronic movement permits during curfews. The creation of such apps made the establishment of efficient data collection methods possible, which were then examined with the use of machine learning and a thorough data analysis to the benefit of citizens.

Foreign Direct Investment (FDI) and cross-border data flows have become increasingly intertwined in the digital age, offering both significant opportunities and potential pitfalls for countries and businesses. Digital FDI and data flows can stimulate economic growth. They enable foreign companies to invest in digital infrastructure, technology, and local businesses, creating jobs and boosting productivity [31]. Cross-border data flows facilitate the exchange of information and ideas, fostering innovation and driving technological advancement. This is particularly important in industries like tech, healthcare, and finance [32]. Cross-border data flows provide access to valuable consumer data and insights, which can be used for better market understanding, targeted marketing, and product development [33]. Moreover, cross-border data flows are critical for the functioning of global supply chains. They enable real-time information exchange and coordination among suppliers and manufacturers.

Some countries, such as Saudi Arabia, impose data localization requirements, mandating that data be stored within their borders. This can increase costs and complicate cross-border operations. Conflicts over data sovereignty, where countries assert control over data generated within their borders, can create tensions in international relations. Digital FDI can sometimes result in the dominance of a few global tech giants, limiting competition and innovation in certain markets [34]. In conclusion, digital FDI and cross-border data flows offer immense potential for economic growth and innovation [35]. However, they also come with significant challenges related to privacy, security, regulation, and access. Striking the right balance between promoting these flows and safeguarding national interests remains a complex task for governments and businesses alike [36]. Effective international cooperation and the development of common standards and regulations can help address some of these challenges.

2.2. Resource-Based View and Dynamic Capabilities View

Research on big data analytics capabilities (BDACs) is a flourishing field, to which the RBV has made important contributions [37]. However, it is not widely understood how businesses use big data analytics to expand their innovation capacity [38]. The term ‘big data’ refers to massive, composite, real-time datasets that demand sophisticated tools for management, analysis, and processing in order to extract useful insights [39,40]. BDACs are defined as “human resources consisting of managerial and technical big data skills and tangible resources (that) include data, technology, and other basic resources (such as time and investments)” [41].

Prior research pointed out that a company’s RBV is defined by how well its technological and human resources are used to generate value [42]. The RBV has been used in most studies to analyze the impact of BDACs on company performance [41]. Scholars have warned that resources might not be enough to generate value for businesses, as these organizations also need the skills to efficiently deploy and use their assets [41,43]. IT embeddedness in dynamic capabilities (DCs) has also received much attention in recent research [44]. Sensing, capturing, and morphing capacities are all part of the DCs [45]. In this case, we explore how the application of big data analytics supported by AI might affect both strategic agility and innovation performance.

Sustainable supply chain management, and the study of its dynamic capacities, is a relatively new and developing discipline. Recent research [46] found that further qualitative and quantitative investigation into the relationship between business sustainability and dynamic skills is warranted. Formal models are rarely applied to this area of study. Prior research used a case study approach to research monitoring, cooperation, and innovation as critical social management capacities for managing social issues and related risks in supply chains [47]. Big data analytics enables manufacturers to collect and analyze vast amounts of data from various sources within their operations. This data can include production metrics, energy consumption, supply chain information, and more. By leveraging analytics tools and techniques, manufacturers can gain valuable insights into their processes, identify inefficiencies, and make data-driven decisions to optimize their operations for sustainability. Manufacturers can use big data analytics to identify opportunities for developing sustainable products and processes. This can lead to the creation of eco-friendly products that meet the consumer demand for sustainability. Across the three pillars of the triple bottom line of sustainability, a previous study empirically evaluates the mediating function of dynamic capacities development for an improved sustainable performance [48]. They discover that collaborative culture and share that planning activities sparks dynamic skills, which in turn have a significant impact on long-term success. Lastly, businesses must realize that adapting to the changing needs of their stakeholders will necessitate new methods of operation and the development of new skills [49]. Therefore, the integration of big data analytics and strategic agility into manufacturing operations can lead to the development and implementation of sustainable practices. These practices not only benefit the environment but also contribute to cost savings, operational efficiency, and improved competitiveness for manufacturing companies. As sustainability continues to be a critical factor in business success, the adoption of data-driven strategies becomes increasingly important in the manufacturing sector.

The dynamic capabilities theory (DCs) refers to a company’s ability to develop new skills and reorganize its existing ones [29,41] as an extension of the RBV. DCs refer to the application and distribution of resources that capitalize on sensing, seizing, and reconfiguring capacities [40]. Thus, in contemporary contexts, DCs are defined as the “capacity to promptly adopt changes and process data and information for actionable knowledge or analytics that enable the effective tackling of changes in the market” [50]. Because capabilities are achieved through the application of resources [51], and organizations build sustainable competitive advantages on the basis of hard-to-imitate resources and capabilities, the RBV and DCs are complementary to one another despite their divergent historical roots. Nonetheless, despite their significance, relatively few studies have looked at the use of BDA-AI in the context of SMEs from both of these angles. Agility was conceived as a DC [52]. According to previous research, organizations that exhibit agility are better able to adjust to external changes and create value as a result [53]. Dynamic capabilities (DCs) are an organization’s “ability to integrate, build, and reconfigure internal and external resources/competencies to address, and possibly shape, rapidly changing business environments” [40]. “If a firm possesses resources/competencies but lacks DC, it has a chance to make a competitive return for a short period, but superior returns cannot be sustained” [29]. The DC view considers an organization’s resource renewal capabilities in light of external developments [40]. A number of internal and external characteristics that make up agility may be linked to DCs, in that they are developed over time [54,55]. Thus, strategic agility is measured by an organization’s propensity to foster an internal environment wherein its people can acquire and hone the knowledge and abilities necessary to respond to shifts in their external environment. Figure 1 demonstrates the conceptual framework of our study.

2.3. The Relationship between Big Data Analytics Powered by Artificial Intelligence, Strategic Agility, and Innovation Performance

According to prior research, “the value of competitive advantage does not lie in the capabilities themselves, but in the way the resources and capabilities are exploited” [56]. This means that functionally similar dynamic capabilities (such as similar BDA technologies powered by AI that can be acquired on the open market) are likely to be common [57]. “Using dynamic capabilities sooner, more astutely, or more fortuitously than the competition to create resource configurations that have the advantage” [51] is, hence, a possible source of a long-term competitive advantage. Despite the rising popularity of emerging technologies like big data analytics and artificial intelligence, it is unclear how the use of business analytics affects corporate performance [58]. In contrast, “the value of competitive advantage does not lie in the capabilities themselves, but in the way the resources and capabilities are exploited” [59]. “Using dynamic capabilities sooner, more astutely, or more fortuitously than the competition to create resource configurations that have the advantage” is, hence, a possible source of a long-term competitive advantage [60].

The impact of business analytics and artificial intelligence on business performance through business processes has been explored in previous studies [61,62]. We propose that using BDA enabled by AI assists the development of a business’s information processing capabilities, in line with dynamic capabilities logic [29,40]. Managers can use this synthesized information to reduce uncertainties in demand, capacity, and supply because it allows them to analyze and incorporate complicated information gathered from diverse sources [63]. Organizations’ profit margins are impacted when they lack such skills and are forced to keep a large stockpile or invest in a more flexible supply chain design [64]. We also argue that BDA-AI can help businesses align with their partners and adapt to changing market conditions by providing new insights into how to best allocate their resources [15,41].

Prior research revealed that big data have the potential to alter the innovation environment by better matching consumer tastes with product characteristics [51]. According to the organizational learning theory, businesses can improve their capacity for learning by using newly available information to develop more precise models of the relationships between strategic decisions and bottom-line results [65]. Decisions based on big data are replacing laboratory-based consumer research and intuition [40]. This means that big data can improve innovation performances by allowing firms to develop and deploy resources more efficiently and effectively. The adoption of BDA-AI may have far-reaching consequences, one of which is an improved strategic agility and innovation performance. Therefore, we postulate:

H1: 

Big data analytics powered by artificial intelligence use have a significant influence on innovation performance.

H2: 

Big data analytics powered by artificial intelligence use have a significant influence on strategic agility.

2.4. The Role of Strategic Agility

A crucial characteristic for achieving agility is the ability to change the configuration of processes with little to no additional investment in capital [66]. Supply chain process changes both internally and externally and the ability to “sense” and “respond” to consumers’ wants are essential for meeting a wide variety of customers’ demands [67]. This aspect of capability is in keeping with process innovation through teamwork. Most small and medium-sized enterprises (SMEs) in the manufacturing sector are suppliers whose main focus is on catering to industrial clients. For suppliers to be responsive, they must have a good working connection with their customer. As a result, a supplier focused on agility should seek out chances for customization, where products and processes are designed and altered in response to customers’ stipulations [46].

Agility in marketing helps fuel innovation, for a number of reasons. To begin, agile businesses actively identify latent and developing client requirements, giving them a leg up on the competition by discovering and capitalizing on previously unanticipated openings [68,69]. Second, companies that have demonstrated a high level of marketing agility tend to be more dedicated to experimentation. Cultural growth that encourages innovation is fostered by adaptable approaches to resource management and production [70]. Third, as a dynamic capacity, marketing agility helps with the change, reconfiguration, and renewal of processes and encourages innovation to better suit the environment.

A common subject in the business literature is the empirical connection between strategic decision implementation and performance [71]. Most research on agility has been theoretical, but the link between agility and endogenous performance indicators has been empirically tested, with varied results. According to a straightforward correlation analysis, value chain agility correlates positively with return on asset and negatively with time-to-market [72]. According to a previous examination, a company’s agility can increase its profitability [73]. Their research into the links between IT alignment, agility, and performance demonstrates that the latter has a favorable impact on the former, with the benefit being reduced by the presence of environmental volatility [74]. When examining the relationship between manufacturing agility and financial performance, however, previous studies find no meaningful correlation [75]. Using metrics from the fields of marketing and finance, a previous study investigated whether or not there is a correlation between a company’s responsiveness to customers (its agility) and its overall performance [76]. The customer sensing capability is found to positively impact business performance, but the customer responding capability is found to have no such effect. Given the contradictory findings in the literature, we propose to investigate the link between the use of big data analytics supported by AI and the performance of innovation through the lens of strategic agility. Thus, we suggest the following:

H3: 

Strategic agility has a significant influence on innovation performance.

H4: 

Strategic agility mediates the link between big data analytics powered by artificial intelligence use and innovation performance.

2.5. The Moderating Role of Market Turbulence

Dynamic capabilities, which contribute to change, are said to be more valuable in such a setting, and this value increases in disturbed contexts [77]. The ability to adapt to change and gain a competitive edge is what we mean when we talk about agility. Contradictory scenarios are also presented in the literature. Dynamic capabilities become experiential and are weakly associated with performance [78], although some scholars claim that in extremely unstable situations, it is difficult to forecast future developments, revealing why enterprises rely on external knowledge [79]. In line with other research [80], we find that more marketing-savvy businesses are better able to quickly assimilate new information and put it to use. As a result, performance improves.

The primary function of dynamic capabilities is to refresh and reorganize static ones. Traders rely on them more frequently in unstable markets [81,82]. Previous empirical research demonstrates that firms operating in highly chaotic environments benefit more strongly from the favorable benefits of dynamic skills on ordinary capabilities [83]. That is, participating in frequent sensing and promptly responding to new information are crucial tactics when a market is very volatile because of the increased opportunities and potential for increased capability. There may be enough upsides to investing in the marketing capacity to justify the expense [41,84]. If businesses want to keep their normal capabilities aligned with the external environment in a turbulent setting [41], they need access to timely, relevant information. In order for businesses to adapt to their surroundings and reorganize their capacity for innovation, marketing agility is crucial. One of the primary drivers of long-term success, especially in dynamic markets, is innovation because it allows businesses to adapt to changing external conditions [85]. Indeed, greater turmoil means dangers for businesses but also greater opportunities. A strong innovation capability allows businesses to grasp and serve consumers’ growing and new needs as the market shifts and new opportunities present themselves. However, if businesses fail to adapt to shifting consumer preferences, they risk seeing their market share rapidly eroded. Therefore, we propose the following hypotheses:

H5: 

Market turbulence moderates the relationships between big data analytics powered by artificial intelligence and strategic agility.

H6: 

Market turbulence moderates the relationships between strategic agility and innovation performance.

3. Methods

3.1. Sampling and Data Collection

We asked a reputable Saudi Arabian market list company with over sixty thousand manufacturing companies in its database to recruit our qualified respondents. This consulting firm offered us a sample of the 1500 businesses from all sectors recorded in its database. We distributed our cross-sectional survey to 1500 manufacturing firms picked at random from a database of such firms. Because of the varied backgrounds of our respondents, we were able to collect an accurate representation of society. Respondents were members of the selected companies’ highest management teams, including “the Board of Directors, CEO, CTO, COO, Executive General Manager and General Manager”. We received 631 viable responses out of a total of 1500 surveys (a response rate of 42%).

Table 1. Participant demographics.

Details	Category	Frequency	Percentage %
Designation	General manager Senior manager Manager Junior manager	123 309 108 91	20 49 17 14
Experience (Years)	Above 20 10–20 Below 10	311 204 116	49 32 19
Firm’s Industry	Pharmaceuticals Electrical equipment Automotive components Machinery and industrial equipment Food Chemicals Pulp and paper Consumer goods	69 102 131 60 106 37 26 100	11 16 21 9 17 6 4 16
Age of the Firm (Years)	>20 15–20 10–14 5–9 Below 5	127 109 121 162 112	20 17 19 26 18

presents the respondent demographics (at the company level): “Pharmaceuticals” (19.62%), “electrical equipment” (16.42%), “automotive components” (16.34%), “machinery and industry equipment” (12.45%), “food” (11.26%), “chemicals” (9.85%), “pulp and paper” (8.15%), and “consumer goods” (5.91%).

Following the recommendations of Armstrong and Overton [86], we checked for a non-response bias. Using a t-test, we compared first-responders to all respondents. No statistically significant differences were found between the two groups on any of the survey items (p > 0.05). Second, we looked for a performance bias by using paired sample t-tests to compare the companies’ returns on assets to the median returns on assets in their respective industries. The p-value for the lack of significance was greater than 0.05. These two findings indicate that the non-response bias is not a major issue.

3.2. Conceptualization of Measures

Our study variables were measured using a valid and reliable scale adopted from previous studies.

Dependent construct—“innovation performance” refers to how well an organization creates value through new ideas for products, services, processes, markets, and production techniques. The six-item innovation performance evaluation was created by prior research [87]: for instance, agreeing that “new products or services have been successfully developed or introduced by our firm”.

Independent construct—“big data analytics powered by artificial intelligence use” refers to the use of big data analytics powered by artificial intelligence to extract more meaningful information with which organizations can improve their decision-making skills. It was measured using ten items adopted from prior research [15]: for instance, “We routinely use data visualization techniques to assist users or decision makers to understand complex information”.

Mediator construct—“strategic agility” was measured using eight items adopted from previous research [88]. For instance, “How easily and quickly can your firm perform the following actions?” (e.g., 1. “respond to changes in aggregate consumer demand”; 2. “customize a product or service to suit an individual customer”).

Moderator construct—“market turbulence” was measured using three items adopted from a prior study [54]: for instance, “In our markets, customer preferences change quickly”.

3.3. Common Method Bias

Due to the potential for common method bias (CMB) in survey-based cross-sectional data [88], we adhered to stringent measures to reduce the impact of this drawback. We began by conducting the standard one-factor Harman’s test (this component alone explained roughly 23.8 percent of the overall variance). However, there are many writers on management who argue that Harman’s one-factor test is insufficient and should not be taken as proof. Therefore, we implemented the second method, the correlation marker methodology, proposed by a previous study [89]. To help filter out CMB-induced correlations, we included a new independent variable. Following the advice of the same previous study [90], we also derived the statistically significant correlation values. The correlations, both adjusted and uncorrected, are very close to one another. Therefore, we assume from these statistical findings that CMB does not significantly affect our remaining outcomes.

We calculated the “nonlinear bivariate causality direction ratio” (NLBCDR) following the advice of a previous study [91]. According to prior research, “the NLBCDR measures the extent to which bivariate nonlinear coefficients of association provide support for the hypothesized directions of the causal links in the proposed theoretical model” [92]. A NLBCDR of 0.93 was found, which is above the minimum required value of 0.7. This justifies our claim that the topic of causality is irrelevant. In addition, we present the corresponding values for the model fit and quality indices (see, for example, average R2 = 0.631; Tenenhaus GoF = 0.708) that underpin this conclusion.

4. Analysis and Results

AMOS 26.0 was used to conduct a structural equation modelling (SEM) analysis using a maximum likelihood estimation. We used an SEM strategy with two phases [93]. In the first, we used the CFA (“confirmation factor analysis”) to verify the reliability of the measurement scheme. Model fit was evaluated using several different indices, such as the “normed chi-square” (χ2/DF), “root mean square error of approximation” (RMSEA), “standardized root-mean square residual” (SRMR), “normed fit index” (NFI), and “comparative fit index” (CFI) [80]. Second, we used a structural model to verify the hypothesized relationships between the study variables.

4.1. Measurement Model

Model statistics for the measurements are reported in

Table 2. Measurement statistics of construct scales.

Construct/Indicator	Standard Loading	CR	VIF	Cronbach’s α	AVE	Mean	SD	t-Statistic	Skewness	Kurtosis
Innovation Performance (INN) INN1 INN2 INN3 INN4 INN5 INN6	0.97 0.95 0.98 0.93 0.96 0.92	0.96	1.479	0.95	0.570	2.56 3.09 2.97 3.01 2.36 2.17	0.816 0.829 0.805 0.814 0.890 0.841	21.09 26.08 17.09 31.09 23.15 9.54	−1.890 −2.806 −2.136 −1.783 −2.085 −1.567	1.457 1.051 2.267 1.985 2.078 2.160
Strategic agility (AGT) AGT1 AGT2 AGT3 AGT4 AGT5 AGT6 AGT7 AGT8	0.97 0.94 0.95 0.95 0.90 0.95 0.97 0.96	0.94	1.780	0.92	0.608	2.70 3.08 3.27 2.75 2.90 2.46 2.10 3.07	0.84 0.83 0.86 0.83 0.87 0.85 0.82 0.86	21.08 17.34 24.16 14.09 22.06 23.19 26.97 18.93	−2.06 −1.64 −1.88 −2.34 −1.65 −2.06 −2.11 −2.01	2.76 2.90 1.45 1.23 1.66 2.02 2.15 1.78
Big data analytics-AI BDA-AI1 BDA-AI2 BDA-AI3 BDA-AI4 BDA-AI5 BDA-AI6 BDA-AI7 BDA-AI8 BDA-AI9 BDA-AI10	0.95 0.98 0.93 0.95 0.94 0.93 0.96 0.92 0.96 0.97	0.97	1.309	0.95	0.509	2.28 3.09 2.21 2.47 2.12 2.30 2.24 2.05 2.22 2.19	0.84 0.83 0.80 0.81 0.83 0.85 0.82 0.81 0.79 0.82	12.09 21.02 14.36 18.30 23.16 22.57 25.12 19.45 26.16 21.57	−1.67 −2.02 −1.21 −1.24 −1.25 −1.20 −1.15 −1.27 −1.06 −1.25	1.05 1.28 1.31 2.14 2.02 2.16 2.07 2.18 1.69 2.56
Market turbulence (MRT) MRT1 MRT2 MART3	0.93 0.96 0.92					2.39 3.05 2.15	0.86 0.84 0.86	24.48 19.30 17.23	−1.57 −1.70 −2.12	1.59 1.16 2.17

: X2/DF = 429.16/362 = 1.186, RMSEA = 0.04, SRMR = 0.05, NFI = 0.94, and CFI = 0.98. In addition, we checked the model’s accuracy by gauging the validity and reliability of each factor. Cronbach’s alpha (CA) and composite reliability (CR) were first used to assess reliability. We inferred construct reliability because all values for CA and CR were more than 0.70 [94]. Standardized factor loadings were used to assess convergent validity. Convergent validity was established by the fact that all factor loadings were statistically significant and greater than 0.70 [95]. We also used AVE to evaluate convergent validity at the construct level (see

Table 3. Discriminant validity of the correlations between constructs.

Constructs	Correlations and Square Roots of AVE
	INN	AGT	BDA-AI	MRT
INN	0.755 a
AGT	0.408 b	0.779
BAD-AI	0.349	0.399	0.713
MRT	0.416	0.401	0.560	0.833

a Composite reliability is along the diagonal, b Correlations.

). This was greater than 0.50 across the board [95], indicating that the factors accounted for more than 50% of the variation across their items and confirming convergent validity [94]. Finally, we compared the correlation between the latent factors and the square root of the AVE to determine discriminant validity [94]. The requirement for discriminant validity was met when we compared the square root of AVE to the inter-factor correlation and discovered that the former was larger.

4.2. Structural Model

We used the SEM models with a 5000 bootstrap to test the proposed hypotheses. Model fit statistics for the full structural model were as follows: X2/DF = 1.276, RMSEA = 0.05, SRMR = 0.04, NFI = 0.95, and CFI = 0.97. Given the high degree of correlation between several variables, we used “variance inflation factors” (VIFs) to look for evidence of a multicollinearity problem. When compared to the most stringent cutoff value of 3.00 [93], the highest VIF value recorded was 1.48. Each path in the research model’s standard parameter estimations and significance levels is listed in

Table 4. Results of testing hypotheses.

Path Directions	β	t	Support
BDA → INN	0.62 ***	21.720	Yes
BDA → AGT	0.43 ***	11.218	Yes
AGT → INN	0.59 ***	19.267	Yes
Model fit indices	X2/DF = 1.276, RMSEA = 0.05, SRMR = 0.04, NFI = 0.95, and CFI = 0.97

*** significant at p < 0.001.

.

4.2.1. Direct and Indirect Effects

The analysis revealed a significant and positive influence of big data analytics powered by artificial intelligence on innovation performance (β = 0.62, t = 21.720, p < 0.001), strategic agility on innovation performance (β = 0.59, t = 19.267, p < 0.001), and big data analytics powered by artificial intelligence on strategic agility (β = 0.43, t = 11.218, p < 0.001), supporting H1, H2, and H3. The results also revealed an indirect effect of big data analytics on innovation performance via strategic agility (β = 0.19, t = 8.502, p < 0.05). Thus, H4 was supported.

4.2.2. Moderating Effect

To test the moderating effect of market turbulence, we used the PROCESS macro method [96].

Table 5. Results of PROCESS macro.

Effects	Estimates	LL 95% CI	UL 95% CI
Direct effects Big data analytics-AI Innovation performance Big data analytics-AI Strategic agility Strategic agility Innovation performance Indirect effects Indirect effect Total effect Conditional indirect effects Market turbulence High-level market turbulence Low-level market turbulence	0.62 0.43 0.59 0.14 0.29 0.41 0.18 0.07	0.23 0.18 0.21 0.02 0.06 0.13 0.08 0.01	0.41 0.33 0.37 0.21 0.32 0.39 0.31 0.11

demonstrates the results of testing the moderating influence of market turbulence. When strategic agility is our dependent construct, the results revealed a significant interactions coefficient between market turbulence and big data analytics (β = 0.41, t = 23.209, p < 0.001), supporting H5. It also indicated that when innovation performance is our dependent construct, a significant interactions coefficient between market turbulence and strategic agility was found (β = 0.36, t = 17.836, p < 0.001), supporting H6.

5. Discussion and Implications

5.1. Key Findings

The main purpose of this paper was to explore the direct and indirect effect of big data analytics powered by artificial intelligence on innovation performance through strategic agility using the perspective of dynamic capabilities. It also explores the moderating role played by market turbulence on these relationships. Five hypotheses were proposed and tested using SEM/AMOS and the results supported all the hypotheses.

The results revealed that big data analytics powered by artificial intelligence have a significant and positive influence on innovation performance. Our results are consistent with prior research, which found that big data analytics are a key driver of performance [15,51,79]. This relationship is also supported by prior research that discerned a significant and positive impact of big data analytics powered by artificial intelligence on business performance [44].

Our research also shows that the use of AI-driven big data analytics has a notable and beneficial effect on strategic agility. Big data analytics driven by AI in previous research have been linked to agility [16] and this finding is consistent with such research. For instance, prior examination revealed that technological capabilities are a key driver of strategic agility, which in turn can improve business performance [97].

Another interesting finding of our work is that strategic agility mediated the relationship between big data analytics and innovation performance. This finding is supported by a prior exploration that demonstrated the mediating role of strategic agility on the link between digital capabilities and firm performance [98]. A previous study also found that strategic agility mediates the relationship between technological capabilities and performance [44].

Finally, our analysis revealed that market turbulence moderated the relationships between big data analytics driven by AI, strategic agility, and innovation performance. These results are in line with previous studies that indicated that market turbulence moderates the relationship between agility and business performance [54]. Another study found that market turbulence plays a moderating role on the link between digital capabilities and business performance [99].

5.2. Theoretical Implications

We have followed the reasoning of a particular prior study and looked at the issues with technologies and big data analytics that matter greatly to business performance [15,41,100]. The use of artificial intelligence (AI) to power big data analytics capabilities has been hotly debated in recent years as a means of boosting corporate efficiency and business success. Most of the studies, however, have either neglected or ignored the scientific rigor necessary for doing empirical research [41]. To address these obstacles, we conducted an extensive empirical study to learn more about the impact of activities like big data analytics powered by AI on innovation performance via strategic agility.

By combining strategic agility and IT management, our research contributes in three important ways to the existing literature. First, we have investigated the link between AI-driven big data analytics and operational nimbleness. Our premise is based on an interpretation of organizations’ capacities as dynamic systems. According to our research, strategic flexibility is greatly aided by big data analytics fueled by AI. The theories of reasoned action [101] and planned behavior [102], and the technological acceptance model [77], have all been used by researchers to better comprehend how and why new technologies are adopted. However, a fresh viewpoint is offered by taking the dynamic capabilities approach to the introduction of new technologies like big data analytics fueled by AI. Therefore, we contend that BDA-AI is a valuable skill since it allows an organization to foresee the need for, and plan for, investments in strategic agility.

The fields of innovation also benefit from our work. Our findings imply that not all sorts of data contribute equally to an organization’s learning process. Organizations can benefit most from learning via innovation when they have access to a wide range of data collected in real time. Diversity in the workplace can spur creativity and new ideas. Greatly varied data may be more favorable to developing new ways for organizations to compete. Innovation relies on the recombination of knowledge and imagination.

Second, we examined the moderating impacts of market turbulence on the routes of BDA-AI-strategic agility and innovation performance, and our results suggested that the influence of BDA-AI on strategic agility and innovation performance is stronger when markets are very turbulent. Our research shows that BDA-AI can significantly affect strategic agility and innovation performance in environments with high levels of market instability.

Third, by factoring in the effect of market turbulence, this research contributes to the ongoing debate about the importance of dynamic capabilities in a variety of contexts [40]. Some academics appear to be confusing the presence of dynamic capacities with environmental conditions, and this mistake persists in the literature. The ability to adapt quickly to new circumstances is one example of a dynamic capability [47,49,53]. But, previous studies revealed that one of the major concerns is to equate external factors with an organization’s internal strengths and weaknesses [51]. Not enough research has been done on the mechanism by which dynamic capabilities influence performance under different levels of market turbulence [40], even though some researchers have investigated the contingent effect of dynamic capabilities and firm performance relationships [101]. Our findings are consistent with those of prior examinations: that dynamic capacities are more significant in dynamic situations, as shown by the results of the simultaneous moderation and mediation test [103,104]. Our findings contradict those in previous studies which concluded that dynamic capabilities were necessary across all environmental levels [54], while their use may become most apparent in environments with moderate to low levels of market turbulence.

5.3. Managerial Implications

There are many ways in which manufacturing companies and their management might profit from this research. First, our research encourages production managers to implement BDA management practices and boost employee innovations for more strategic agility in their companies. Our research could also help executives understand how client input is the key to a company’s exceptional strategic agility. An industrial company’s decision-making processes could be jeopardized by insufficient data visualization capabilities. The incorporation of knowledge from external players has been identified as a problematic and significant part of the digital change transition in a number of earlier studies. Our research demonstrates that managers can cut down on order-to-delivery times by implementing new digital transformation methods if they adopt an agile approach and use our recommended methodology. Therefore, strategic agility-focused businesses should think about implementing BDA-AI to improve their innovation performance.

This research demonstrates that big data analytics involves more than merely spending money on tools, amassing huge datasets, and giving the IT department leeway to try out new analytical methods. In addition to the aforementioned factors, it is also crucial to embed big-data decision-making into the fabric of the organization, recruit people who have a solid technical and managerial understanding of big data and analytics, and cultivate a culture of organizational learning. In this way, these assets work together to help a company build a BDAC and increase its value. That calls for a wide range of initiatives, which in turn necessitates buy-in from upper management and a well-defined strategy for rolling out big data analytics across the board.

Understanding where and how to establish strategic agility is crucial, as is knowing how to use BDA-AI to boost innovation performance. Additionally, market turbulence may modify how BDA-AI affects strategic agility and innovation performance throughout adoption. Therefore, it is important for managers to comprehend how different levels of market volatility affect the efficacy of BDAAI adoption and consequently, the effects on agility and performance. The empirical findings of this study provide a comprehensive knowledge of BDA-AI and strategic agility, which in turn helps clarify the role that market turbulence plays in determining the usefulness of dynamic capabilities. Our study’s data-driven research has additional benefits, including lending credence to our recommendations for practitioners and providing extra information that may be used to evaluate how well those recommendations are being carried out.

A data-driven culture is one of the most ephemeral of these factors. Many recent company publications have stressed the significance of a data-driven culture in extracting value from big data investments. Companies with a data-driven culture believe that decisions supported by an empirical data analysis should be given more weight. Increasing the number of data-informed decisions made by managers, broadening the types of decisions influenced by data, and encouraging decision-making that blends analytical insight with intuition are all steps towards becoming a data-driven firm, which is receiving increasing attention from the practitioner community.

Big data analytics and artificial intelligence (AI) have the potential to bring about significant benefits to society, but they also raise important ethical considerations. The collection and analysis of large amounts of data can lead to privacy concerns. It is important to ensure that individuals’ personal information is handled with care and that data is anonymized when necessary. Consent for data collection and transparent data usage policies are crucial. Many AI systems, especially deep learning models, can be seen as “black boxes” because their decision-making processes are not easily explainable. This lack of transparency can be problematic when decisions have significant consequences. Developing AI systems that are more transparent and providing explanations for their decisions is a growing area of research. Big data stores and AI systems can be vulnerable to cyberattacks. Protecting these systems from malicious actors is crucial to prevent misuse and data breaches. The development and deployment of AI and big data technologies often cross international borders. Establishing global ethical standards and guidelines can help ensure consistency and fairness in their use. Addressing these ethical concerns requires collaboration between policymakers, technologists, ethicists, and other stakeholders. It is essential to strike a balance between innovation and responsible-use to harness the full potential of big data analytics and AI while minimizing harm and ensuring ethical practices. Ethical frameworks and regulations are continually evolving to keep pace with these rapidly advancing technologies.

Finally, our proposed theoretical model can serve as a practical approach by which manufacturing organizations may boost performance outcomes, given the growing concerns surrounding strategic agility and its management. Our research also suggests that in order to attain leadership in innovation, performance managers should include customers in the data analysis. When customers are involved in the decision-making process, the company gains valuable operational information that can be used to spot trends in the market, predict future demand, and meet the specific needs of its customers in terms of product customization and time between order placement and delivery. Several prior studies have shown that customer involvement is a key mechanism for firm innovation [19,44]; therefore, it is no surprise that customer involvement can boost innovation performance [22]. Involving customers in the data analysis also helps businesses absorb all the insights and information they can glean from their data, which should boost creativity. Institutional vacuums, such as those seen in emerging countries [26,103], provide significant obstacles for businesses, elevating the importance of external elements such as consumers (industrial customers—e.g., lead firms for suppliers). Prior research revealed that businesses rely on informal sources and external actors to generate value [23,104]. We contend that the manufacturer’s ability to create value from big data, which in turn leads to strategic agility, is facilitated by the involvement of customers in BDA. According to the DC view, a company can develop novel insights from existing assets, which is important to the interaction between BDACs and strategic agility. Therefore, customer involvement could play a critical role as a key organizational external knowledge mechanism suited to understanding how organizations may balance their capabilities to develop manufacturing agility.

6. Limitations and Future Research Directions

Our study was restricted to manufacturing firms in Saudi Arabia, which limits the study’s generalizability. Additional studies in the future may confirm our proposed model in other places and deepen our understanding of these problems. Customers’ participation as data analysts could lead to novel insights that help predict market demand and expand access to existing data, which could be shared across divisions to facilitate the adaptation by digital transformation. The impact of BDA-AI on preexisting organizational structures and on radical and incremental innovation capacities should be explored in future research. Moreover, social capital (i.e., “trust, support, information and knowledge exchange among partners, such as focal enterprises and suppliers”) should be taken into account when enlisting customers as data analysts. Especially when talking about the moderating role of customers as data analysts, future research should broaden this topic by integrating the role played by social capital in the suggested framework. Furthermore, exploring the main antecedents of BDA-AI would help us to understand how to foster it. Future studies can examine the influence of entrepreneurial orientation as a driver of BDA-AI. In addition, our study operationalized innovation performance without considering the various dimension of it. Future studies could explore the varied dimensions of innovation, such as radical versus incremental innovation, which might offer more insights in the study context.