43 Advances in Production Engineering & Management ISSN 1854-6250 Volume 20 | Number 1 | March 2025 | pp 43–60 Journal home: apem-journal.org https://doi.org/10.14743/apem2025.1.526 Original scientific paper Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Liang, P.P. a,* , Jili, H.H. a , Lv, Y.Q. a,* , Xu, Y. a a School of Economics and Management, Anqing Normal University, Anqing, P.R. China A B S T R A C T A R T I C L E I N F O In an increasingly complex and turbulent global environment, achieving resil- ience in manufacturing supply chains has become a critical strategic priority. Drawing on a sample of 300 manufacturing firms, this study examines both the net and configurational effects of supply chain governance mechanisms and dynamic digital capabilities on supply chain resilience. Using structural equation modeling and fuzzy-set qualitative comparative analysis (fsQCA), the findings reveal that: Contractual governance, relational governance, digital sensing capability, digital resource integration, digital-driven innovation, and digital-enabled business capabilities each have a positive impact on manufac- turing supply chain resilience. In the overall sample, only relational govern- ance demonstrates a relatively strong individual effect, while none of the six governance or digital capability dimensions serve as necessary conditions for high resilience in subsample analyses. For high-tech manufacturing firms, two resilient configurations are identified: 1) basic digital enablement with strong governance synergy, and 2) advanced digital enablement with strong govern- ance synergy. In contrast, non-high-tech firms exhibit three distinct resilient configurations: 1) digital integration–driven, 2 advanced digital enablement with relational governance dominance, and 3) dual-core digital enablement with robust governance synergy. These insights provide nuanced theoretical contributions and practical implications for configuring governance and digi- tal strategies to build sustainable supply chain resilience in the manufacturing sector. Keywords: Manufacturing supply chain; Resilience; Supply chain governance; Dynamic digital capability; Fuzzy-set qualitative comparative analysis (fsQCA); Configuration analysis *Corresponding author: 614lpp@163.com (Liang, P.P.) lvyouqing2011@163.com (Lv, Y.Q.) Article history: Received 12 November 2024 Revised 16 April 2025 Accepted 20 April 2025 Content from this work may be used under the terms of the Creative Commons Attribution 4.0 International Licence (CC BY 4.0). Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. 1. Introduction Manufacturing supply chain resilience (MSCR) is intrinsically linked to national security and economic development. Over the past decade, manufacturing supply chains (MSCs) have been increasingly vulnerable to external shocks, including trade protectionism, technological embar- goes, and global public health crises. Unanticipated disruptions—such as core component cut- offs due to sanctions—can severely impair operational performance and even threaten the sur- vival of individual firms, triggering cascading effects across the entire supply chain network. The specialization and fragmentation of production processes have further increased systemic risks, exacerbated by the need for tighter coordination and real-time demand-supply matching. Numerous real-world cases underscore the disruptive potential of such events. For instance, the 2023 strikes in France—sparked by public dissatisfaction with pension reforms—caused major disruptions to both maritime and inland freight flows. Similarly, the COVID-19 pandemic dramati- cally reshaped global supply chain structures. According to Dun & Bradstreet, up to 94 per cent of Liang, Jili, Lv, Xu 44 Advances in Production Engineering & Management 20(1) 2025 the world's top 1,000 companies experienced supply chain disruptions, with the automotive and electronics manufacturing sectors being the most affected. While some disruptions may be man- ageable in the short time, others pose significant threats to the long-term resilience and competi- tiveness of MSCs. As a result, the development of robust, long-term resilience mechanisms has garnered increasing attention from both industry practitioners and academic researchers. Largely, previous organizations relied significantly on supply chain governance (SCG) to miti- gate disruptions and build resilient supply chains. Specific SCG measures can fall into two broad categories, namely contractual governance and relational governance. The former aligned with transaction cost theory emphasizes the use of contracts to safeguard against opportunism and conflict. The latter grounded in the social exchange theory (SET) and relational exchange theory, aiming to curb opportunism by instituting relationship-based norms and developing trust – building mechanisms. Nevertheless, the inherent looseness of supply chain structures, coupled with the bounded rationality of decision-makers, amplifies the vulnerability to opportunistic risks. This, in turn, undermines the efficacy of supply chain governance, particularly within dy- namically evolving environments. According to the dynamic capabilities view (DCV) proposed by Teece et al., organizations of- ten develop dynamic capabilities to mitigate the impact of unexpected risks on supply chain per- formance [1], particularly under highly competitive pressures and in dynamic environments, by integrating, building, and reconfiguring resources. Based on Dubey et al. [2] we argue that dy- namic capabilities are multi-faceted, encompassing both the ability to capture new opportunities and risks and the ability to utilize available resources and technologies to address them. Crucial- ly, firms are not necessarily strong across all types; the appropriate response to supply chain disruptions is to leverage the specific competencies in which they excel. Moreover, digital- enabled technologies as a key option in crisis scenarios play a significant role in improving sup- ply chain resilience. To maintain competitiveness during turbulent times, organizations are re- quired to develop their digital capabilities for enhancing supply chain resilience to remain com- petitive in the digital era. Integrating the above two perspectives, MSCR features multiple concurrent causalities and encompasses different levels. This necessitates a configuration perspective to uncover the mul- tiple equivalent configurations that build supply chain resilience. To better address issues such as “enterprises being ‘willing but unable’ when facing risks” or “possessing strong dynamic ca- pabilities yet remaining "powerless to reverse the situation”, this study incorporates supply chain governance into the configuration analysis framework and matches it with dynamic capa- bility. This approach aims to explore the influencing factors and their configuration mechanisms of supply chain resilience. The main problems to be solved in this paper are as follows. • How does supply chain governance initiative and dynamic digital capabilities affect MSCR? • Which factor configurations may constrain MSCR? • What paths to achieving high MSCR with different technological levels? Compared with the extant literature, this study makes three primary contributions: • We develop and empirically validate a theoretical framework elucidating the synergistic mechanisms through which supply chain governance and dynamic capabilities jointly en- hance MSCR. • Distinguishing from net effect studies, we innovatively adopted the fsQCA approach to ex- plore the configuration effect of multiple factors on MSCR, in response to the call from ac- ademics for mixed studies of mainstream statistical methods and qualitative comparative analysis methods. • Considering the "causal complexity" behind supply chain resilience management, the equivalent driving mechanisms for achieving high MSCR (e.g., different routes the same destination) are revealed, which can provide an actionable scenario framework for en- hancing MSCR. Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 45 The rest of this study is organized as follows. Section 2 presents the theoretical framework and the hypotheses development. Section 3 outlines the research methodology, including the questionnaire design and data-collection process. In Section 4, we conduct an empirical analysis of MSCR using SEM, while Section 5 examines the MSCR mechanism through a hybrid approach combining NCA and fsQCA. Finally, the main findings and conclusions are presented and discus- sed in the final two Sections. 2. Theoretical framework and research hypotheses 2.1 Supply chain resilience (SCR) SCR refers to the ability of interconnected supply chain enterprises to maintain their own sys- tem stability and avoid chain breakage when exposed to internal and external shock risks, as well as the ability to anticipate and react to future uncertainty [3]. Subsequently, some scholars have further extrapolated this concept across the dimensions: recovery from disruptions, risk resistance, and complexity adaptation [4]. The SCR measurement metrics can be categorized into four groups, respectively: core capability indicators (e.g., supply chain flexibility, visibility agility), recovery metrics (degree/time to restore original state), financial performance, and network topology metrics [5, 6]. Currently, the research paradigm on SCR strategies has evolved from "static to dynamic" and" traditional to complex". Early studies grounded in the static resource-based theory (RBT), em- phasized cooperation production, supply chain network structure design, supply chain redun- dancy design, contract design and governance [7], etc. To address RBT’s limitations in analyzing technological shifts, changing consumer preferences, and dynamic competition, scholars have begun to apply DCV to reveal the antecedents, processes and outcomes of SCR [8]. The DCV pos- its that mere possession of scarce resources is insufficient for competitive advantages—these resources must be reconfigured and deployed effectively. 2.2 SCG and MSCR Contractual governance relies on written agreements to regulate relationships among manufac- turing supply chain members. These contracts include explicit terms that clearly define the re- sponsibilities and obligations of each party [9]. When unexpected disruptions occur, clearly de- fined responsibilities help prevent task shirking and interpersonal conflicts. They also facilitate effective information sharing through standardized parameters such as price, quantity, logistics, and quality, thereby enhancing the efficiency of uncertainty management [10]. Moreover, com- prehensive contracts address a broad spectrum of potential risks and corresponding counter- measures. This provides partners with predetermined rules and procedures, which in turn re- duces decision-making uncertainty and promotes supply chain stability. Legally enforceable contracts also ensure compliance, as violations trigger timely corrective actions or penalties. This mechanism deters opportunistic behavior during crises and strengthens the supply chain’s systemic resistance to risk [11]. Hypothesis H1a: Contractual governance positively affects MSCR. Relational governance emphasizes coordinating each other's behaviors and developing long- term relationships, through the construction of social mechanisms such as trust, commitment and reciprocity [12]. As a cornerstone of social exchange, trust motivates supply chain partners to share critical information and resources, facilitating joint actions for rapid operational adap- tation [13]. Consequently, institutionalizing trust mechanisms is pivotal for cultivating risk- resilient manufacturing supply chain. Relational commitment as another ingredient of SET, in- stills confidence in manufacturing supply chain members, put forth the essential effort and en- hances MSCR by creating reciprocally beneficial exchanges [14]. According to SET, reciprocity is mutual exchanges that partners consider fair and provide long term gratification because behav- ior by an exchange partner will encourage reciprocal action by other partners. From a long-term view, reciprocity mechanisms can significantly enhance MSCR by facilitating resource sharing, Liang, Jili, Lv, Xu 46 Advances in Production Engineering & Management 20(1) 2025 risk sharing, and collaborative innovation, enabling partners to better cope with uncertainties [15, 16]. Hypothesis H1b: Relational governance positively affects MSCR. 2.3 Dynamic digital capability and MSCR Digital capability is commonly defined as the abilities endowed by digital technologies that re- spond quickly to environmental changes. Following Teece et al. [17] and Sousa-Zomer et al. [18], a defining attribute of digital technologies amid continuous disruption is their proactive envi- ronmental scanning capability. Digital sensing capability refers to an organization's ability to collect, analyze and interpret digital information from its internal and external environments [19]. This helps with scanning the external environment for unexpected disruptions and taking preventive actions. For instance, manufacturers with strong digital sensing can predict a sudden surge in demand for a particular product and adjust production accordingly. They can also min- imize the impact of disruptions on their supply chain, by proactively managing inventory, adjust- ing production schedules, and collaborating with suppliers. Hence, digital capabilities must pos- sess the seizing ability. Hypothesis H2a: Digital sensing capabilities positively affect MSCR. Sirmon and Hitt [20] argue that resource integration refers to an organization’s ability to create economic value by assembling, combining, optimizing and rationally allocating the internal and external resources. Digital resources are a key source for building dynamic digital capabilities. This dimension focuses on the ability to combine and optimize digital resources across the en- tire manufacturing supply chain, involving integrating data from different systems and process- es to create a unified operational view [21]. When manufacturers are capable of effectively inte- grating digital resources, they can eliminate redundancies, enhance communication among partners, and respond swiftly to changes, resulting in improved coordination and adaptability amid disruptions. Digital resource integration capability focuses on data management, resource orchestration and process integration, whereas digital-driven innovation capability concentrates on how digital technologies can be leveraged to drive innovation [22]. In the manufacturing sec- tor, digital-driven innovation refers to embedding digital technologies into the manufacturing process to drive improvements and create new opportunities [23]. This form of innovation goes beyond simply adopting digital tools, it entails a fundamental transformation in how manufac- turing operations are conceived, executed, and managed. For instance, Internet of Things (IoT) sensors collect real-time data from devices and installations, providing valuable insights to make decisions for optimizing processes, predicting maintenance needs and addressing quality devia- tions. What's more, the innovative application of blockchain technology enables greater trans- parency and traceability in manufacturing supply chains, thereby reducing the risk of disruption. Hypothesis H2b: Digital resource integration capacity positively affects MSCR. Hypothesis H2c: Digitally- driven innovation capability positively affects MSCR. In practice, numerous manufacturing firms have successfully leveraged digital technology to achieve business transformation, and digital-driven business capability has gained significant attention from organizational scholars [24]. As an important component of dynamic digital ca- pability, digital-driven business capability refers to an organization’s proficiency in utilizing digi- tal technologies, data resources and digital mindset to drive business growth, optimization and transformation [25]. This capability manifests in various ways, including innovating business models, formulating more effective marketing strategies, expanding sales channels and custom- er bases, and optimizing business resourcing through various digital means [26, 27]. For in- stance, supported by digital technologies, firms can generate new value growth through busi- ness transformation, collaboratively address uncertainties and risks by breaking down infor- mation silos, and improve operational efficiency by creating a more agile MSCs network. It fur- ther helps to enhance MSCR through improved agility and resistance. Hypothesis H2d: Digitally-driven business capability positively affects MSCR. Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 47 The SEM-based theoretical model is shown in Fig. 1. Moreover, it's clear that a dynamic pro- cess of developing supply chain capacity can enhance MSCR. However, some cases disclose that not all supply chains with dynamic digital capabilities are necessarily capable of actively tackling supply chain disruptions. Taking Motorola, once a leading player in the mobile phone industry, as an example, exhibited poor coordination with suppliers during the product design and manu- facturing phases and was incapable of responding promptly to market demands. Eventually, it underwent multiple acquisitions and restructurings. This implies that even if an enterprise possesses outstanding dynamic digital capabilities, without choosing effective governance initiatives, it can still lead to the enterprise's cessation of operation. Consequently, scholars have increasingly recognized that the dominant role of SCG initiatives should not be ignored and have attempted to deeply explore the internal mechanisms of MSCR from the relationship management perspective [28]. As a matter of fact, dynamic digital capabilities can provide more advanced tools and means for supply chain governance. Converse- ly, appropriate supply chain governance initiatives can facilitate the effective application and integration of digital resources, thereby avoiding resource waste and information silos. These two dimensions exhibit a superior synergistic effect, which has not been adequately considered in existing research [2, 29]. Therefore, this paper proposes a conceptual model of MSCR from the configurational perspective, as presented in Fig. 2. Contractual governance Relational governance Digital sensing capability Digital resource integration capacity Digitally- driven innovation capability Digitally- driven business capability Manufacturing supply chain resilience H1a H1b H2d H2c H2b H2a Fig. 1 The SEM-based theoretical model Supply chain governance Supply chain governance Dynamic digital capability Dynamic digital capability Synergistic linkage Synergistic linkage Configuration matching Relational governance Contractual governance Digital resource integration capacity Digital sensing capability Digitally-driven business capability Digitally-driven innovation capability High MSCR High MSCR Non-high MSCR Non-high MSCR Fig. 2 Conceptual model based on a configurational perspective 3. Research methodology 3.1 Method of hybrid NCA and fsQCA As a case-oriented method, Qualitative Comparative Analysis (QCA) is designed to capture the causal complexity and interdependence among multiple conditions in configurational research. Among its variants, fsQCA has emerged as a mainstream analytical paradigm, as it accommo- dates continuous variables and partial membership, thereby enhancing both analytical practical- ity and generalizability. However, fsQCA primarily identifies configurations of antecedent condi- Liang, Jili, Lv, Xu 48 Advances in Production Engineering & Management 20(1) 2025 tions associated with an outcome, offering qualitative insight rather than quantitative assess- ment of the degree to which specific conditions are necessary. Especially in fuzzy-set contexts, necessity is not a binary concept (“yes” or “no”) but rather a matter of degree. To address this limitation, Necessary Condition Analysis (NCA) can be integrated with fsQCA. NCA quantitatively assesses the extent to which a condition is necessary for a particular outcome, thereby comple- menting fsQCA and enriching the explanatory power of social science theories. This hybrid ap- proach significantly enhances both descriptive precision and theoretical robustness. This study begins by using SEM to explore the effects of dynamic digital capabilities and SCG initiatives on MSCR. Following this, the NCA method is employed to identify whether certain dimensions of digital capabilities or SCG initiatives constitute necessary conditions for high MSCR, and if so, to what degree. Concurrently, the QCA approach is applied to verify the robust- ness of these necessary condition findings. Finally, fsQCA is utilized to delve into the complex causal mechanisms through which dynamic digital capabilities and SCG initiatives shape high levels of MSCR. A heterogeneity analysis is also conducted across industries with differing levels of technological sophistication. As a configurational method, fsQCA conducts cross-case compar- ative analysis from a holistic perspective. It aims to uncover which combinations of conditions lead to the presence—or absence—of the outcome. This approach is well-suited for examining the multifactorial and complex formation mechanism of MSCR. 3.2 Questionnaire design We employ the questionnaire survey data from manufacturing firms to study the impact of the configuration mechanism between dynamic digital capabilities and SCG initiatives on MSCR. To ensure the sample reliability and validity, this questionnaire mainly draws on the mature con- tent that has been published in domestic and foreign literature (as shown in Table 1). Primary data collection utilized a five-point Likert scale, where values ranging from 1 (“strongly disa- gree”) to 5 (“strongly agree”), capturing progressive agreement levels across all variables. Table 1 Measurement items of each variable Constructs Measurement items References Contractual governance (CG) CG1: Sign an agreement with supply chain partners. CG2: The agreement improves product quality. CG3: The agreement ensures that both sides understand the product. CG4: The agreement improves communication efficiency with partners. [30, 31] Relational governance (RG) RG1: Have close cooperative relationship with supply chain partners. RG2: Share the long-term and short-term plans with partners. RG3: Trust the commitments made by partners. [31] Digital sensing capability (DSC) DSC1: Accurately predict industry technology trends leveraging digital technology. DSC2: Fully track the changes and trends of customer needs leveraging digital technology. DSC3: Identify opportunities brought by competitive changes leveraging digital technology. DSC4: Identify opportunities brought by supply and demand changes (e.g., changes in supplier quotations, emerging supply markets, and changes in consumer preferences) leveraging digital technology. [2] Digital resource integration capability (DRIC) DRIC1: Be able to effectively achieve the transfer and combination of digital resources. DRIC2: Be able to effectively allocate and utilize data resources. DRIC3: Be able to obtain abundant data resources from the supply chain network. [22, 32] Digitally-driven innovation capability (DIC) DIC1: Have a high tolerance for losses stemming from innovation. DIC2: Use digital means to introduce more new products and services. DIC3: Use digital means to continuously improve the manufacturing pro- cess. DIC4: Use digital means to transform production mode at a faster speed. [33] Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 49 Table 1 (Continuation) Digitally-driven business capability (DBC) DBC1: Enabled by digital technology, we can promptly execute counter- measures once major competitors target our customers with promotional activities. DBC2: Digital technology empowers us to execute timely and effectively marketing strategies. DBC3: Leveraging digital technology, we can proficiently acquire and assim- ilate fundamental and pivotal business technologies. DBC4: Digital technology facilitates the continuous development of initia- tives aimed at reducing production costs. DBC5: Digital technology allows for the efficient organization of production processes. DBC6: Digital technology enables the efficient allocation of resources across production and other departments. [25] Manufacturing supply chain resilience (MSCR) MSCR1: Preparedness for potential disruption impacts across the supply chain. MSCR2: Rapid respond to supply chain disruption events MSCR3: Maintain basic business operations in the event of disruption. MSCR4: Preserve the desired level of control over structure and function in the event of disruption. MSCR5: Recover speed to its original state after being disrupted. MSCR6: Adaptive transformation to an improved post-disruption state. [34] 3.3 Sample selection and data collection To empirically test the proposed hypotheses, data were collected from manufacturing firms in China. The survey participants included top and middle managers, and confidentiality of their responses was strictly maintained. The qu items were adapted from validated measurement scales in prior research, with item wording carefully adjusted to align with our research context. A pilot test was conducted with 20 enterprises to finalize the questionnaire, which was refined for a large-scale distribution. These procedures ensured reliability and validity. Data collection employed both field research and online distribution methods, yielding a total of 300 valid re- sponses, including 102 from field surveys. Sample characteristics are summarized in Table 2, while Table 3 presents the descriptive statistical for all variables. The classification of industry technology level follows the OECD’s high-tech industry classifi- cation standard, aligned with China’s Industrial Classification of National Economy (GB/T 4754- 2017). According to this criterion, manufacturing sectors with relatively high R&D intensity are categorized as high-tech manufacturing industries. These encompass six major categories: aero- space vehicle and equipment manufacturing, electronic and communication equipment manu- facturing, computer and office equipment manufacturing, pharmaceutical manufacturing, medi- cal equipment and instrument manufacturing, and information chemical manufacturing. The remaining industries are classified as non-high-tech manufacturing industries. Table 2 Descriptive statistics of the sample Sample characterization Norm Sample size Percentage (%) Firm information Firm size (no. of employees) ≤ 50 51-200 201-500 501-1000 > 1000 7 63 114 74 42 2.3 21.0 38.0 24.7 14.0 Firm age 1-3 4-6 7-9 ≥ 10 10 33 62 195 3.3 11.0 20.7 65.0 Industry technology level High-technology Non-high-technology 104 196 34.7 65.3 Respondent information Educational attainment Below bachelor's degree Bachelor's degree Master's degree or above 18 235 47 6 78.3 15.7 Current position Top manager Middle manager 20 280 6.7 93.3 Liang, Jili, Lv, Xu 50 Advances in Production Engineering & Management 20(1) 2025 Table 3 Descriptive statistical analysis of variables Statistical indicators Antecedent condition Outcome variable Supply chain governance Dynamic digital capability Manufacturing supply chain resilience CG RG DSC DRIC DIC DBC / Average value 4.242 4.231 3.970 4.148 3.961 4.176 3.968 Median value 4.500 4.667 4.250 4.667 4.200 4.500 4.333 Standard deviation 0.770 0.874 0.944 0.921 0.806 0.890 0.927 Minimum value 1.250 1.333 1.250 1.333 1.400 1.500 1.500 Maximum value 5.000 5.000 5.000 5.000 5.000 4.833 5.000 4. Empirical analysis of MSCR mechanism based on SEM 4.1 Reliability and validity The reliability of the scale data is typically assessed using two indicators: internal consistency coefficient (Cronbach's α coefficient) and composite reliability (CR value). In this study, SPSS 26.0 software was used for analysis. The results presented in Table 4 show that the Cronbach's α and CR value for all variables exceed 0.8, indicating that the scale used in this study has good reliability. Validity analysis includes four aspects: content validity, structural validity, convergent validi- ty, and discriminant validity. Content validity has been addressed previously. Structural validity, convergent validity, and discriminant validity were examined using confirmatory factor analysis (CFA) with Amos 26.0 software for testing. Prior to CFA, KMO and Bartlett's sphericity test were firstly carried out by using SPSS 26.0 software. The results of the KMO and Bartlett's sphericity test indicate that the data sample is suitable for factor analysis (the KMO value is 0.917 > 0.6; the Bartlett's sphericity test is significant with 𝑝𝑝 = 0.000 < 0.05). CFA results demonstrated a good model fit: 𝑥𝑥 2 /𝑑𝑑𝑑𝑑 = 1.128< 3, 𝑅𝑅 𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅 = 0.038 < 0.05, 𝐺𝐺 𝐺𝐺𝐺𝐺 = 0.914, 𝑁𝑁𝐺𝐺𝐺𝐺 = 0.922, 𝐶𝐶 𝐺𝐺𝐺𝐺 = 0.990, 𝐺𝐺 𝐺𝐺𝐺𝐺 = 0.990, 𝑇𝑇 𝑇𝑇𝐺𝐺 = 0.989, all of which are greater than 0.9, indicating that the model overall fit was good, and the scale had excellent structural validity. The factor loadings of each question item were all greater than 0.6, and the combined reliability CR was greater than 0.8, indicating that the aggregation validity of the scale basically met the standard. The square root of AVE for each variable was greater than the correlation coefficient of that variable with the rest of the variables, indicating that the scale used in this study had good discriminant validity. Table 4 Reliability and validity analysis CG RG DSC DRIC DIC DBC MSCR CG 0.579 RG 0.371*** 0.644 DSC 0.379*** 0.352*** 0.673 DRIC 0.292*** 0.394*** 0.371*** 0.69 DIC 0.321*** 0.368*** 0.382*** 0.34*** 0.555 DBC 0.295*** 0.353*** 0.297*** 0.346*** 0.336*** 0.676 MSCR 0.462*** 0.566*** 0.485*** 0.508*** 0.513*** 0.533*** 0.641 Cronbach's Alpha 0.845 0.844 0.891 0.867 0.861 0.926 0.914 CR 0.846 0.845 0.892 0.869 0.862 0.926 0.914 The square root of AVE 0.761 0.802 0.820 0.831 0.745 0.822 0.801 4.2 Common method bias test The data samples used in this study are mainly micro-level data obtained through research. However, a single data source has the potential to cause common method bias and thus affect the research results. In view of this, we apply one-way validation factor analysis to test the data for common method bias using MSCR as a latent factor. As can be seen in Table 5, compared to the original fitted model, the model after the one-way validated factor analysis was poorly fitted and did not meet the reference standard; therefore, this study does not suffer from a serious common method bias problem. Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 51 Table 5 Common method bias test Indicator One-way validated factor analysis model Original fitted model Reference standard 𝑥𝑥 2 / 𝑑𝑑𝑑𝑑 7.285 1.128 <3 GFI 0.510 0.914 >0.9 RMSEA 0.145 0.038 <0.08 CFI 0.504 0.990 >0.9 NFI 0.469 0.922 >0.9 IFI 0.506 0.990 >0.9 RMR 0.130 0.038 <0.05 4.3 Hypothesis testing We used Amos 26.0 to conduct the structural equation model test. Consequently, we got the path analysis diagram shown in Fig. 3, and the specific results are shown in Table 6. As can be seen from Table 6, CG( 𝛽𝛽 = 0.134, 𝑝𝑝 = 0.015), RG( 𝛽𝛽 = 0.247, 𝑝𝑝 = 0.000), DSC( 𝛽𝛽 = 0.165, 𝑝𝑝 = 0.002), DRIC( 𝛽𝛽 = 0.143, 𝑝𝑝 = 0.010), DIC( 𝛽𝛽 = 0.185, 𝑝𝑝 = 0.001), DBC( 𝛽𝛽 = 0.242, 𝑝𝑝 = 0.000) all have a significant positive impact on MSCR. Thus, hypotheses H1a∼H2d are all supported. CG CG RG RG DSC DSC DRIC DRIC DIC DIC DBC DBC MSCR MSCR MSCR1 MSCR1 MSCR2 MSCR2 MSCR3 MSCR3 MSCR4 MSCR4 MSCR5 MSCR5 MSCR6 MSCR6 CG1 CG1 CG2 CG2 CG3 CG3 CG4 CG4 RG1 RG1 RG2 RG2 RG3 RG3 DSC1 DSC1 DSC2 DSC2 DSC3 DSC3 DSC4 DSC4 DRIC1 DRIC1 DRIC2 DRIC2 DRIC3 DRIC3 DIC1 DIC1 DIC2 DIC2 DIC3 DIC3 DIC4 DIC4 DBC1 DBC1 DBC2 DBC2 DBC3 DBC3 DBC4 DBC4 DBC5 DBC5 DBC6 DBC6 0.134** 0.247*** 0.242*** 0.185*** 0.143*** 0.170*** 0.802 0.777 0.787 0.670 0.815 0.768 0.824 0.788 0.858 0.831 0.802 0.754 0.854 0.878 0.764 0.764 0.763 0.741 0.798 0.795 0.885 0.849 0.775 0.827 0.749 0.809 0.866 0.857 0.814 0.693 Fig. 3 SEM path diagram Table 6 Hypothesis testing results Hypothetical path Hypothesis Unstandardized coefficient Standardized coefficient S.E. C.R. P CG → MSCR H1a 0.186 0.134 0.077 2.423 ** RG → MSCR H1b 0.248 0.247 0.059 4.179 *** DSC → MSCR H2a 0.165 0.170 0.054 3.089 *** DRIC → MSCR H2b 0.129 0.143 0.050 2.600 *** DIC → MSCR H2c 0.201 0.185 0.061 3.302 *** DBC → MSCR H2d 0.253 0.242 0.056 4.548 *** Notes ：** indicates P-value < 0.05；*** indicates P-value < 0.01. 5. MSCR mechanism analysis based on hybrid NCA and fsQCA methods 5.1 Necessary condition analysis The NCA method adopted in this article can not only identify whether a specific condition is a necessary condition for a certain result but also quantify the effect size of that necessary. The effect size, also termed the bottleneck level, represents the lowest level of a necessary condition required to achieve a particular result, ranging between 0 and 1. Generally, two methods, ceiling Liang, Jili, Lv, Xu 52 Advances in Production Engineering & Management 20(1) 2025 regression (CR) and ceiling envelopment (CE), can be used for estimation. A condition is deemed necessary if its effect size (d) is ≥ 0.1 and the permutation-based Monte Carlo simulation test yields a significant result. Table 7 presents the NCA results. Overall, none of the antecedent con- ditions within dynamic digital capabilities meet both criteria. In terms of SCG, in the total sam- ple, only RG has an effect size greater than 0.1 and a significant result, which is a necessary con- dition for MSCR with a medium-level effect. For this reason, we conducted a further sub-sample analysis and obtained 𝑑𝑑 𝐶𝐶𝐶𝐶 = 0.093, 𝑑𝑑 𝐶𝐶𝐶𝐶 = 0.04 in the high-tech manufacturing sample, and 𝑑𝑑 𝐶𝐶𝐶𝐶 = 0.038, 𝑑𝑑 𝐶𝐶𝐶𝐶 =0.046 in the non-high-tech manufacturing sample. The effect sizes are all less than 0.1, so that CR is not a necessary condition for high MSCR. Table 8 presents the bottleneck level analysis for antecedent conditions. The results show that there is a bottleneck of dynamic digital capabilities for MSCR level, and to achieve 70 % level (member membership score > 0.7) of MSCR requires 35.1 % level of DBC, 23.5 % level of DIC, and 9.3 % level of RG. No bottleneck effects were found for DRIC, DSC, and CG at this level. However, to reach a high MSCR level of 90 % (membership score > 0.9), higher thresholds are needed: 9.3 % for DRIC, 39.9 % for DBC, 50 % for DSC, 41.2 % for DIC, 9.3 % for RG, and 66.7 % for CG. In addition, we conducted an analysis of the necessity of individual conditions based on the fsQCA method and further examined the necessary conditions for high MSCR and non-high MSCR in the high-tech and non-high-tech manufacturing industries, and the results are shown in Table 9. The consistencies of the antecedent conditions in the four dimensions of digital dynamic capabilities and the two dimensions of supply chain governance are all less than 0.9, indicating that none of the above eight antecedent conditions are necessary conditions for high MSCR and non-high MSCR. Table 7 Results of NCA analysis Antecedent Estimation method Ceiling zone Precision (%) Effect size P-value CG CR CE 1.284 1.708 96.2 100 0.098 0.130 0.057 0.021 RG CR CE 1.295 1.388 99.0 100 0.101 0.108 0.042 0.049 DSC CR CE 1.162 1.535 99.0 100 0.095 0. 125 0.002 0.002 DRIC CR CE 1.006 1.272 99.0 100 0.078 0.099 0.110 0.095 DIC CR CE 1.580 1.932 98.1 100 0.133 0.162 0.060 0.004 DBC CR CE 1.539 2.083 98.1 100 0.132 0.179 0.095 0.000 Notes: (1) The data are the calibrated fuzzy-set membership values. (2) Range of effect size (d): 0 < d < 0.1 is re- garded as “small effect”, and 0.1 ≤ d < 0.3 is regarded as “medium effect”. (3) Permutation test (test. rep = 10,000), when p is within the range of less than 0.05, it is significant. Table 8 Bottleneck level analysis results (%) MSCR CG RG DSC DRIC DIC DBC 0 NN NN NN NN NN NN 10 NN NN NN NN NN NN 20 NN NN NN NN NN NN 30 NN NN NN NN NN NN 40 NN NN NN NN NN NN 50 NN NN NN NN NN NN 60 NN NN NN NN NN NN 70 NN 9.3 NN NN 23.5 35.1 80 NN 9.3 14.3 9.3 35.3 35.1 90 66.7 9.3 50.0 9.3 41.2 39.9 100 93.3 91.0 92.9 91.0 82.4 95.2 Notes: The analytical method is CR, and NN means "not necessary". Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 53 Table 9 Necessary condition test for QCA methodology in high-tech and non-high-tech industries Antecedent condition Outcome Variable High MSCR Non-high MSCR Supply Chain Governance CG 0.816 (0.841) 0.599 (0.595) ~CG 0.417 (0.389) 0.688 (0.672) RG 0.714 (0.729) 0.512 (0.553) ~RG 0.528 (0.562) 0.785 (0.784) Dynamic digital Capability DSC 0.778 (0.814) 0.521 (0.583) ~DSC 0.466 (0.470) 0.780 (0.746) DRIC 0.802 (0.711) 0.631 (0.563) ~DRIC 0.472 (0.623) 0.706 (0.824) DIC 0.751 (0.800) 0.478 (0.571) ~DIC 0.449 (0.450) 0.767 (0.719) DBC 0.684 (0.866) 0.535 (0.623) ~DBC 0.625 (0.463) 0.846 (0.759) Note: The values in parentheses are the analysis results for non-high-tech manufacturing industries. 5.2 Configuration analysis Using fsQCA 3.0 software, we analyzed and extracted distinct configurational pathways leading to high MSCR, illustrating the principle of equifinality, where different paths lead to the same destination”. A total of 104 valid samples were collected from the high-tech industry. According- ly, the case frequency threshold was set to 3, the original consistency threshold to 0.8, and the PRI consistency standard to above 0.6. For the non-high-tech industry, the case frequency threshold was set to 4, the original consistency threshold to 0.8, and the PRI consistency stand- ard to above. Core conditions in each grouping were identified by comparing the nested rela- tionship between the intermediate and simple solution via counterfactual analysis: if a grouping appears in both the intermediate solution and the simple solution, it is a core condition, and if it appears only in the intermediate solution, it is an auxiliary condition. (1) Configurational analysis for high-tech manufacturing industries Table 10 presents the results following the standard QCA configuration format. We identified two distinct configurations (M1a, M1b, and M2) that consistently generate high MSCR. Notably, each configuration demonstrates a consistency scores exceeding 0.9, signifying their status as sufficient conditions for achieving high MSCR. The overall solution coverage is 0.615, surpassing the 0.5 threshold, which highlights the strong explanatory power of these configurations. To succinctly capture the core attributes and highlight the uniqueness of each configuration, we performed a qualitative analysis of representative cases and took the intensity of SCG and the elementary or advanced dynamic digital capability as the "anchors" for naming the configurations. Table 10 Sufficiency analysis of condition configuration - High-tech industries Antecedent condition High MSCR M1a M1b M2 CG ● ● ● RG ● ● ● DSC ● ●  DRIC ● ● ● DIC ●  DBC  ● Consistency 0.937 0.975 0.907 Raw coverage 0.402 0.468 0.237 Unique coverage 0.021 0.105 0.028 Solution consistency 0.953 Solution coverage 0.615 Notes:  and ● respectively indicate that the level of antecedent conditions is not high and relatively high; The large circle represents the core condition, and the small circle represents the auxiliary condition; a blank space indicates that the condition is not important for the generation of results. Liang, Jili, Lv, Xu 54 Advances in Production Engineering & Management 20(1) 2025 a) Configuration M1 (Elementary digital application—Strong governance synergy type). Configu- ration M1 demonstrates that core conditions—CG, RG, DSC and DRIC—jointly drive high MSCR with DBC or DIC serving as auxiliary conditions. In this configuration, the attributes of the high- tech industry highlight the critical role of strong governance. It suggests that focal firms in the manufacturing supply chain should foster both CG and RG to leverage complementary govern- ance advantages. Essentially, digital sensing is the initial detection of digital value, while integra- tion is the optimal combination of value carriers (digital resources). Together, these two capabil- ities establish a foundational level for extracting digital value, paving the way for more advanced capabilities, such as DIC and DBC. Configuration M1 comprises two distinct paths: − Path M1a: The antecedent construct is represented as "CG*RG*DSC*DRIC*~DBC". − Path M1b: The antecedent construct is represented as "CG*RG*DSC*DRIC*DIC". Configuration M1a exhibits a consistency score of 0.937, with an original coverage of 0.402 and a unique coverage of 0.021. This configuration accounts for approximately 40.2 % of the cases, primarily in highly regulated sectors such as shipbuilding, aviation, aerospace, and equipment manufacturing. In these sectors, a robust governance structure empowers supply chain partners to navigate complex regulatory and market environments, ensuring compliance and transparency throughout the digital transformation process. In this context, DSC plays a vital role. It enables enterprises to swiftly capture shifts in market demand while providing pro- active data support for addressing potential risk events. Additionally, DRIC serves as a collabora- tive tool for enhancing SCR. By facilitating digital resource integration, it fosters synergistic ef- fects that improve the overall elasticity and stability of the manufacturing supply chain. Configuration M1b features a consistency of 0.975, an original coverage of 0.468, and a unique coverage of 0.105, explaining the largest number of cases at 46.8 %. These cases are mainly con- centrated in industries such as electronic and communication equipment manufacturing, com- puter and office equipment manufacturing, pharmaceutical manufacturing, and medical instru- ment and device manufacturing. A notable difference between M1a and M1b is that, M1b incor- porates DIC as a auxiliary condition, while M1a includes DBC. This difference can be attributed to significant variations in market environment, product characteristics, and innovation de- mands between these two industry types. The industries covered by configuration M1b are characterized by rapid technological iteration, intense market competition, and a strong orienta- tion toward mass-market consumer products. In this setting, firms must proactively drive inno- vation and application of digital technologies to sustain their competitive edge. In contrast, sec- tors such as shipbuilding and aerospace typically produce highly customized products with sta- ble user demands and longer R&D cycles. Consequently, these industries place less direct reli- ance on DIC and prioritize developing foundational digital capabilities to ensure product reliabil- ity and compliance. b) Configuration M2 (Dual-core digital-driven—Strong governance synergy type). The antecedent construct is "CG*RG*DRIC*DBC*~DSC*~DIC". This configuration indicates that CG, RG, DRIC and DBC are core conditions, while DSC and DIC are absent auxiliary conditions in achieving high MSCR. This configuration yields a consistency of 0.907, with raw coverage of 0.237 and unique coverage of 0.028, explaining approximately 23.7 % of the observed cases. These cases are pre- dominantly situated in the information chemical manufacturing industry, which features long, complex supply chains and high sensitivity to fossil fuel price fluctuations. This configuration highlights the importance of digital optimization and execution across production and opera- tional processes. Specifically, the combined strength of DRIC and DBC facilitates agile resource allocation and rapid response under external shocks, such as war, geopolitical conflict, or abrupt price volatility. These digital capabilities enable supply chain nodes to maintain visibility, redis- tribute constrained resources, and reconfigure operations in real time. When such capabilities are embedded in a governance structure with CG and RG, firms are better positioned to absorb shocks, contain their propagation, and swiftly restore operational continuity. Therefore, the syn- Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 55 ergy between strong governance redundancy and dual-core digital capability serves as a critical resilience mechanism against severe external disruptions. (2) Configurational analysis for non-high-tech manufacturing industries As shown in Table 11, there are three types of configurations (L1a, L1b, L2, and L3) that contrib- ute to high MSCR. All of their consistencies exceed 0.9, indicating that these configurations are sufficient conditions for achieving high MSCR. Additionally, the solution coverage is 0.697, which is significantly above the threshold of 0.5, demonstrating strong explanatory power. Combined with theory and industry cases analysis, we took DRIC, RG and the elementary or advanced dy- namic digital capability as the "anchors" for configuration naming that takes into account both integrity and uniqueness. Table 11 Sufficiency analysis of condition configuration – non-high-tech industries Antecedent condition High MSCR L1a L1b L2 L3 CG ● ● ● RG ● ● ● DSC ●  DRIC ● ● ● DIC ● ●  DBC ● ● ● Consistency 0.959 0.950 0.937 0.910 Raw coverage 0.534 0.511 0.520 0.180 Unique coverage 0.070 0.056 0.056 0.018 Solution consistency 0.917 Solution coverage 0.697 a) Configuration L1 (Digital resource integration dominant type). This configuration highlights the core role of digital resource integration capability in enhancing MSCR, as it effectively breaks down information silos, facilitates collaboration among stakeholders, and optimizes resources. This, in turn, enhances the decision-making flexibility and market responsiveness of the supply chain, thereby improving the resilience of member enterprises in dynamic market environments. Configuration L1 comprises two distinct paths: − Path L1a: The antecedent construct is represented as "CG*DRIC*DBC*DIC". − Path L1b: The antecedent construct is represented as "CG*RG*DSC*DRIC". Configuration L1a identifies DRIC as the core condition, with DBC, DIC, and CG serving as aux- iliary conditions contributing to high MSCR. The consistency of this configuration is 0.959, with a raw coverage of 0.534 and unique coverage of 0.07. This configuration accounts for approxi- mately 53.4 % of the cases, primarily within the agricultural and food processing sectors. These industries operate in a highly demand-driven market, where seasonal variations and fluctua- tions in consumer preferences directly influence production and inventory decisions. Moreover, growing societal concerns over food safety highlight the pivotal role of DRIC in enabling real- time monitoring and traceability across the supply chain. DBC and DIC function as auxiliary con- ditions that support enterprises in optimizing processes and driving innovation. However, in such a responsive market, their effectiveness relies on the foundational support provided by DRIC. Additionally, CG ensures collaboration and compliance among supply chain partners, fur- ther enhancing MSCR. Configuration L1b also identifies DRIC as the core condition, but includes DSC, RG and CG serving as auxiliary conditions. This configuration has a consistency of 0.95, with a raw coverage of 0.511 and unique coverage of 0.056. It accounts for approximately 51.1 % of the cases, pri- marily within the chemical fiber, rubber and plastic manufacturing, non-ferrous metal smelting, and metal manufacturing sectors. These industries share common characteristics, including complex production processes, high dependence on raw materials, and frequent fluctuations in market demand. Therefore, in practical operational management, on one hand, the focus is on leveraging DRIC and DSC to optimize process flows and enhance market responsiveness, aiming to achieve cost reduction, efficiency improvement, and risk mitigation. On the other hand, the Liang, Jili, Lv, Xu 56 Advances in Production Engineering & Management 20(1) 2025 collaborative advancement of CG and RG provides a more comprehensive management frame- work, enhancing the cooperation efficiency and adaptability of supply chain members. b) Configuration L2 (Advanced digital-driven—Relationship-oriented governance synergy type). The antecedent construct is represented as "CG*RG*DIC*DBC", where CG serves as an auxiliary condition, while the others are considered core conditions. The consistency of this configuration is 0.937, with a raw coverage of 0.52 and unique coverage of 0.056. This configuration accounts for approximately 52 % of the cases, primarily within general and specialized equipment, auto- motive manufacturing, and electrical machinery and equipment manufacturing sectors. In such an industrial environment with complex products and a highly dependent supply chain, DIC and DBC can effectively work in coordination, drive business process reengineering, and facilitate innovation and optimization in product design, production, and services. This is of vital im- portance for maintaining a competitive advantage and achieving a high MSCR. Additionally, compared with CG, RG can accelerate knowledge flow and promote collaborative innovation, while CG is more of a support for this relationship. This distinction is particularly significant for understanding the condition configuration in which they jointly achieve high resilience with DIC. c) Configuration L3 (Dual-core digital-driven—Relationship-prioritized governance synergy type), has its antecedent construct represented as "CG*~DSC*DRIC*~DIC*DBC". The consistency of this configuration is 0.91, with a raw coverage of 0.18 and unique coverage of 0.018. Approxi- mately 18 % of the cases can be explained by this configuration, primarily those in the paper- making, paper products and printing industries, as well as in the manufacturing sectors of cul- tural, educational, sports and entertainment products. In the digital era, these industries regard the assetization of digital resources as both a foundation and a strategic direction for develop- ment. At the same time, real-world cases of digital transformation further underscore the neces- sity of leveraging digital technologies to optimize operations, enhance customer experience, and improve marketing strategies. Thus, it is evident that DRIC and DBC emerge as the dual-core digital drivers for achieving high MSCR. This configuration also suggests that an RG model should be prioritized to enhance the adaptability and flexibility of manufacturing supply chain by building long-term trust and reciprocity among partners, rather than a CG model that focuses only on short-term compliance. 5.3 Test of robustness QCA is a set-theoretic approach that is considered robust when slight adjustments to the opera- tion, with subset relationships between the results produced, do not change the substantive ex- planation of the research findings. We evaluated the robustness of the antecedent configuration that achieves high MSCR by increasing the case frequency threshold and consistency. First, the case frequency thresholds for high-tech manufacturing and non-high-tech manufacturing were adjusted upward by 1, resulting in new configurations that are fundamentally subsets of the original configurations, with no significant changes in core conditions. Second, by increasing the consistency from 0.80 to 0.90, the resulting configurations remained consistent with the original configurations, with no changes in consistency or coverage. The robustness test indicates that the results are robust. 6. Discussion 6.1 Interpretation of findings This study empirically confirms that supply chain governance and dynamic digital capabilities serve as dual drivers of high MSCR. SEM results show both positively contribute to resilience outcomes, while NCA indicates that no single factor—whether contractual governance, relation- al trust, or digital innovation— is indispensable, highlighting the causal complexity involved. FsQCA further identifies five distinct high-resilience configurations across high-tech and non- high-tech sectors. These results reveal that resilience does not stem from isolated excellence, but emerges from context-specific combinations of governance and capability elements. High-tech Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 57 firms tend to benefit from strong governance paired with either foundational or advanced digital enablement, whereas non-high-tech firms rely more on digital resource integration and rela- tional governance. Overall, the findings underscore that high MSCR is shaped by configuration fit and strategic alignment, not from uniform solutions. Firms must tailor strategies to their techno- logical and organizational contexts, embracing configuration logic in place of one-size-fits-all models. 6.2 Theoretical implications Several theoretical implications are noteworthy. First, in contrast to prior empirical summaries and conceptual models, our study verifies the positive influence mechanism of SCG and dynamic digital capability on enhancing MSCR through SEM. It highlights the critical importance of an organization’s dynamic digital capabilities in addressing supply chain risks [2], and also empha- sizes the strategic value of effective SCG in securing competitive advantage in volatile environ- ments. These findings enrich our understanding of the logical linkages among the multiple di- mensions of SCG, dynamic digital capability, and MSCR. Second, prior research on SCR has predominantly emphasized the net effects of individual fac- tors, often overlooking how multiple resources and capabilities may interact in a configurational manner to drive resilience. Beyond the conventional SCG initiatives, developing dynamic digital capabilities has emerged as a critical option for building SCR in uncertain and turbulent envi- ronments. However, existing studies have seldom investigated the synergistic configuration of SCG and digital capabilities within a complex systems framework. By addressing this gap, the present study offers new theoretical insights that enrich and refine the conceptual foundations of SCR, particularly under conditions shaped by digital transformation. Finally, this research presents the "causal complexity" of constructing MSCR by identifying multiple, equally effective configurations that lead to high resilience. This finding aligns with Fiss’s configuration theory, which posits that similar outcomes can emerge from divergent caus- al paths [34]. It underscores the industry-specific and context-dependent nature of resilience- building strategies. Firms should adopt a configuration mindset to identify and leverage their core capability combinations to address specific market environments and challenges. 6.3 Managerial implications This research presents three key managerial implications. First, adapting governance models to industry characteristics is essential. Enterprises are advised to select appropriate governance models based on their technological attributes. The industry heterogeneity analysis reveals no universally applicable conditional configuration, indicating that high-tech and non-high-tech manufacturing sectors follow distinct paths toward achieving high MSCR. Specifically, high-tech industries ought to accentuate the synergy between CG and RG to ensure that all stakeholders remain coordinated amidst technological shifts and market fluctuations. Conversely, non-high- tech manufacturing industries should leverage governance methods suited to their resource profiles and capability structures. Actively fostering trust, commitment, and reciprocity through informal governance can further facilitate the efficient flow of knowledge and resources, thereby enhancing the stability and synergy of the supply chain. Second, firms should prioritize the development of digital capabilities based on strategic needs. In high-tech sectors, investment should focus on DSC and digital DRIC. DSC enables agile detection of technological trends and shifting market demands, while DRIC enhances the coordi- nation of internal and external data to support supply chain synergy. For non-high-tech manu- facturing industry, the strategic priority lies in customer value creation. Here, the emphasis should be placed on developing DBC and DRC. By introducing IoT, big data analysis, and other digital technologies to optimize product design, production processes, and service models, firms can foster business model innovation and improve the responsiveness and flexibility of the sup- ply chain. Third, managers should adopt a configuration-oriented mindset in decision-making. Rather than relying on single-factor approaches, firms should consider how different combinations of SCG and digital capabilities contribute to MSCR. This study reveals that, equivalent configurations may fea- Liang, Jili, Lv, Xu 58 Advances in Production Engineering & Management 20(1) 2025 ture either different core conditions or identical core conditions with varying auxiliary conditions. The existence of such multiple-condition configuration reflects the complexity of MSCR manage- ment. Accordingly, firms should avoid blindly replicating successful strategies from other contexts. Instead, they should assess their own technological orientation, resource base, and market condi- tions to develop adaptive, context-specific capability-governance portfolios. 6.4 Limitations and future research While this study provides valuable theoretical and practical insights, several limitations should be acknowledged. First, the cross-sectional design limits causal inference, limiting the ability to capture temporal dynamics in resilience development. Future research could adopt longitudinal designs or temporal QCA to explore how configurations evolve over time. Second, resource en- dowment, particularly the constraints faced by small enterprises, represents a critical contextual factor influencing the feasibility of resilience configurations. Due to the limited representation of small firms in the current sample, conducting a dedicated fsQCA for this subgroup was not feasi- ble. Future research should consider expanding the sample size of small firms and incorporate firm size as a moderating variable to further clarify how resource constraints shape configura- tion selection and performance. Third, cultural and institutional contexts may influence the ef- fectiveness of relational governance mechanisms such as trust and reciprocity, suggesting the need for cross-cultural comparative studies. Finally, the study focuses on governance and digital capabilities, leaving other potential factors—such as policy support or organizational learning— for future exploration. Addressing these areas may enhance the robustness and generalizability of configuration-based resilience research. 7. Conclusion This study rigorously examines the dual influence of SCG and dynamic digital capabilities on MSCR by integrating SEM, NCA, and fsQCA. Empirical results confirm that both governance mechanisms and digital capabilities significantly enhance MSCR, however, it also reveals inher- ent causal complexity: no single factor can ensure resilience alone. Five distinct, context-specific configurations across high-tech and non-high-tech sectors were identified, indicating that resili- ence emerges from tailored combinations of governance and capability elements rather than uniform solutions. These findings deepen theoretical understanding by framing MSCR as a con- figurational outcome shaped by the interplay of multiple factors under varying contextual condi- tions. Managerially, they highlight the imperative for firms to align governance models and digi- tal capability development with their specific industry characteristics and strategic priorities, enabling the creation of context-sensitive strategies to withstand supply chain disruptions. Ulti- mately, this research moves the resilience discourse forward by demonstrating that strategic alignment and configuration fit—not generic prescriptions—are fundamental to sustaining competitive advantage amid escalating supply chain uncertainties. Acknowledgement This work is partly supported by Philosophy and Social Science Planning Project of Anhui Province (AHSKQ2020D22), and Scientific Research Foundation of Education Department of Anhui Province (2023AH040066, 2024AH052759). References [1] Teece, D.J., Pisano, G., Shuen, A. (1997). Dynamic capabilities and strategic management, Strategic Management Journal, Vol. 18, No. 7, 509-533, doi: 10.1002/(SICI)1097-0266(199708)18:7<509::AID-SMJ882>3.0.CO;2-Z. [2] Dubey, R., Bryde, D.J., Dwivedi, Y.K., Graham, G., Foropon, C., Papadopoulos, T. (2023). Dynamic digital capabili- ties and supply chain resilience: The role of government effectiveness, International Journal of Production Eco- nomics, Vol. 258, Article No. 108790, doi: 10.1016/j.ijpe.2023.108790. [3] Ponomarov, S.Y., Holcomb, M.C. (2009). Understanding the concept of supply chain resilience, The International Journal of Logistics Management, Vol. 20, No. 1, 124-143, doi: 10.1108/09574090910954873. Configuring supply chain governance and digital capabilities for resilience: Evidence from the manufacturing sector Advances in Production Engineering & Management 20(1) 2025 59 [4] Ivanov, D., Dolgui, A. (2020). Viability of intertwined supply networks: Extending the supply chain resilience angles towards survivability. A position paper motivated by COVID-19 outbreak, International Journal of Produc- tion Research, Vol. 58, No. 10, 2904-2915, doi: 10.1080/00207543.2020.1750727. [5] Behzadi, G., O'Sullivan, M.J., Olsen, T.L. (2020). On metrics for supply chain resilience, European Journal of Opera- tional Research, Vol. 287, No. 1, 145-158, doi: 10.1016/j.ejor.2020.04.040. [6] Liang, P.P., Li, C.W. (2019). Impact of cooperation uncertainty on the robustness of manufacturing service system, Advances in Production Engineering & Management, Vol. 14, No. 2, 189-200, doi: 10.14743/apem2019.2.321. [7] Maududy, R., Nurdin, A.M. (2024). An architecture framework for supply chain management systems integrated with supervisory control and data acquisition functionality, Journal of Logistics, Informatics and Service Science, Vol. 11, No. 5, 38-51, doi: 10.33168/JLISS.2024.0503. [8] Kähkönen, A.K., Evangelista, P., Hallikas, J., Immonen, M., Lintukangas, K. (2023). COVID-19 as a trigger for dy- namic capability development and supply chain resilience improvement, International Journal of Production Re- search, Vol. 61, No. 8, 2696-2715, doi: 10.1080/00207543.2021.2009588. [9] Um, K.-H. (2024). Strategic governance dynamics in manufacturing firms: Navigating operational performance through contractual and relational mechanisms in the face of product complexity, Journal of Manufacturing Technology Management, Vol. 35, No. 3, 502-523, doi: 10.1108/JMTM-09-2023-0411. [10] Samant, S., Kim, J. (2023). Best foot forward? The importance of contractual governance mechanisms for innova- tion from alliances, Technovation, Vol. 127, Article No. 102828, doi: 10.1016/j.technovation.2023.102828. [11] Zhang, Q., Jin, J.L., Yang, D.F. (2020). How to enhance supplier performance in China: Interplay of contracts, rela- tional governance and legal development, International Journal of Operations & Production Management, Vol. 40, No. 6, 777-808, doi: 10.1108/IJOPM-02-2020-0093. [12] Keller, J., Burkhardt, P., Lasch, R. (2021). Informal governance in the digital transformation, International Journal of Operations and Production Management, Vol. 41, No. 7, 1060-1084, doi: 10.1108/IJOPM-09-2020-0660. [13] Giannoccaro, I., Iftikhar, A. (2022). Mitigating ripple effect in supply networks: The effect of trust and topology on resilience, International Journal of Production Research, Vol. 60, No. 4, 1178-1195, doi: 10.1080/00207543. 2020.1853844. [14] Yang, Q., Geng, R., Jiang, Y., Feng, T. (2021). Governance mechanisms and green customer integration in China: The joint effect of power and environmental uncertainty, Transportation Research Part E: Logistics and Transpor- tation Review, Vol. 149, Article No. 102307, doi: 10.1016/j.tre.2021.102307. [15] Yang, J., Liu, Y., Kholaif, M.M.N.H.K. (2024). The impact of relationship management on manufacturer resilience in emergencies, Kybernetes, Vol. 53, No. 3, 960-989, doi: 10.1108/K-08-2022-1198. [16] Valiušis, O. (2025). Digitalization modeling of production processes in paper packaging sector companies, Jour- nal of Management Changes in the Digital Era, Vol. 2, No. 1, 67-76, doi: 10.33168/JMCDE.2025.0105. [17] Teece, D., Peteraf, M., Leih, S. (2016). Dynamic capabilities and organizational agility: Risk, uncertainty, and strategy in the innovation economy, California Management Review, Vol. 58, No. 4, 13-35, doi: 10.1525/ cmr.2016.58.4.13. [18] Sousa-Zomer, T.T., Neely, A., Martinez, V. (2020). Digital transforming capability and performance: A microfoun- dational perspective, International Journal of Operations and Production Management, Vol. 40, No. 7-8, 1095- 1128, doi: 10.1108/IJOPM-06-2019-0444. [19] Christofi, M., Khan, H., Zahoor, N., Hadjielias, E., Tarba, S. (2024). Digital transformation of SMEs: The role of entrepreneurial persistence and market sensing dynamic capability, IEEE Transactions on Engineering Manage- ment, Vol. 71, 13598-13615, doi: 10.1109/TEM.2022.3230248. [20] Sirmon, D.G., Hitt, M.A. (2003). Managing resources: Linking unique resources, management, and wealth creation in family firms, Entrepreneurship Theory and Practice, Vol. 27, No. 4, 339-358, doi: 10.1111/1540-8520.t01-1- 00013. [21] Ghosh, S., Hughes, M., Hodgkinson, I., Hughes, P. (2022). Digital transformation of industrial businesses: A dy- namic capability approach, Technovation, Vol. 113, Article No. 102414, doi: 10.1016/j.technovation.2021. 102414. [22] Zhang, Z.G., Ye, B., Hu, A.T., Chen, L. (2024). How do manufacturing enterprises integrate data resources to ena- ble product innovation performance? The role of calculative and relational inter-organizational trust, Studies in Science of Science, Vol. 42, No. 4, 649-659, doi: 10.16192/j.cnki.1003-2053.20230922.003. [23] Liu, Y., Dong, J.Y., Mei, L., Shen, R. (2023). Digital innovation and performance of manufacturing firms: An af- fordance perspective, Technovation, Vol. 119, Article No. 102458, doi: 10.1016/j.technovation.2022.102458. [24] Eller, R., Alford, P., Kallmünzer, A., Peters, M. (2020). Antecedents, consequences, and challenges of small and medium-sized enterprise digitalization, Journal of Business Research, Vol. 112, 119-127, doi: 10.1016/j.jbusres. 2020.03.004. [25] Lin, J., Lin, S., Benitez, J., Luo, X.R., Ajamieh, A. (2023). How to build supply chain resilience: The role of fit mecha- nisms between digitally-driven business capability and supply chain governance, Information & Management, Vol. 60, No. 2, Article No. 103747, doi: 10.1016/j.im.2022.103747. [26] Bag, S., Dhamija, P., Luthra, S., Huisingh, D. (2023). How big data analytics can help manufacturing companies strengthen supply chain resilience in the context of the COVID-19 pandemic, The International Journal of Logis- tics Management, Vol. 34, No. 4, 1141-1164, doi: 10.1108/IJLM-02-2021-0095. [27] Xu, Q., Tan, Z., Zhang, Y. (2024). An evolutionary game analysis of digital decision making in manufacturing en- terprises under reward and punishment mechanism, Economic Computation and Economic Cybernetics Studies and Research, Vol. 58, No. 1, 52-69, doi: 10.24818/18423264/58.1.24.04. Liang, Jili, Lv, Xu 60 Advances in Production Engineering & Management 20(1) 2025 [28] Faruquee, M., Paulraj, A., Irawan, C.A. (2024). The dual effect of environmental dynamism on proactive resilience: Can governance mechanisms negate the dark side?, Production Planning & Control, Vol. 35, No. 15, 2113-2130, doi: 10.1080/09537287.2023.2291378. [29] Li, Y., Wang, X., Gong, T., Wang, H. (2023). Breaking out of the pandemic: How can firms match internal compe- tence with external resources to shape operational resilience?, Journal of Operations Management, Vol. 69, No. 3, 384-403, doi: 10.1002/joom.1176. [30] Zhou, X., Zhu, Q., Xu, Z. (2023). The role of contractual and relational governance for the success of digital tracea- bility: Evidence from Chinese food producers, International Journal of Production Economics, Vol. 255, Article No. 108659, doi: 10.1016/j.ijpe.2022.108659. [31] Poppo, L., Zenger, T. (2002). Do formal contracts and relational governance function as substitutes or comple- ments?, Strategic Management Journal, Vol. 23, No. 8, 707-725, doi: 10.1002/smj.249. [32] Li, M., Li, Z., Huang, X.D., Qu, T. (2021). Blockchain-based digital twin sharing platform for reconfigurable social- ized manufacturing resource integration, International Journal of Production Economics, Vol. 240, Article No. 108223, doi: 10.1016/j.ijpe.2021.108223. [33] Liao, M.C., Jiang, Y.S., Jing, J.M. (2023). Digital innovation capability under innovation ecosystem: Re- conceptualization and scale development, Soft Science, Vol. 37, No. 5, 62-70, doi: 10.13956/j.ss.1001-8409.2023. 05.09. [34] Aslam, H., Khan, A.Q., Rashid, K., Rehman, S.-U. (2020). Achieving supply chain resilience: The role of supply chain ambidexterity and supply chain agility, Journal of Manufacturing Technology Management, Vol. 31, No. 6, 1185-1204, doi: 10.1108/JMTM-07-2019-0263. [35] Fiss, P.C. (2011). Building better causal theories: A fuzzy set approach to typologies in organization research, Academy of Management Journal, Vol. 54, No. 2, 393-420, doi: 10.5465/amj.2011.60263120.