1. Introduction
Businesses are increasingly concerned about environmental protection and achieving sustainable development goals (SDGs) has become a global priority, owing to their critical significance in economic development and well-being [
1]. Interest in SGDs has resulted in a significant increase in investment in research in enterprises that face shareholder pressure to implement sustainable development policies [
2,
3,
4]. Sustainable supply chain management (SSCM) attained academic and practitioner attention in achieving SDGs. There is a focus on waste generation, ecosystem disruption, and natural resource depletion. If an organization’s actions result in irrevocable harm to the ecosystem and fail to ensure safety, security, a living wage, healthcare, better employee working environment, and an improved standard of living for the neighboring community and society at large, there is a question mark over the organization’s sustainable performance [
5]. This trend has triggered academics and practitioners to pursue SSCM. SSCM has become an essential way for organizations to achieve SDGs and boost organizational performance [
6,
7,
8].
SSCM is referred to as “the management of material and information flows, as well as enterprise interaction, along the supply chain (SC) while taking into account all important components of sustainable development” [
9]. In contrast, many global SCs have been demonstrated to be opaque and prone to potential sustainability violations from a sustainability aspect [
10]. They argue that even if those breaches were not sufficient in and of themselves, they would still be sufficient to warrant a reevaluation of current SC practices, which negatively influence sustainable organizational performance. Traditional supply chain management (SCM) is unable to provide a suitable response to current stakeholder needs, not only because of its adverse environmental effects but also because of widespread public awareness of environmental issues in all areas of the world, including those that are difficult to reach [
11,
12,
13]. The green ecological approach and its emphasis on sustainability thus become an alternative method of administering public requests to regulate resource consumption in SC [
14,
15]. The study of Das (2017) conceptualized, developed, and validated a scale to measure the SSCM practices followed by a company and evaluated the organization’s performance on multiple dimensions of SSCM, among other things.
On the other hand, stakeholder theory gives a suitable theoretical viewpoint to understand stakeholder affiliations and has grown as the leading concept in the literature on corporate social responsibility (CSR) [
16,
17]. Similarly, an organization’s application of CSR activities does not stop at meeting the needs of its employees and the local community; it frequently results in a considerable improvement in SCM practices and performance [
5,
18]. However, we must provide answers to questions such as how SSCM occurs from the stakeholder theory perspectives (e.g., CSR); otherwise, the unknown nature of these difficulties would hamper our general knowledge of SSCM. Organizations considering adopting and implementing SSCM practices might benefit from the findings of a collaborative investigation of SSCM practices and outcomes from the perspective of CSR [
19]. Additionally, the integration of information technology has significantly aided and shaped organizational development in all areas involving SCM and sustainable performance [
5,
20,
21,
22]. Prior studies illustrate the critical significance of resources, competencies, and skills in ensuring a company’s competitiveness and performance in current competitive settings [
23]. Meanwhile, according to the dynamic capability theory, an organization’s competitiveness is determined by its capacity to utilize its capability. However, what type of capability can continue to provide value in a fast-changing environment [
21]?
Simultaneously, the importance of big data analytics (BDA) in steering organizational decision-making has garnered considerable attention in recent years [
24]. An increasing number of businesses are accelerating their BDA projects to gain essential insight that will finally offer them a competitive advantage and impact sustainable organizational performance [
25,
26,
27,
28]. According to the findings of [
25], the versatility of the BDA infrastructure is defined by its connectivity, compatibility, and adaptability. Numerous studies indicate that the accumulation of data has prompted many businesses to build analytical tools such as BDA in order to translate the data into valuable information that may aid in decision making and SC performance [
26,
27,
28,
29].
SSCM practices combine CSR and green SCM goals to assist organizations in achieving their operational and environmental performance at the micro-level, and ultimately improve the organization’s image in the eyes of stakeholders to improve its sustainable performance. Although the researchers described the relationship, no attempt has been made empirically to put the framework into practice in a real-world scenario, particularly the impact of BDA capabilities. The current study aims to close this gap by identifying the management perception of the impact of CSR (e.g., internal and external CSR) on the development of SSCM practices, which leads to long-term organizational success while using BDA capabilities.
To this end, this research aims to address and analyze the subsequent questions.
How does CSR (internal and external) affect SSCM practices in developing countries?
What effect do SSCM practices have on sustainable organizational performance?
What role do BDACs play in mediating the relationship between SSCM practices and sustainable organizational performance?
Both a theoretical discussion of the mechanisms behind the relationship of SSCM and organizational performance and empirical findings that could aid in identifying the mechanisms that prevail in various settings and contexts are lacking in the literature. Thus, this research aims to close a knowledge gap by analyzing the possible influence of CSR on SSCM practices that improve organizational performance via BDACs.
The rest of this article is organized as follows. The second part reviews the relevant literature and analyzes the limitations of previous studies; in addition, the hypotheses and framework of this paper are proposed. Part three is the method of data collection and analysis, and part four is the analysis and detailed discussion of the collected data results. Finally, the fifth part summarizes the research results, puts forward the implications of this research, and points out the limitations and prospective.
4. Results and Discussion
4.1. Respondent’s Profile
Table 1 summarizes the respondents’ demographic characteristics (gender, experience, and education). In total, 73.8 percent of respondents were male, while 26.3 percent were female. The findings indicate that 25.6 percent of respondents had between one and three years of job experience, 37.2 percent had between four and six years of work experience, and 37.2 percent had more than six years of work experience. A total of 14.4 percent of respondents held an undergraduate degree, 35.3 percent held a graduate degree, and 28.1 percent held a postgraduate degree. Only 22.2% of respondents reported having additional degrees or professional education. We can assume from these findings that our respondents are well educated and experienced enough to comprehend the questionnaire.
4.2. The Measurement Model
4.2.1. Reliability and Convergent Validity
Convergent validity describes the degree to which distinct indicators within the same structure are correlated [
101]. This study employed Smart-PLS v3 to perform confirmatory factor analysis (CFA) on each item to determine its convergent validity. The reliability and convergence validity analyses for this investigation are summarized in
Table 2. Cronbach’s alpha values for all constructs were between 0.890 and 0.949, above the suggested level. According to the threshold values, the composite reliability (CR) ranged from 0.924–0.957, and the average variance extracted (AVE) was 0.698–0.764. As a result, the findings in this research demonstrate that there are no issues with convergence validity or reliability.
4.2.2. Discriminant Validity
Discriminant validity is defined as “the extent to which variables differ empirically” [
101]. Three approaches were used in this study to assess discriminant validity. First, the researchers related each factor’s association to the square root of AVE. Second, they assessed the relevance of the survey items using item loadings and cross-loadings. Thirdly, they determined the heterotrait–monotrait ratio (HTMT) [
101,
105,
106].
As illustrated in
Table 3, the link between the constructs and the square root of AVE was used to determine the instrument’s validity, a criterion known as the Fornell–Larcker criterion. In
Table 3, the diagonal values suggest that the square root of AVE is greater than the correlation coefficient between the variables. The findings indicate no discriminant validity problem [
107].
Prior research has used cross-loading criteria to determine discriminant validity [
20,
106,
108]. According to the literature, each entry’s load should be greater than the burden of the succeeding construct, and the entry load is also regarded as a threshold. The item loads and cross-loads for all linked values are shown in
Table 4, demonstrating that the factor item loads are more significant than those of other potential factors. This implies that the distinction is sufficiently valid to satisfy the cross-loading criterion.
Finally, the ratio of HTMT is near to one, indicating a lack of discriminant validity in path testing [
106]. The HTMT approximates the correlation between numerous variables (more precisely, the upper boundary). Henseler et al. (2016) advised that HTMT values should be less than one. As a result, we also use the HTMT ratio; as shown in
Table 5, the highest value is 0.604, which is less than the suggested limit, indicating that discriminant validity is sufficient.
4.3. Method Bias and Multicollinearity
The researchers utilized SPSS v26 software to conduct Harman’s single factor test to determine whether this study had a common method bias (CMB) issue [
109,
110,
111]. The scores signaled that the first factor explained 39% of the variation, less than the 50% requirement for CMB [
110,
112]. Simultaneously, the inner variance inflation factor (VIF) was employed in Smart-PLS v3 to identify any CMB issues. Kock (2015) states that this number should not exceed 3.3. These values vary between 1.00 and 1.728, demonstrating that CMB was not an issue in this investigation. Meanwhile, the outer VIF was employed to test for multicollinearity. The highest VIF value found in this study was 4.82, which is under the suggested threshold value of 10 [
113,
114]. As a result, no evidence of significant multicollinearity was discovered.
4.4. The Structural Model
After confirming the model’s reliability and validity, Smart-PLS v3 was used to quantify the hypothetical connectivity along the standardized path [
101,
109]. Compared with covariance-based SEM, this software is better suited for dealing with potential paths of formative and reflected [
104,
115]. It is a powerful tool for measuring indirect relationships of path models using PLS-SEM [
103]. The beta coefficients are depicted in
Figure 2. The SEM path’s significance level was determined using bootstrapping, with a total of 5000 resampling’s. The suggested model’s descriptive capacity can be quantified by the explanatory deviation of the results (i.e., the R-squared value).
SSCMP’s adjusted R-squared value was 0.404, indicating that the internal and external CSR accounted for 40.4 percent of the variation in the SSCM practices. Meanwhile, the R-square for BDAC was 0.317, indicating that SSCM practices accounted for 31.7 percent of BDAC variation. OP’s R-squared value was 0.367, while EP’s R-squared value was 0.361, showing the factors’ effective participation.
As per the SEM results in
Figure 2, all exogenous constructs in this investigation are positively linked with endogenous structures. The values of the beta coefficients for bootstrapping, the significance of the direct impacts, and path analysis are all included in
Table 6. The results suggest that the t-statistic value is greater than the recommended value of 1.96, indicating the existence of a meaningful association between the quasi-variables [
102,
103]. Additionally, the
p-value is included to indicate its significance.
As shown in
Table 6, the SEM analysis confirms that the path’s analysis coefficient between ICSR and SSCM practices is 0.293, which is statistically significant at the 0.000 level. These findings indicate that internal CSR has a positive effect on SSCM practices. The beta coefficients between ECSR and SSCM practices are 0.439 (
p = 0.000), suggesting that external CSR strongly affects SSCM practices. Based on these findings, we may conclude that organizational CSR activities, whether internal or external, significantly and positively contribute to developing SSCM practices (as a second-order construct) in automotive organizations. Therefore, H1a and H1b are statistically supported based on these findings.
The SSCM practices’ beta coefficient also demonstrated a strong positive effect on BDAC. In the path analysis, the measurement value was 0.563 (p = 0.000). As a result, H2a is supported. SSCM practices were linked to OP and EP positively. The beta coefficient demonstrates a statistically significant and positive association between SSCM practices and OP (b = 0.409, p = 0.000) and between SSCM practices and EP (b = 0.362, p = 0.000). As a result, H2b and H2c are approved.
Additionally, the findings indicate that BDAC significantly impacts the automotive organization’s OP and EP. According to
Table 6, the association between BDAC and OP is 0.272 (
p = 0.000), whereas the relationship between BDAC and EP is 0.178 (
p = 0.002). This indicates that BDAC has a considerable impact on an organization’s sustainable performance. As a result, H3a and H4a are likewise endorsed.
Additionally, this study examined the mediating effect of BDAC on the connection between SSCM practices and sustained organizational performance. The results suggest that BDAC mediated the association between SSCM practices and sustainable organizational performance positively and significantly. However, the direct association between SSCM practices and sustainable organizational performance (OP and EP) remains strong—a partial mediation—thus, H3b and H4b are supported. Furthermore, the results indicate that improving an organization’s operational performance can positively affect its environmental performance. The correlation between OP and EP was determined as 0.166, with a significance level of 0.006.
Overall results show that all proposed hypotheses are supported and statistically significant because the t-statistics and p-values meet the threshold. We can conclude that the proposed model is statistically significant and acceptable from the results.
5. Conclusions, Implications, and Future Directions
5.1. Conclusions
This study investigated the influence of CSR, namely internal and external CSR on SSCM practices, which eventually results in achieving sustainable organizational performance (OP and EP) using its BDACs. SSCM practices were measured using second-order formatting constructs (EMP, OPR, and SCI). For this objective, the authors performed an empirical analysis using data acquired from Pakistani automotive manufacturing managers through a survey questionnaire to evaluate the proposed hypotheses.
Overall, the findings of this study suggest that CSR (both internal and external CSR) has a positive impact on SSCM practices in automotive manufacturing organizations. The findings show that internal or external CSR activities can improve organizational SSCM practices. To improve SSCM, organizations should establish and improve their internal CSR and external CSR activities. These findings are also consistent with the prior studies [
17,
116]. Meanwhile, as described in the second research question of this study, the most pressing problem we encounter is identifying the influence of SSCM practices on organizational performance, and the outcomes prove that SSCM practices have both direct and indirect impacts on sustainable organizational performance. These findings are also supported by the previous studies of [
14,
117].
The advancement of information technology and sensor technology has made it possible to collect large amounts of data from each SC partner, which has been beneficial in reducing the lack of knowledge about sustainability in SCs and improving the overall performance of organizations [
80,
118]. Therefore, according to the findings of this study, BDAC positively mediates the association between SSCM practices and organizational performance (OP and EP). It can be projected to empower managers to manage their SCs more effectively while improving organizational performance.
5.2. Theoretical Implications
This work adds to the intellectual property of the SSCM and SDG literature by forming prior theoretical and empirical investigations on SCM. First, this study extends stakeholder theory in SSCM by probing the association between two types of CSR (internal and external) and SSCM practices. While previous research has concentrated on reactive responses to stakeholder pressure in various fields, including SC [
17,
39,
43,
119], this study applied stakeholder theory to the field of SSCM to uncover previously unknown conclusions. The empirical outcome of the proposed framework reveals that this research contributes to expanding the notion of SSCM practices. Thus, we anticipated that internal and external CSR would significantly enhance organizational SSCM practices, ultimately assisting in achieving sustainable performance. These findings may aid in advancing our understanding of the link between CSR and SSCM.
Second, this study empirically explores and inflates the literature on the link between SSCM practices and sustainable organizational performance using BDAC, adding to the body of knowledge on this subject. Several prior researchers have demonstrated that BDACs play an increasing role in decision-making for sustainable development and performance [
17,
25,
63,
120]. Thus, this study adds to the body of knowledge on SSCM by empirically examining the mediating influence of BDAC on the link between SSCM and sustainable organizational performance.
Finally, this study examined SSCM practices from three distinct perspectives—EMP, OPR, and SCI—and established their significance in ensuring the sustainable performance of automotive manufacturing organizations (one of the largest manufacturing sectors). It may add a new dimension to established theoretical concepts of SSCM and sustainable performance, which can be employed in a wide range of diverse contexts.
5.3. Practical Implications
On the practical level, this study has the following implications. First, this study describes how CSR activities in manufacturing companies increase SSCM practice. Internal and external CSR initiatives encourage firms to adopt SSCM techniques to accomplish SDGs and improve organizational performance because organizations are responsible for business and social actions within their own premises and practices outside their own premises. Therefore, organizational policymakers should emphasize internal and external CSR initiatives that build a force to become involved in SSCM practices.
Second, our research demonstrates that organizations can reap the benefits of BDAC to improve their overall performance further. BDAC could play a constructive role in assisting policymakers and researchers design and implement policies, strategies, and practices that enable organizations to achieve both operational and environmental performance. This conceptual paper advises that organizations, managers, and entrepreneurs align their CSR activities, SSCM practices, and BDAC strategies to achieve sustainability in organizational performance.
Third, manufacturing organizations, especially in developing countries such as Pakistan, need to be aware of the areas of internal and external social responsibility related to achieving SSCM goals. Furthermore, it turns out that organizations need to develop an IT infrastructure capable of handling BDA for ultimate organizational performance.
Finally, this study recommends that policymakers reconsider the country’s environmental policies, as the manufacturing sector is seen as operating at a loss due to poor observable sustainability practices. To encourage organizations to implement environmental policy guidelines, policymakers must improve access to SSCM practices and technological development by building environmental practices for active learning programs, providing financial support, and incentivizing collaboration with customers and SC partners.
5.4. Limitations and Prospectives
The limitations of this study are few and point out the scope of further research. First, the scope of this study was confined to cross-sectional data from automotive manufacturing organizations collected at a single point in time. In the future, a longitudinal investigation will be conducted to better understand the sequence of relationships that exist between CSR, SSCM practices, BDAC, and sustainable organizational performance. Second, the data was gathered in Pakistan, where the working environment of automotive manufacturing organizations may be different from that seen in other nations. Therefore, data from a different geographic area should be used in the future to confirm the conclusions of this study. Finally, the model proposed in this study is a first step toward describing the link between CSR, SSCM practices, BDAC, and sustainable organizational performance. Future studies may incorporate additional relative factors such as external support, environmental instability, and market intensity to understand this concept.