Stakeholder engagement & knowledge digitalization for sustainable performance in the era of artificial intelligence

Abstract

Technological factors and stakeholder engagement promote organizations’ sustainable performance. This study aims to determine sustainable performance through stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, data-driven decision-making (DD), and responsible big data intelligence. Data were collected from 394 managers of Chinese manufacturing organizations via a questionnaire survey conducted between June and August 2025. The results reveal that stakeholder involvement and knowledge digitalization are key factors in DD. Additionally, DD significantly mediates stakeholder engagement, knowledge digitalization, and sustainable performance. Disruptive innovation substantially moderates the stakeholder engagement–DD relationship. Moreover, AI information quality significantly moderates the knowledge digitalization–DD relationship. Responsible big data intelligence moderates the DD–sustainable performance relationship. This study highlights a significant issue in how the manufacturing sector’s management utilizes stakeholder engagement, knowledge digitalization, DD, disruptive innovation, AI quality information, and responsible big data intelligence to determine sustainable performance. This initial research incorporates stakeholder engagement, knowledge digitalization, DD, disruptive innovation, AI quality information, responsible big data intelligence, and sustainable performance from the perspective of the resource-based view.

Introduction

For manufacturing organizations sustainable performance is a key objective, which is why management pays immense attention toward its improvement (Henao & Sarache, 2022; Shiferaw et al., 2025). Manufacturing organizations actively seek relevant technologies to increase sustainable performance in a turbulent environment (Yang et al., 2025). Disruptive innovation fundamentally alters the market landscape by creating new value chains and replacing existing products and firms (Christensen et al., 2018). In the current competitive environment, firms cannot operate in isolation. Stakeholders—that is, individuals or groups who are directly or indirectly impacted by firms’ actions—constitute the pillars of any organization (Yoon & Chung, 2018). Firm–stakeholder collaboration yields substantial information, which requires processing to facilitate effective decision-making (Goodman, 1993). Data-driven decision-making (DD) enables firms efficiently process information and generate actionable outcomes (Giachino et al., 2025), serving as a bridge among stakeholders’ engagement, disruptive innovation, knowledge digitalization, and sustainable performance (Adomako et al., 2021; Hopp et al., 2018; Zong & Guan, 2025). As data analytics as a decision-making tool becomes increasingly embedded in organizations, the model assumes that companies will be significantly more efficient and effective, as well as better aligned with the broader sustainability agenda related to the business. There exists support for the argument that analytics can promote the contextual relevance of the decision-making process to meet sustainability objectives.

Therefore, each enterprise requires an appropriate DD framework to allocate its resources to actively pursue environmentally and economically responsible actions, thereby opening a sustainable development channel (Awan et al., 2021). Apart from providing refined input for the decision-making process, stakeholder engagement, in turn, helps align innovations with society’s expectations (Bal et al., 2013). In this context, artificial intelligence (AI) is an effective tool that facilitates information acquisition and dissemination, as well as reasoned decision-making. Duan et al. (2019) posit that AI-generated information aids decision-making. Helu et al. (2016) state that integrating smart manufacturing technologies improves decision-making. Al-Okaily and Al-Okaily (2025) utilized data from Jordanian firms to study five factors that influence decision-making: technology capability, data capability, information quality, data-driven insights, and financial decision quality. These studies indicate that information quality and big data are key factors for data-based decision-making. Recently, big data has acquired increasing significance as a crucial tool for solving complex social and environmental issues (Xiao & Qu, 2025). In this context, Dubey et al. (2019) note that appropriate managerial skills enable effective utilization of big data to ensure social and environmental sustainability.

This study integrates stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence to determine sustainable performance from the perspective of the resource-based view (RBV)—a viewpoint largely ignored by prior research (Barney, 1991; Cheng et al., 2026; Dal Maso et al., 2017; Dubey et al., 2019; Hopp et al., 2018; Mikalef & Gupta, 2021). The research objectives of the study are as follows:

• RO1: To examine the influence of stakeholders’ engagement and knowledge digitalization on data-driven decision-making and sustainable performance.

• RO2: To examine the mediation mechanisms of data-driven decision-making among stakeholders’ engagement, knowledge digitalization, and sustainable performance.

• RO3: To investigate how disruptive innovation, AI information quality, and responsible big data intelligence shape these relationships.

The current research makes several valuable contributions. The researchers combined key intangible resources for sustainability performance, which shape a firm’s sustainability outcomes (Kumar et al., 2025; Rashid et al., 2025). Moreover, we examined several moderation and mediation mechanisms through the hitherto ignored RBV theoretical lens (De Silva et al., 2025; Mohamed Riyath & Inun Jariya, 2024). The study explored stakeholders’ engagement rather than pressure, adopting a proactive rather than reactive approach (Sahoo, 2024). Data were gathered from manufacturing enterprises in Jiangsu Province, China, including Nanjing, Yangzhou, Xuzhou, Wuxi, Suqian, and Huai'an. The study has practical implications for managers regarding the strategic importance of integrating knowledge and quality information, as well as the effective application of stakeholder knowledge in managerial decision-making. By integrating sustainability goals into business policy, managers can ensure that their companies remain competitive and contribute to society as a whole. It confirms the systematic direction of intelligent manufacturing enterprises in China, which integrate innovation and technology with traditional manufacturing to respond to the new challenges and opportunities presented by the global economy. By leveraging these digital capabilities, organizations can achieve a competitive advantage, which could, in turn, enable them to pursue long-term sustainability and responsible business practices in addition to profit maximization.

Theoretical background and hypothesis development

Resource-based view theory

The RBV is a leading theory that postulates that resources and capabilities are vital assets in attaining sustainable competitive advantage and performance (Barney, 1991). RBV highlighted the association between a firm’s resources and capabilities and its success (Jafari-Sadeghi et al., 2021). Hence, firms should focus on both tangible and intangible components to compete in the market and foster a dynamic relationship with their stakeholders (Dezi et al., 2019; Ferraris et al., 2019). The researchers stated that tangible resources can be easily copied, unlike intangible resources (Rehman et al., 2023). Thus, organizations focus more on unique intangible resources. This study examines several intangible resources, such as stakeholder engagement, knowledge digitalization, AI information quality, and responsible big data intelligence in measuring sustainable performance. Moreover, disruptive innovation and DD are capabilities that help improve sustainable performance.

Stakeholder engagement is considered a valuable resource for organizations and impacts firm performance (Dal Maso et al., 2017). Knowledge digitalization implies a combination of digital technology and knowledge resources (Cheng et al., 2026). Disruptive innovation refers to an innovation that dramatically changes industries or markets in a technological context (Hopp et al., 2018). AI information quality is a significant resource for an organization’s success, but by itself it is unlikely to deliver any competitive advantage (Mikalef & Gupta, 2021). DD is considered a firm’s capability. Research suggests that firms’ capabilities can explain the relationship between firms’ resources (i.e., stakeholder engagement and knowledge digitalization) and sustainable performance (Barney, 1991). Dubey et al. (2019) stated that big data analytics are essential for effectively utilizing big data to ensure social and environmental sustainability. Hence, this study applied the RBV to explain the research framework.

Stakeholder engagement and data-driven decision-making

Stakeholder engagement involves the active participation of individuals or groups who have a vested interest in an organization’s decisions (Kujala et al., 2022). DD refers to the process of making informed decisions based on data analysis rather than relying solely on intuition or observation (Bousdekis et al., 2021). Bag et al. (2024) reported that new technologies, including big data and predictive analytics, can impact an organization’s predisposition to share information on circular economy practices with stakeholders and to become more connected with those stakeholders in the Industry 4.0-era. This improves stakeholder trust, engagement, and social sustainability. By actively engaging stakeholders throughout the decision-making process, organizations can obtain valuable insights and contextual data, ultimately improving decision quality. Moreover, studies posit that when stakeholders are effectively engaged, organizations can leverage their insights to refine data interpretation, which is critical for making well-reasoned decisions after considering multiple aspects of a problem (Watson et al., 2018).

Furthermore, Deverka et al. (2012) proposed a structured framework for stakeholder participation, arguing that effective engagement enriches the decision-making process and enhances the quality of data-derived outcomes. The integration of advanced analytics into stakeholder engagement offers a significant opportunity to improve decision-making. Prior studies identified that stakeholder engagement significantly enhances environmental performance (Huynh & Nguyen, 2025) and project success (Shaukat et al., 2022). Scarce attention has been paid to determining DD through stakeholder engagement (Song et al., 2025). Hence, we strive to nurture this relationship. The relationship between stakeholder engagement and DD highlights how data and insights shared among stakeholders can enhance decision-making. Based on the above discussion, the following hypothesis is proposed:

H₁: Stakeholders’ engagement is related to data-driven decision-making.

Knowledge digitalization and data-driven decision-making

Digital tools can significantly enhance knowledge sharing, ultimately facilitating prompt, data-driven decisions across industries, which renders their application critical (Deng et al., 2023). Heltberg (2022) argued that the digitalization of information management can improve decision-making by enabling organizations to process voluminous data effectively to extract useful information. Pan et al. (2021) developed this thought by stating that data-driven strategies are replacing traditional knowledge-based decision-making strategies as cloud computing and big data technologies advance, driven by digitalization. This change facilitates more rational, evidence-based decision-making, contributing to improved overall organizational effectiveness. Moreover, Zhang (2024) reported that success in the presence of changing dynamics is determined by information orientation within organizations. It is realized in various dimensions at the organizational level, such as information sharing, technology integration, DD, and network resource acquisition. The study investigated how the three independent variables of information orientation, digital adoption, and knowledge digitalization influence the improvement of information management practices via the moderating effect of knowledge management and reported significant results. Additionally, Colombari et al. (2023) identified that digitalization supports DD, ultimately improving firm performance. A few studies have shown that the knowledge digitalization significantly impacts innovation performance (Cheng et al., 2023) and innovation ambidexterity (Zhang et al., 2025). However, researchers have largely ignored the influence of digital knowledge on DD (Han et al., 2024). Knowledge–data analysis integration is essential for decision-making frameworks, especially for organizations that intend to thrive in an increasingly competitive business environment. Based on the above discussion, the following hypothesis is proposed:

H₂: Knowledge digitalization is related to data-driven decision-making.

Data-driven decision-making and sustainable performance

It is well known that DD is intimately associated with sustainable performance across a broad range of industries (Awan et al., 2021). Efficiently utilizing data analytics will not only optimize operations but also make them more environmentally friendly, thereby improving performance (Huang et al., 2023). The authors examine a convergence of the concepts of Lean Six Sigma, DD, and sustainable manufacturing. Their results show that companies that use DD methods achieve significantly better environmental performance, thereby confirming the hypothesis that champions of such integration maintain sustainability programs (Huang et al., 2023). Furthermore, it has been reported that introducing data analytics into business processes significantly improves organizational behavior and sustainable performance. Several researchers have found that DD significantly determines the performance of higher education institutions (Ashaari et al., 2021) and decision-making effectiveness (Cao et al., 2015). Furthermore, Mikalef et al. (2019) show that big data analytics improves company innovation and sustainability. Their findings suggest that dynamic capabilities are crucial and that firms with robust DD frameworks can effectively overcome environmental challenges without compromising competitiveness. This connection underscores the importance of an organizational culture that fosters data-driven strategies and promotes sustainable practices. Thus, the following is the study hypothesis:

H₃: Data-driven decision-making significantly relates to sustainable performance

Mediating role of data-driven decision-making

Studies indicate that information-driven practice is a key factor in enhancing sustainable performance in manufacturing operations. For instance, Abdul-Rashid et al. (2017) argued that sustainable manufacturing practices have a significant effect on sustainability performance, economic viability, social equity, and environmental protection. It implies that manufacturers are more likely to meet sustainability goals when they proactively involve stakeholders and incorporate their feedback into the DD process. The importance of DD is especially pronounced when exploring its potential to enhance stakeholder involvement by facilitating better decision-making. Hussain et al. (2018) suggest that effective corporate governance, based on stakeholder contributions and data analysis, can improve the Triple Bottom Line (TBL) performance. The results suggest that organizations that utilize data not only to comply but also to proactively involve stakeholders are likely to perform well in terms of sustainability. Furthermore, DD effects on sustainable practices are implementing Lean Six Sigma practices. Using data to involve stakeholders in decision-making fosters trust and cooperation, both of which are crucial for achieving industrial sustainability goals (Huang et al., 2023). Budsaratragoon and Jitmaneeroj (2019) reported that corporate sustainability involves stakeholder engagement and informed decision-making in order to recognize and respond to the key drivers of sustainable practices. Their results show that, as manufacturing companies use data as a strategic resource, they improve performance and promote relationships with stakeholders, thereby improving the overall sustainability indicators.

H₄: Data-driven decision-making mediates between stakeholders’ engagement and sustainable performance.

With appropriate knowledge management, aided by digitalization, leaders can make informed decisions that generate long-term value for stakeholders and enhance their sustainability performance. It implies that leaders have access to the organization. Furthermore, Abdul-Rashid et al. (2017) examined the effect of sustainable manufacturing practices on sustainability performance. They concluded that sustainability practices are essential for organizations aiming to improve their sustainability indicators. Knowledge digitalization enables manufacturers to leverage data to optimize their operations, thereby improving performance. Nicolăescu et al. (2015) further developed the sustainability performance measurement by discussing the role of metrics in transforming organizations into sustainable ones. DD enables the collection and analysis of performance data that is central to steering organizations throughout the sustainability process. From an RBV perspective, knowledge digitalization is a crucial resource that can facilitate data-driven insights and improve performance (Barney, 1991). Based on the above discussion, the following hypothesis is proposed:

H₅: Data-driven decision-making mediates between knowledge digitalization and sustainable performance.

Moderating role of disruptive innovation

Disruptive innovation implies innovation that dramatically alters industries or markets through technological change (Hopp et al., 2018). Stakeholder engagement involves examining the relationships between organizations and various stakeholders to elicit insights, enhance these relationships, and ultimately foster innovation (Watson et al., 2018). Disruptive innovation facilitates integrating diverse stakeholder opinions into the process. Chen et al. (2025) highlighted the significant relationship between sustainable innovations and stakeholder interaction. Similarly, stakeholders have been reported to contribute to the production of market-driven innovations. Through multiple views, these advances improve decision-making. Juntunen et al. (2019) reported that the quality of contacts improves sustainable innovation. This is consistent with the claim that the more firms improve their stakeholder engagement strategies, the more they will, wittingly or naturally, depend on data insights to render their DD more informed and responsive. Jäger et al. (2023) showed that involving stakeholders can facilitate complex innovation paths. Involvement of various stakeholders in the process generates a rich knowledge pool, which is essential for DD that accounts for real-world complexities and stakeholder expectations. Based on the above discussion, the following hypothesis is proposed:

H₆: Disruptive innovation moderates between stakeholders’ engagement and data-driven decision-making.

Moderating role of AI information quality

AI plays a significant role in improving information quality and facilitates effective organizational decision-making (Neiroukh et al., 2025). This is especially crucial for manufacturing firms, which require high-quality information to make the right decision at the right time. When AI-generated data is trustworthy, it can significantly enhance decision-making effectiveness and reinforce strategic objectives in the production setting. Thus, AI is crucial for business decision-making. Helu et al. (2016) stated that integrating smart manufacturing technologies would effectively promote decision-making. They suggested that DD can be improved by creating a shared body of knowledge that incorporates AI-created information. In production, appropriately digitalized knowledge is fed into AI systems, thereby improving the quality of insights derived from such data (Olan et al., 2022). When the quality of AI-generated information is high, it enhances decision-making by providing more insights and reducing uncertainty. Nguyen et al. (2022) highlighted the potential effect of AI quality on customer experiences and suggested that this phenomenon can be extended to the organizational decision-making context by referring to the value of information quality across numerous business practices. Based on the above discussion, the following hypothesis is proposed:

H₇: AI information quality moderates between knowledge digitalization and data-driven decision-making.

Moderating role of responsible big data intelligence

Digital intelligence technologies, such as big data analytics, play a significant role for organizations (Huang & Zhou, 2025). New-generation digital technologies, such as big data analytics, are a critical source of innovation (Zhang et al., 2024). Big data analytics are employed to enhance operational efficiency, competitiveness, and performance (Sarwar et al., 2025). Many studies emphasize implementing big data analytics in the organizational context (Meng & Wang, 2023). Furthermore, studies have shown that responsible and ethical management of big data enhances decision-making, resulting in a direct positive impact on sustainability outcomes. Mikalef et al. (2019) explicated why dynamic capabilities are required to optimize the value of big data analytics, suggesting that the key concern for organizations aiming to realize the full potential of DD and implement sustainable practices into their decision-making process should be ethical data handling. Dubey et al. (2019) noted that managerial skills are essential for effectively utilizing big data to ensure social and environmental sustainability, highlighting that effective management of big data analytics can enhance firm ability to address sustainability issues, thereby bridging the relationship between data-driven insights and sustainable performance. Consequently, companies that adopt powerful big data analytics are more likely to make informed decisions that improve performance indicators and advance broader sustainability agendas. In this context, Hassan et al. (2016) reported that integration of sustainable practices into the supply chain can be improved through the application of data-driven approaches. Furthermore, Riaz and Ali (2024) examined the role of responsible innovation and sustainability outcomes, reporting that big data significantly moderates the relationship between responsible innovation and sustainability performance. This aligns with the RBV, which posits that rare resources are crucial for performance. Thus, responsible application of big data is a crucial resource that contributes to sustainable outcomes. Based on the above discussion, the following hypothesis is proposed:

H₈: Responsible big data intelligence moderates the balance between data-driven decision-making and sustainable performance.

Table 1 shows that several studies have been conducted on stakeholder engagement, knowledge digitalization, and DD. However, researchers have largely ignored the significance of DD through stakeholder engagement and knowledge digitalization in the presence of disruptive innovation and AI information quality. Moreover, researchers are unaware of measuring sustainable performance through DD in the existence of responsible big data intelligence. Hence, this study addresses this gap. Fig. 1 portrays the research framework.

Table 1.

Stakeholder engagement and knowledge digitalization, data-driven decision-making, and sustainable performance.

No.	Authors/Year	Country	Exogenous Variable	Sample	Research Focus	Remarks
1	Huynh and Nguyen (2025)	Vietnam	Stakeholder engagement	Listed businesses	Environmental Performance	Significant
2	Song et al. (2025)	Brazil,	Stakeholder engagement	Strategic planners and stakeholders	Sustainable Project Performance	Mixed
3	Zhang et al. (2025)	China	Knowledge Digitalization	Employees	Innovation ambidexterity	Significant
4	Han et al. (2024)	China	Knowledge Digitalization	High-tech firms	Firms Performance	Significant
5	Cheng et al. (2023)	China	Knowledge digitalization	Furniture Enterprises	Innovation performance	Significant
6	Shaukat et al. (2022)	Pakistan	Stakeholder engagement	Project Management Professionals	Project Success	Insignificant
7	Ashaari et al. (2021)	Malaysia	Data-driven decision-making	Higher education institutions	Higher education institutes' performance	Significant
8	Cao et al. (2015)	United Kingdom	Data-driven decision-making	Medium and Large Companies	Decision-making effectiveness	Significant

Fig. 1.

Research Framework. Note: SE = stakeholder engagement; DD = data-driven decision-making; SP = sustainable performance; KD = knowledge digitalization.

Methodology

Design and participants

Data were gathered from manufacturing enterprises in Jiangsu Province, China, including cities such as Nanjing, Yangzhou, Xuzhou, Wuxi, Suqian, and Huai'an, as these organizations rigorously adhered to green practices and complied with the stipulations of the Environmental Protection Law. The 10-times rule is applied to the sample size. This rule states that the minimum sample size should be 10-times the maximum number of arrowheads to the latent construct in the PLS path model (Hair et al., 2021a). Two arrows lead DD in Fig. 1, which makes it 20. Similarly, it was reported that the sample size for PLS-SEM must be at least 10-times the number of indicators per variable in the model (Peng & Lai, 2012). Responsible big data intelligence has the most items/indicators (15 in this study). The recommended minimum sample size is 15 × 10, or 150. A self-administered questionnaire was utilized to collect data from selected firms, employing purposive sampling. Data were collected directly from employees involved in implementing sustainable practices. Surveyors were engaged for data acquisition. In accordance with the directives given to the surveyors, they informed all respondents of the research objective and subsequently requested that they complete the questionnaire. Upon agreement, they were provided with the questionnaire to complete. A total of 425 employees among the selected organizations submitted the surveys, and 394 valid responses were included for data analysis. Participation was voluntary, and participants were assured of the anonymity and confidentiality of their responses. Table 2 presents a breakdown of various categories related to manufacturing sector managers; the sample comprises almost equal proportions of male (50.25%) and female (49.75%) managers. In terms of age, the majority (69.54%) fall within the 26–45-year range, while a smaller percentage is in the 0–25-year (20.05%) and 46–55-year (10.41%) age groups. Regarding educational qualifications, most managers hold a bachelor’s degree (54.06%), followed by those with a master’s degree (35.79%). A smaller number have completed a diploma or equivalent (7.87%), and only 2.28% possess a PhD degree. In terms of employment type, a majority of the managers (58.63%) are employed in permanent positions, while 41.37% hold contractual roles. Regarding service duration, most managers (51.27%) have been with their firms for 2–5 years, followed by those with 5–10 years of experience (30.46%). A smaller proportion has up to 1 year (10.40%) or over 10 years (7.87%) of service experience.

Table 2.

Demographic profile of respondents.

Demographics	Frequency	(%)
Gender
Male	198	50.25%
Female	196	49.75%
Age Group
Up to 25 years	79	20.05%
26 to 45 years	274	69.54%
46 to 55 years	41	10.41%
Qualification
Bachelor’s Degree	213	54.06%
Master’s Degree	141	35.79%
Others (Diploma, etc.)	31	7.87%
Ph.D.	9	2.28%
Employment Type
Permanent Job	231	58.63%
Contractual Employment	163	41.37%
Length of Service
Up to 1 year	41	10.40%
2–5 years	202	51.27%
5–10 years	120	30.46%
10+ years	31	7.87%

Questionnaire and pre-test

All measuring instruments applied to assess the underlying constructs were derived from prior studies. Stakeholder engagement encompasses eight key items, as outlined by Ansong (2017). Knowledge digitalization comprises seven items (Cheng et al., 2023; Han et al., 2024). Disruptive innovation encompasses five key elements, as outlined by Govindarajan and Kopalle (2006). AI information quality encompasses six key items (Nguyen & Malik, 2022; Wixom & Todd, 2005). DD encompasses four key elements (Ashaari et al., 2021; Khan & Fatima, 2025). Responsible big data intelligence comprises 15 items: 10 items for responsible AI derived from Wang et al. (2023), and five items for big data analytics capability (Chen & Liang, 2023). A recent study applied 15 items to measure responsible big data intelligence (Al Zaabi et al., 2025). Sustainable performance encompasses three key dimensions: economic, social, and environmental performance. Economic performance includes five items from Zhu et al. (2013). Social performance comprises five items, as outlined by Paulraj (2011). Finally, environmental performance includes five items derived from Laosirihongthong et al. (2013) (see Appendix A for details of items).

Content validity was established by distributing the measurement instrument to three experts, comprising two from selected organizations and one assistant professor in the relevant field. Minor modifications were implemented based on their recommendations. A pilot study was conducted to ensure clarity, during which a questionnaire was administered to 38 respondents to gather their views. The surveys necessitated no modifications whatsoever. A reliability analysis was conducted to evaluate the internal consistency of the scales employed in the study.

Operationalization of constructs

Stakeholder engagement is defined as “practices the firm undertakes to engage stakeholders positively in firms’ activities” (Greenwood, 2007). Furthermore, knowledge digitalization leverages advanced information and communication technologies, such as sensors, digital systems, and machine vision, to convert comprehensive knowledge, including production, business, technical, and marketing information, into a digitally accessible format (Cheng et al., 2023). Moreover, new products introduced are those that mainstream customers eventually found attractive, as they satisfied the requirements of the mainstream market over time—a phenomenon termed disruptive innovation (Govindarajan & Kopalle, 2006). AI information quality refers to AI tools’ ability to produce accurate, latest, and error-free information (Nguyen & Malik, 2022; Wixom & Todd, 2005). Similarly, DD refers to the process of using facts, data, and analytics to guide business decisions rather than relying on guesswork or intuition (Ashaari et al., 2021). Responsible big data intelligence integrates ethical considerations into AI-driven data analytics, guiding decision-making across sectors (Al Zaabi et al., 2025). Finally, sustainable performance implies an organization’s ability to attain financial goals while simultaneously making a positive, long-term impact on the environment and society (Rehman et al., 2021).

Results

Bias control

To evaluate data accuracy, we performed several analyses to detect potential biases. To evaluate late-response bias, we performed an independent t-test comparing the replies of early responders with those of late responders (Armstrong & Overton, 1977). The results indicated no statistically significant differences between the two groups. Consequently, late-response bias was not a substantial cause for concern. An independent-samples t-test was conducted to assess non-response bias by comparing the attributes of responding and non-responding managers across education, experience, company size, and age. No statistically significant difference was observed between the two groups. Consequently, the study did not experience a substantial problem with non-response bias. Furthermore, we corroborated a portion of the survey data by cross-referencing it with secondary sources, as applicable, to ensure precision. Common method bias (CMB) may probably impact our findings owing to our research design, as data were collected from a single source—specifically, managerial workers. Hence, there exists a possibility that CMB can arise and disturb study data (Kraus et al., 2020). CMB is recognized as a serious concern often noted in surveys (Podsakoff & Organ, 1986). Statistical and procedural methods are employed to mitigate CMB issues. From a procedural perspective, researchers assure respondents that their information will not be disclosed to any third party without their consent, and that the questionnaire’s language is simple (Podsakoff et al., 2012). This study applies Herman’s single-factor and variance inflation factor (VIF) for CMB. Herman’s single-factor variance value is 41.215%—below 50%. According to Kock (2015), prior research has sought to assess common technique bias by examining VIFs derived from a comprehensive collinearity test. The VIF scores indicate that a cut-off value of 3.3 or higher suggests the estimated model may be affected by CMB, while values below 3.3 imply the model is CMB free. Hence, both criteria are fulfilled, and there exists no CMB issue. Finally, to assess potential endogeneity issues, we employed the Gaussian copula approach outlined by Hult et al. (2018). All values of the Gaussian copulas in our findings were negligible, indicating that endogeneity is unlikely to pose a substantial issue in our study.

Measurement model assessment

Previous research has shown that PLS-SEM is the most efficient method for analyzing models with mediation, moderation, and inherent complexity (Preacher & Hayes, 2004). A confirmatory factor analysis was conducted within the measurement model using SmartPLS 4. Table 3 illustrates the findings of the measurement model. Figs. 2 and 3 depicts the measurement model at first and second orders, respectively. All metrics exhibit satisfactory levels of construct dimensionality, composite reliability, average variance extracted (AVE), and discriminant validity. Table 3 shows that the minimum loading is 0.631, and the maximum is 0.995. The results surpass the established criterion of 0.5 (Hair et al., 2021b). Internal consistency is evaluated using composite reliability (CR) and Cronbach’s alpha—both values are expected to be at least 0.70 (Becker et al., 2023). Hair et al. (2021b) contend that the AVE should surpass 0.50. Table 3 indicates that all AVE, CR, and alpha values are within the required range.

Table 3.

Convergent validity.

First-Order Constructs	Second-Order Constructs	Items	Factor Loading	α	CR	AVE	VIF
Stakeholders Engagement		SE1	0.935	0.947	0.956	0.733	2.341
		SE2	0.912
		SE3	0.912
		SE4	0.768
		SE5	0.722
		SE6	0.831
		SE7	0.878
		SE8	0.868
Disruptive Innovation		DI1	0.852	0.888	0.918	0.690	1.543
		DI2	0.828
		DI3	0.840
		DI4	0.805
		DI5	0.829
Knowledge Digitalization		KD1	0.756	0.930	0.945	0.712	1.298
		KD2	0.744
		KD3	0.730
		KD4	0.918
		KD5	0.911
		KD6	0.917
		KD7	0.902
Data Driven Decision Making		DD1	0.828	0.824	0.875	0.637	1.489
		DD2	0.859
		DD3	0.761
		DD4	0.737
AI Information Quality		AQ1	0.995	0.991	0.993	0.959	2.322
		AQ2	0.977
		AQ3	0.976
		AQ4	0.982
		AQ5	0.984
		AQ6	0.960
Responsible Big Data Intelligence		RAI1	0.675	0.946	0.953	0.576	1.489
		RAI2	0.695
		RAI3	0.670
		RAI4	0.861
		RAI5	0.851
		RAI6	0.829
		RAI7	0.845
		RAI8	0.643
		RAI9	0.677
		RAI10	0.631
		RAI11	0.816
		RAI12	0.805
		RAI13	0.768
		RAI14	0.805
		RAI15	0.755
	Sustainable Performance	Economic Performance	0.777	0.824	0.895	0.740
		Social Performance	0.947
		Environmental Performance	0.848
Sustainable Economic Performance		ECP1	0.887	0.927	0.945	0.774
		ECP2	0.875
		ECP3	0.857
		ECP4	0.835
		ECP5	0.942
Sustainable Social Performance		SP1	0.886	0.921	0.940	0.760
		SP2	0.847
		SP3	0.848
		SP4	0.843
		SP5	0.931
Sustainable Environmental Performance		ENP1	0.880	0.921	0.941	0.760
		ENP2	0.848
		ENP3	0.847
		ENP4	0.838
		ENP5	0.942

Note: α = Cronbach’s alpha, CR = Composite Reliability, AVE = Average Variance Extracted, VIF = Variance Inflation Factor.

Fig. 2.

Measurement model-first order.

Fig. 3.

Measurement model- second order.

This research evaluates the discriminant validity of the components in accordance with the principles of Henseler et al. (2015) using the heterotrait-monotrait (HTMT) ratio. Tables 4 and 5 present the HTMT ratios at first and second orders, respectively. The HTMT results confirmed the measurement’s discriminant validity. As indicated in Tables 4 and 5, all values are below the established cut-off of 0.85, as articulated by Henseler et al. (2015), confirming that values below 0.85 affirm the measurement’s discriminant validity. Overall, both outcomes confirmed that discriminant validity is not a concern in the present research.

Table 4.

HTMT-first order.

Constructs	AIQ	DD	DI	ECP	ENP	KD	RAI	SP	SE
AI-Information Quality
Data-Driven Decision Making	0.300
Disruptive Innovation	0.445	0.489
Economic Performance	0.592	0.385	0.565
Environmental Performance	0.388	0.635	0.513	0.448
Knowledge Digitalization	0.203	0.648	0.523	0.377	0.486
Responsible Big data Intelligence	0.242	0.623	0.542	0.367	0.555	0.691
Social Performance	0.414	0.650	0.836	0.760	0.770	0.578	0.613
Stakeholders Engagement	0.772	0.374	0.452	0.797	0.428	0.243	0.221	0.566

Note: AIQ = AI-Information Quality, DD = Data-Driven Decision Making, DI = Disruptive Innovation, ECP = Economic Performance, ENP = Environmental Performance, KD = Knowledge Digitalization, RAI = Responsible Big Data Intelligence, SP = Social Performance, SE = Stakeholders Engagement.

Table 5.

HTMT-second order.

Constructs	AIQ	DD	DI	KD	RAI	SE	SP
AI-Information Quality
Data-Driven Decision Making	0.300
Disruptive Innovation	0.445	0.489
Knowledge Digitalization	0.203	0.648	0.523
Responsible Big data Intelligence	0.242	0.623	0.542	0.691
Stakeholders Engagement	0.772	0.374	0.452	0.243	0.221
Sustainable Performance	0.579	0.682	0.788	0.591	0.630	0.740

Note: AIQ = AI-Information Quality, DD = Data-Driven Decision Making, DI = Disruptive Innovation, ECP = Economic Performance, ENP = Environmental Performance, KD = Knowledge Digitalization, RAI = Responsible Big Data Intelligence, SP = Social Performance, SE = Stakeholders Engagement.

Structural model assessment

This section relates to the structural model illustrated in Fig. 4. SmartPLS 4 was applied with 5000 bootstrap subsamples. Using a two-tailed test at the 5% significance level, Table 5 aids in assessing the study’s hypotheses. Table 6 presents a significant association between SE and DD (β-value = 0.270; p < 0.001), supporting H₁. Similarly, knowledge digitalization is significantly related to DD (β = 0.441; p < 0.001), supporting H₂. The analysis of the influence of DD on sustainable performance revealed a significant relationship between DD and sustainable performance (β-value = 0.458; p < 0.001), thereby validating hypothesis H₃. The mediation of DD between SE and SP (β = 0.124, p = 0.002) was significant, supporting H₄. Similarly, the mediation of DD between knowledge digitalization and SP (β-value = 0.124; p-value = 0.002) was significant and supported H₅. The moderation analysis of stakeholders’ engagement indicated significant moderation (β = 0.131; p = 0.010), thereby supporting H₆ (see Fig. 5). Similarly, the moderation of AI information quality also yielded a significant effect (β = 0.121, p = 0.001) and supported H₇ (see Fig. 6). Finally, the moderation of responsible big data intelligence reported significant results (β-value = 0.214; p < 0.001) and supported H₈ (see Fig. 7). Cohen (1988) developed the f² statistic, categorizing it into three levels: small (f² = 0.02), moderate (f² = 0.15), and considerable (f² = 0.35). Table 7 shows that all the values are within the established thresholds.

Fig. 4.

Structural model.

Table 6.

Hypotheses testing.

Hypotheses	Paths	β-values	t-values	p-values	f-square	Remarks
H1	SE -> DD	0.270	2.555	0.000	0.049	Supported
H2	KD -> DD	0.441	9.211	0.000	0.271	Supported
H3	DD -> SP	0.458	8.296	0.000	0.276	Supported
H4	SE -> DD -> SP	0.124	2.306	0.002	–	Supported
H5	KD -> DD -> SP	0.202	3.182	0.002	–	Supported
H6	SE x Disruptive Innovation -> DD	0.131	2.580	0.010	0.021	Supported
H7	AIQx KD -> DD	0.121	3.380	0.001	0.022	Supported
H8	RAI x DD -> SP	0.214	4.117	0.000	0.071	Supported

Note: AIQ = AI-Information Quality, DD = Data-Driven Decision Making, DI = Disruptive Innovation, ECP = Economic Performance, ENP = Environmental Performance, KD = Knowledge Digitalization, RAI = Responsible Big Data Intelligence, SP = Social Performance, SE = Stakeholders Engagement.

Fig. 5.

Moderating role of disruptive innovation.

Fig. 6.

Moderating role of AI-information quality.

Fig. 7.

Moderating role of responsible big data intelligence.

Table 7.

Explanatory and predictive relevance of the study model.

Constructs	R^2	Q² (=1-SSE/SSO)
Data-Driven Decision Making	0.449	0.421
Sustainable Performance	0.497	0.504

Note: R² = Coefficient of Determination, Q² = Predictive Relevance.

To assess the explanatory and predictive relevance of our study model, we employed two methodologies: R² and Q². Falk and Miller (1992) stated that the coefficient of determination (R²) must exceed 10%. The R² statistic quantifies the extent to which a research model accounts for the observed data. The R² value in Table 7 indicates substantial significance. The R² values for DD and SP are 0.449 and 0.497, respectively. Geisser (1974) asserts that existing literature demonstrates that Q² must exceed zero. Table 7 presents empirical evidence indicating that the Q² values for DD and sustainable performance exceed zero. Consequently, the criteria for predictive relevance have been fulfilled.

Discussion and conclusion

This study examines sustainable performance through stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence in manufacturing organizations. The RBV theory is applied to explain the research framework.

The results indicate that stakeholder engagement has a significant positive impact on DD, supporting H₁. Raji et al. (2024) stated that data-driven insights, stakeholder participation, and analytical tools can expedite operations, leading to improved competitiveness. These results are in line with the RBV theory, as stakeholder engagement constitutes a valuable, rare, and inimitable resource, which embeds unique knowledge and information flow (Barney, 1991). Knowledge digitalization significantly determines DD, which supports H₂. According to the RBV, knowledge digitalization acts as a robust strategic resource that is rare and cannot be replicated. Colombari et al. (2023), while studying US automotive firms, reported that digitalization supports DD, which ultimately enhances firm performance. The integration of knowledge and data analysis is essential for decision-making frameworks; hence, organizations that intend to emerge in a more competitive world are imperative. Both H₁ and H₂ align with RBV, suggesting that firms’ resources (i.e., stakeholder engagement and knowledge digitalization) significantly contribute to their DD (Barney, 1991). Moreover, DD significantly enhances sustainable performance, facilitating effective resource utilization, risk minimization, and strategic decision-making—thus, H₃ is supported. DD is a decisive competence that enhances environmental, social, and economic performance through evidence-based interventions. Prior studies have indicated that DD has a significant impact on the performance of higher education institutions (Ashaari et al., 2021). Their findings suggest that dynamic capabilities are crucial and that firms with robust DD frameworks can effectively navigate environmental challenges without compromising their competitiveness. Furthermore, this study reported that DD mediates the relationships among stakeholder engagement, knowledge digitalization, and sustainable performance—thus, H₄ and H₅ are supported. The RBV supported this mediation that firms’ capabilities (i.e., DD) can explain the relationship between firms’ resources (i.e., stakeholder engagement and knowledge digitalization) and sustainable performance (Barney, 1991).

Disruptive innovation significantly moderates the stakeholder engagement–DD relationship, supporting H₆. Disruptive innovation refers to innovation that dramatically alters industries or markets through technological changes (Hopp et al., 2018). Fig. 5 illustrates that stakeholder engagement increased by 13.1% in DD with the involvement of disruptive innovation. Furthermore, disruptive innovation reinforces the interdependence between stakeholder engagement and DD by providing advanced tools and adaptable structures for accessing stakeholder inputs. Under disruptive conditions, stakeholder knowledge becomes more important and practical through innovative digital solutions. Thus, increased disruptive innovation enables companies to transform stakeholder involvement into rational and coherent DD.

Similarly, the study reported that AI information quality significantly moderates the knowledge digitalization–DD relationship, supporting H₇. AI information defines the extent to which digitalized knowledge is converted into valuable insights. The quality of AI information also increases data reliability, relevance, and interpretability. Consequently, more effective DD regarding AI information quality is possible in the context of digitalized knowledge. Quality information is vital for organizational success, as it enables making the right decisions at the right time. Reliable AI-generated data can boost decision-making effectiveness and reinforce strategic objectives in the production setting. Helu et al. (2016) stated that integrating innovative manufacturing technologies would enable decision-making. Fig. 6 illustrates that knowledge digitalization changes by 12.1% in DD because of AI information quality. Finally, responsible big data intelligence significantly moderates the DD–sustainable performance relationship, supporting H₈. This implies that DD increased by 21.4% because of responsible big data intelligence, leading to improved sustainable performance. Responsible big data intelligence ensures that the data utilization process remains ethical, transparent, and sustainability-oriented. Whenever big data intelligence is implemented responsibly, it is more likely that data-driven decisions will create long-term economic, social, and environmental value. In this manner, the beneficial influence of DD on performance sustainability is rendered more prominent within the highly responsible big data intelligence levels. The results favor the RBV, suggesting that responsible big data intelligence enhances the DD–sustainable performance relationship (Barney, 1991).

Theoretical implications

This study has several theoretical implications. This research extends the study by Han et al. (2024), who determined firm performance through knowledge digitalization, entrepreneurial orientation, novel business model innovation, and efficient business innovation. This study incorporates stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence to determine sustainable performance from the hitherto ignored RBV perspective (Barney, 1991; Cheng et al., 2026; Dal Maso et al., 2017; Dubey et al., 2019; Hopp et al., 2018; Mikalef & Gupta, 2021).

Previous research has explored the influence of stakeholder engagement on environmental performance based on the stakeholder theory (Huynh & Nguyen, 2025). Additionally, stakeholder engagement disclosure is employed to measure corporate financial performance in the context of the stakeholder theory (Galeotti et al., 2025). Knowledge digitalization is employed to measure high-tech firms’ performance from the perspective of the knowledge-based view (Han et al., 2024). Furthermore, studies have measured innovation ambidexterity through knowledge digitalization within the technology–organization–environment framework (Zhang et al., 2025). Studies have investigated the DD–firm performance relationship, with data integration as a moderator, from the perspective of the organizational information processing theory (Colombari et al., 2023). Prior research has largely ignored the significance of sustainable performance through stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence from the perspective of the RBV (Barney, 1991; Cheng et al., 2026; Dal Maso et al., 2017; Dubey et al., 2019; Hopp et al., 2018; Mikalef & Gupta, 2021). This research addresses this gap.

The theoretical contributions of this study significantly broaden the RBV of the firm in the context of digitalization and innovation. This research integrates several resources (i.e., stakeholder engagement, knowledge digitalization, AI information quality, and responsible big data intelligence) and capabilities (i.e., disruptive innovation and DD) to determine sustainable performance (Barney, 1991). The RBV traditionally treats concrete resources, such as physical capital, as a factor in the firm’s performance. However, the research identifies stakeholder engagement and knowledge digitalization as strategic intangible assets that play a key role in enhancing firm performance—especially in the manufacturing sector. The research broadens the horizons of the RBV through these elements; therefore, it is suggested that tangible and intangible resources also contribute to the formation of competitive advantage in the digital era (Mikalef et al., 2020). Moreover, the study delineates the mediating role of DD in linking stakeholder engagement and digital knowledge with sustainable performance. This supports the RBV, as the availability of valuable resources is not the only factor driving high performance; it also involves the capacity of those resources to be used effectively through strategic decision-making processes. One can also trace this back to the concept of dynamic capabilities, in which the RBV’s key postulate is the firm’s ability to realign its resources to meet evolving environmental requirements. The other significant contribution is the moderating role of disruptive innovation and the quality of AI information, which characterizes the impact of stakeholder engagement and knowledge digitalization on DD. It is but one method of modulating the old RBV frame, which, in many respects, entails the exploitation of internal resources, and the novel idea that external conditions and stakeholder relations can impact the performance of those resources.

Practical implications

The study has several implications for industry, academia, and policymakers. Policymakers may design policies to facilitate the integration of new technologies, such as AI, and disruptive innovation. In this manner, they will be able to support the digitalization of businesses and stimulate the creation of responsible big data intelligence. Providing enabling ecosystems that promote knowledge digitalization and stakeholder participation would also advance sustainable business practices, particularly in countries such as China, where the manufacturing industry is a major economic stakeholder. The AIQ augments the effect of knowledge digitalization in DD. The advantages for managers include ensuring that digitalized knowledge is digested by AI systems that provide accurate, timely, and relevant insights. This enhances the accuracy of decision-making, resource allocation, and responsiveness, as well as the translation of knowledge into actionable strategies. By managing AI quality through appropriate governance and oversight, managers can derive the optimal value from knowledge digitalization for sustainable, competitive performance.

Similarly, these findings can enable managers implement strategies to improve stakeholder engagement and stimulate knowledge digitalization. Moreover, organizations can utilize digital technologies to expand the boundaries of knowledge integration and assimilation. Stakeholder engagement and management’s focus on digitalizing knowledge indirectly improves organizations’ sustainable performance. This study outlines the direct impact of these factors on DD, which is a key factor in achieving sustainable performance. Managers need to focus on building robust relationships with stakeholders and remain keen to utilize data analytics to inform their business decisions. Managers will be able to position their firms ahead of competitors in a dynamic, competitive digital environment by adopting AI-based insights and promoting responsible data management. Moreover, disruptive innovation also offers managers a similar opportunity to mediate breakthrough innovations that distinctly position their businesses in the market. By integrating digitalization and DD values at the core of their business, companies can achieve greater efficiency, streamline their processes, and align their strategies and approaches with sustainable goals. Additionally, disruptive innovation would enable companies to adopt new business models, achieve a competitive advantage, and ensure the responsible and ethical utilization of big data. Managers should embrace new technologies and processes to rapidly capture, analyze, and respond to stakeholder inputs. This will render engagement activities actionable and enhance strategic decision-making, agility, and competitive positioning. With the development of an innovation culture toward enhanced stakeholder relations, managers can optimize the value of engagement in the promotion of data-driven and sustainable performance. Finally, digitalization, DD, and innovation adoption will empower all stakeholders in the business sector to establish a more sustainable and responsible business environment, not only benefiting individual companies but the society at large over the long term.

Limitations and future recommendations

This study has several limitations that future studies can address. First, this is a cross-sectional study, and it is challenging to assume that stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence will sustainably improve long-term performance. Future studies may examine these constructs longitudinally to mitigate the impact of several methodological biases. Moreover, future studies may examine the framework in a broader context; for instance, by conducting a cross-country comparison. Second, the narrow definitions of stakeholder engagement, knowledge digitalization, disruptive innovation, AI information quality, DD, and responsible big data intelligence may be clarified or expanded to analyze their long-term influence on sustainable performance. Third, DD serves as a mediating variable between stakeholder engagement, knowledge digitalization, and sustainable performance. Future studies can utilize technological intensity as a mediator between stakeholder engagement, knowledge digitalization, and sustainable performance. Fourth, this study has a singular focus—on managers. Other stakeholder views (e.g., employees or customers) could also be included in future studies to obtain a holistic understanding of how digitalization can drive sustainable performance across levels within the firm. Fifth, the moderating role of AI information quality and responsible big data intelligence is also presented in the study. However, it is possible to learn about other factors that may influence DD’s effectiveness in the future, such as corporate culture, leadership style, or external market conditions. Sixth, the research data on Chinese manufacturing companies are collected only from company managers, which may not be generalizable across other industries and geographic contexts. Future research may include other firms from multiple industries and countries to investigate the universality of the identified relationships. Finally, the study employed PLS-SEM for data analysis; future studies may implement CB-SEM or other models.

Funding

This work was supported by the Jiangsu Province Excellent Postdoctoral Program (2022ZB648).

CRediT authorship contribution statement

Muhammad Saeed Ashraf: Writing – review & editing, Writing – original draft, Conceptualization. Muhammad Usman: Writing – original draft, Software, Methodology, Formal analysis, Conceptualization. Mingxing Li: Writing – review & editing, Writing – original draft, Supervision, Funding acquisition, Formal analysis, Conceptualization. Ing Luboš Smrčka: Writing – review & editing, Writing – original draft, Conceptualization. Zhiqiang Ma: Validation, Supervision, Investigation, Funding acquisition.