Microelement Integration Drives Smart Manufacturing: A Mixed Method Study

Abstract

Smart manufacturing is an important initiative to promote the transformation and upgrading of industries and the high-quality development of the economy. However, the current situation of digitalized smart transformation in manufacturing enterprises is not optimistic, which is primarily attributed to the ambiguity surrounding the pathways. This study is based on the technology-organization-environment-individual (TOE-I) analytical framework; it selects 20 case studies of advanced manufacturing enterprises; and employs case studies and necessary condition fuzzy set qualitative comparative research methods (NCA and fsQCA) to investigate the pathways through which technology, organization, the environment, and individual microelements synergistically drive smart manufacturing from a configurational perspective. The study reveals that digital technology breakthroughs, digital infrastructure, digital talent, digital sharing, organizational resilience, organizational culture, and the entrepreneurial spirit are the core influencing factors in advancing smart manufacturing for manufacturing enterprises, and four implementation paths driven by smart manufacturing are analyzed. Among them, digital technology breakthroughs and digital infrastructure have a potential substitutive relationship in the “technology + talent” empowerment organizational model. Organizational resilience, organizational culture, and the entrepreneurial spirit are important safeguards for successful advancements in smart manufacturing. In contrast, digital infrastructure plays a more indirect, supporting role. Accordingly, this paper provides theoretical reference and practical guidance.

1. Introduction

Industry 4.0 heralds a new era for the global manufacturing industry to step into smart manufacturing [1]. Smart manufacturing is of substantial importance for facilitating industrial upgrading and enhancing competitiveness as well as for achieving future sustainable development [2]. The systematic implementation of smart manufacturing requires manufacturing enterprises to undergo comprehensive digitalization and smart transformation in the fields of production, management, and services [3,4]. However, in the current situation, the implementation for the transformation of manufacturing enterprises is not optimal [5]. According to McKinsey’s “Industrial Intelligence Index: An Industrial 4.0 Assessment Tool for Operational Excellence,” by the end of 2020, approximately 74% of manufacturing companies were stuck in transformation dilemmas; this further indicates that the lack of a clear path is the main obstacle for most manufacturing enterprises in the process of transformation. Therefore, how to drive the digitalized smart transformation of manufacturing enterprises and promote better implementation of smart manufacturing has become a key issue on which the academic community and manufacturing enterprises need to focus.

Smart manufacturing, as a product of the integration of digital technology and the manufacturing industry, facilitates the transition to smart manufacturing, which can help enterprises enhance the efficiency and quality of production and operations, optimize resource allocation, and increase the level of industrial structure sophistication [6,7]. Enterprises, as the main drivers of industrial development, play a pivotal role in the construction of smart manufacturing by addressing the needs of manufacturing enterprises while facilitating the convergence of various resources toward these enterprises and stimulating their drive and integrated development vitality [8,9]. Despite a strong inclination toward smart manufacturing, manufacturing enterprises are still facing difficulties in the actual process of transformation. Specifically, their understanding of smart manufacturing is superficial, with a tendency to equate it with the mere application of digital technologies and to uncritically adopt the digital enablement paths of other companies [10]. This approach does not integrate the internal elements of the manufacturing enterprise with the enablement path, leading to unsatisfactory transformation outcomes. Some enterprises, recognizing the importance of smart manufacturing, have failed in their attempts due to unclear driving elements and paths [11]; this not only prevents the recovery of initial investment costs but also may lead to a decrease in a company’s market competitiveness [12].

Scholars have noted that digital technologies, such as big data, artificial intelligence, and blockchain, play a pivotal role in propelling smart manufacturing within manufacturing enterprises [13]. Specifically, these technologies can assist in optimizing production processes, reshaping supply chain management methods, and achieving upgrades in high-quality, efficient, and high-value-added industrial value chain segments, thus promoting continuous innovation in smart manufacturing [14]. The utilization of digital technology infrastructures to expand the scope of external resource utilization and enable collaborative sharing of data resources provides an impetus for the transformation and upgrading of smart manufacturing in manufacturing enterprises [15]. Concurrently, it can reconfigure enterprise resource allocation, overcome the imbalance of supply and demand matching, and aid enterprises in transitioning from mass production to intelligent, customized manufacturing, achieving the integration of production and consumption [16]. The relationship between smart manufacturing and enterprise factors has also garnered widespread attention. Scholars have indicated that smart manufacturing relies on the tacit knowledge of internal leaders at various levels, including concepts, experience, and behavioural patterns, as well as the explicit knowledge of professional and technical teams and R&D personnel [17,18]. Given the long-term and uncertain nature of smart manufacturing, which does not yield short-term benefits, leaders may avoid digital and intelligent transformation to maintain short-term profitability and performance rewards. However, through proactive cognition and systematic planning by leadership, supplemented by supportive and collaborative enterprise management, effective guidance and the continuous advancement of enterprise smart manufacturing can be achieved [19,20]. Furthermore, Hu et al. noted that internal professional technical personnel are crucial for carrying out key core technology innovation activities, shaping advantages for the rapid development of smart manufacturing [7]. A resilient organizational system facilitates the transition of departments to smart upgrades, enhances effective organizational collaboration, and drives smart manufacturing [21,22].

Although scholars have confirmed the key role of digital technology in driving smart manufacturing, the focus has mainly been on theoretical discussions of the driving process of smart manufacturing [14,15,16]. Some scholars have conducted empirical studies via quantitative analysis methods, but these methods are often based on macro industry data [23]. The inherent difficulties in obtaining industry data, such as high acquisition costs and insufficient indicators, lead to a lack of in-depth research on the driving factors of smart manufacturing in manufacturing enterprises. In addition, existing research has revealed the effectiveness of factors in driving smart manufacturing at the enterprise level, but most of these studies focus on the role of single factors [20,21]. Building on this, some scholars have begun to explore the interactive effects of factors influencing smart manufacturing, but mainly in the context of constructing business models for smart manufacturing [24]. However, there is still a research gap regarding how to integrate multidimensional factors at the micro level of enterprises to jointly drive the systematic implementation of smart manufacturing.

Fuzzy-set qualitative comparative analysis (fsQCA) is suitable for revealing the impact mechanisms of multiple influencing factors working in conjunction from a configurational perspective, and it can effectively explain the causal complexity of multiple enterprise elements and smart manufacturing. The technology-organization-environment-individual (TOE-I) framework is widely applied to reveal the multifactorial conditions affecting corporate digitalized smart transformation [25] and is conducive to analyzing the key factors influencing smart manufacturing advancement in the context of manufacturing enterprises and exploring effective implementation paths to enhance smart manufacturing.

Therefore, to examine how multiple factors interact in configurations and to understand the complex mechanisms that can drive smart manufacturing, this article is based on the TOE-I analytical framework. It adopts 20 typical Chinese advanced manufacturing enterprises as its research subjects and uses case study methods to explore the core factors affecting smart manufacturing from four dimensions, namely, technology, organization, the environment, and individuals; it subsequently employs the methods of necessary condition analysis (NCA) and fsQCA to investigate the multifaceted pathways of the “technology-organization-environment-individual” dimensions that promote smart manufacturing from a configurational perspective and provides case enterprises’ smart manufacturing implementation processes as references. This not only opens the “black box” of the interactive relationship between enterprise elements and smart manufacturing from a theoretical level but also offers guidance for the digitalized smart transformation of manufacturing enterprises from a practical standpoint.

2. Theoretical Framework

The TOE framework, proposed by Tornatzky and Fleischer in 1990, is a theoretical framework that comprehensively considers the impact of technological, organizational, and environmental factors on technological innovation [26]. Initially aimed at analyzing technology adoption issues at the organizational level, subsequent studies have indicated that incorporating individual factors makes the application of the TOE framework more effective [27,28], leading to the proposal of the TOE-I framework [29]. Smart manufacturing, as a complex outcome of corporate digitalized smart transformation, involves the interaction and alignment of multidimensional factors such as digital technology, the organizational structure, the external environment, and internal personnel. For instance, Chang et al. proposed the impact of individual digital literacy on transformation [30]. Xu et al. drew on upper echelons theory and suggested that leaders’ academic cognition affects corporate decisions in smart manufacturing transformation [31]. Therefore, this article uses the TOE-I analytical framework to conduct exploratory research on the mechanisms that promote smart manufacturing in typical manufacturing enterprises. The aim of this work is to uncover potential influencing factors and identify important conditional variables.

2.1. Technological Perspective

Promoting comprehensive penetration into the industry of new production factors, such as digital technology and data resources, with innovation breakthroughs and application expansion of digital technology as the main thread [32], is conducive to advancing the digitalization of industries and the industrialization of digital technology, thereby achieving comprehensive empowerment for digitalized smart manufacturing. Breakthrough innovations in digital technologies, such as big data, artificial intelligence, the Internet of Things, and blockchain, serve as crucial pathways for the digital and smart information application management of traditional enterprises [33]. Breakthroughs and innovations in digital technology can facilitate the optimization of corporate factor structures, the transformation of kinetic structures, and the upgrading of industrial structures [34]. Furthermore, their groundbreaking applications can help enterprises continuously update and expand new products or services, offering competitive advantages and sustained value [35,36].

The digital infrastructure represents staunch support for digital technology that relies on the robust technical support capabilities of application support-type technological infrastructure to achieve breakthroughs in key core technologies and ensure autonomous control in critical areas and key links, which has become an important focal point for advancing digitalized smart manufacturing [37]. The synergy between these two elements continuously facilitates the penetration and function of various enterprise segments, such as R&D innovation, manufacturing, user services, and operational maintenance [38]. Therefore, this paper posits that breakthroughs in digital technology and digital infrastructure have a significant effect on enhancing the effectiveness of smart manufacturing.

2.2. Organizational Perspective

Smart manufacturing is inseparable from the deep integration of smartization with the internal organization of manufacturing enterprises. Organizational resilience, as the capacity to anticipate potential threats, adapt to external shocks, and modulate the impact of risks and uncertainties, can mitigate the damage caused by crises and enable swift rebound and recovery [39]. Leveraging digitalization and smartization to reshape the core elements of an organization can enhance organizational resilience, thereby facilitating corporate transformation and upgrading [40]. Moreover, organizational culture can facilitate the strengthening of forward-looking cognition and the spread of a shared vision within the enterprise while guiding the direction and process of smart manufacturing [41]. For example, Sany Group, as an advanced global equipment manufacturing enterprise, has conveyed smart manufacturing cultural cognition at all levels of the company through learning activities, expert lectures, and visits to other enterprises, promoting the implementation of smart manufacturing concepts within the organization, and has now achieved smart manufacturing. Therefore, organizational resilience and organizational culture are key factors that drive smart manufacturing.

2.3. Environmental Perspective

The successful development of smart manufacturing is the result of enterprises and multiple stakeholders working together to achieve efficient allocation of resources. A robust digital sharing environment facilitates more collaboration and allows enterprises to gain more knowledge and information, which aids in strategic planning and transformation [42]. Data sharing helps enterprises transmit information in a timely and accurate manner, breaks down information silos, optimizes cooperative supply chain coordination, and aids in optimizing production processes and improving corporate decision-making efficiency [43]. Digital sharing is inseparable from the exchange of information within digital networks constructed by the external social relationships of enterprises and the integration and openness of internal digital platforms [44,45]. In-depth communication between enterprises, government departments, and research institutions has clarified the opaque and imprecise phenomena that occurred in the past technology achievement demand docking process, enhancing the efficiency and effectiveness of sharing scientific and technological achievements. The public disclosure(s) of internal information platforms and system data have improved the effectiveness of corporate data resource applications and strengthened control over the external digital environment.

2.4. Individual Perspective

The entrepreneurial spirit is associated with the development of enterprises and encompasses attributes such as opportunity sensitivity and leadership [46]. Leaders with an entrepreneurial spirit are often able to seize external opportunities, effectively advance the implementation of corporate development concepts and plans, form an internal growth dynamic and achieve external opportunities for development [47]. In the process of smart manufacturing, digital leadership not only helps enterprises identify opportunities and risks in the external digital environment while enhancing control over uncertain environments, leading to the transformation of smart manufacturing in enterprises; it also aids in increasing the participation of digital talent in innovative activities and the future development of such activities [48]. With extensive exposure to advanced concepts and new technologies in their industry, digital talent can adeptly master the use of various digital technologies and tools. Their skills, experience, and knowledge level, as the most valuable resources of their enterprise [49], promote the effective utilization of data elements and the optimal allocation of other factors of production, ensuring the long-term effectiveness and sustainability of smart manufacturing.

3. Methodology

3.1. Case Study, NCA and fsQCA Hybrid Methods

Case studies constitute an important research method in the field of management and are commonly used for analyzing exploratory issues [50] and investigating complex and specific problems. Through the comparison and analysis of cases, this method reveals differences and similarities under different contexts, underlying logic, and causal relationships [51].

NCA and QCA are two logical methods for analyzing the necessity and sufficiency of multicausal phenomena. NCA quantitatively reflects the degree of necessary conditions, that is, to what extent a certain antecedent condition is a necessary condition for a certain result, whereas QCA is used to analyze whether a certain condition is necessary for the occurrence of a result and focuses on the combination of conditions [52]. The QCA method can be used to analyze the complexity of multicausal phenomena from a configurational perspective and explore the causal logic behind complex phenomena through qualitative and quantitative analysis [53].

Therefore, this study adopts the method of necessary condition fsQCA, following the sequence of “coding and measuring variables on the basis of the TOE-I framework—conducting necessary condition analysis on the basis of NCA and fsQCA methods—conducting condition variable combination on the basis of the fsQCA method—revealing the process mechanism on the basis of the case study method” to explore the impact of the interactive matching relationship of conditional factors on smart manufacturing as derived from encoding. The necessary condition fsQCA is a case-oriented method that can be used to analyze the complexity of multicausal phenomena from a configurational perspective in a limited number of cases and explore the causal logic behind complex phenomena through quantitative and qualitative analysis. Figure 1 shows the analysis process underlying the research approach.

3.2. Sample Selection

The case samples for this study were selected from China’s advanced manufacturing enterprises, and ultimately, 20 enterprises were confirmed as the case sample set, as shown in

Table 1. Basic information of the case enterprises.

Corporate Name	Core Business	Corporate Name	Core Business
SAIC Volkswagen	Automobile Manufacturing	Zhengzhou Coal Mining Machinery Group	Coal Mining Equipment Manufacturing
Ningbo Oriental	Cable Manufacturing	Hangzhou Gear Group	Gearbox and Transmission Device Manufacturing
Maanshan Iron and Steel	Steel Product Manufacturing	Shanghai Electric	Power Generation Equipment Manufacturing
Qiqihar Equipment Company	Railway Freight Car Manufacturing	AVIC Chengdu Aircraft Industry Group	Aviation Equipment Manufacturing
Changkai	Low-voltage Electrical Equipment Manufacturing	Aeolus Tyre	Tire Manufacturing
Hangzhou Cigarette Factory	Cigarette Manufacturing	Tangshan Iron and Steel	Steel Product Manufacturing
Jiuzhou Group	Military Equipment Manufacturing	Hangzhou Oxygen	Air Separation Equipment Manufacturing
14th Research Institute	Military Electronic Equipment Manufacturing	Wuhan Shipbuilding Industry	Bridge Steel Structure Manufacturing
Changfei Group	Aviation Industry Manufacturing	Hebei Iron and Steel	Steel Product Manufacturing
Changan Industries	Ordnance Equipment Manufacturing	China Southern Industries	Military Equipment Manufacturing

. The specific criteria for selection are as follows: First, these case companies are all established Chinese manufacturing enterprises that have embarked on smart manufacturing and achieved successful experiences and insights; they are continuously exploring new directions for smart manufacturing, thus meeting the requirements for the typicality of the research subjects. Second, the case enterprises cover many fields of the manufacturing industry and are all influential, large enterprises in their respective fields with strong representativeness, which helps to more accurately analyze the influencing factors in the process of smart manufacturing. Finally, the information on the case enterprises is accessible. The research team has followed up with these case enterprises for a long time and can obtain relevant internal information, ensuring the authenticity of the information and the validity of the research conclusions.

3.3. Data Collection

The data for this study were sourced from investigative interviews conducted with 20 case enterprises. The interviewees included senior management personnel, assistants to senior executives, and relevant personnel from departments such as production and operations. The interview content is primarily divided into two parts: the first part concerns the process of smart manufacturing, which includes the overall strategic planning of smart manufacturing, the establishment of complete systems, significant technological breakthroughs, and a review of key events; the second part pertains to the outcomes of smart manufacturing as well as to the experiences and insights gained. The content of the collected interview data was sent back to the enterprise personnel for review, with revisions made according to their feedback. On the basis of the interview materials, areas not covered by the interviewees or information unclear within the interviewees’ purview were supplemented through the China National Knowledge Infrastructure (CNKI), the China Enterprise Management Innovation Achievements Database, the China Case Sharing Center, the China Enterprise Confederation, and corporate websites. The multilevel, multisource collection of data complements each other, ensuring the completeness and accuracy of the data and thus enhancing the reliability and validity of the study.

3.4. Measurement, Calibration and Configuration Model

NVivo is computer-assisted qualitative data analysis software developed by QSR International in 1999 and has evolved to version NVivo11.0 as of now. This study employed NVivo 11.0 software and utilized the three-level coding method of grounded theory to analyze the case enterprise research data. A total of nineteen subcategories related to the factors influencing smart manufacturing were extracted. A hierarchical relationship analysis was subsequently conducted on the nineteen subcategories, selecting category associations and attributes with the same items and ultimately classifying them into eight core categories as variables and naming the variables in reference to existing scholarly research. Among these, the first three subcategories (digitalized smart production, digitalized smart management, and digitalized smart service) originate from the effects of the enterprise’s smart manufacturing and constitute the outcome variables. The remaining sixteen subcategories were derived from the process of smart manufacturing and served as the condition variables. The variables, secondary indicators, phenomenon definitions, and scholarly references for naming are shown in

Table 2. Variable coding results and phenomenon definition.

Variable Types	Variable	Secondary Indicators	Phenomenon Definition	Scholarly References
Outcome Variable	Smart Manufacturing	Digitalized Smart Production	Digital and smart production	Zhou et al. [4]
		Digitalized Smart Management	Digital and smart management
		Digitalized Smart Service	Customer needs and enterprise product digital and smart Services
Condition Variables	Digital Technology Breakthrough	Digital Technology Research Innovation	Research or innovation in emerging digital technologies such as artificial intelligence, big data technology, and smart systems	Onifade et al. [32]
		Digital Technology Application Breakthrough	The capability of utilizing emerging digital technologies such as 5G+ industrial internet, e-commerce, internet finance, image understanding, and smart data analysis has achieved breakthroughs
	Digital Infrastructure	Digital System	Possessing digital systems, information systems, and smart systems	Ma and Lin [37]
		Digital Platform	Possessing digital platforms, information platforms, and smart platforms
		Digital Software	Possessing digital and smart applications, software, and apps
	Digital Talent	Digital Talent Workforce Service Station	Possessing research workstation, digital technology centre and research institute of technology	Huang et al. [49]
		Digital Elite Team Building	Introducing or cultivating highly educated, high-level digital talents, digital technology teams, and key digital technology personnel
	Digital Sharing	Data Openness	Data integration, real-time data sharing, and transparent and open data	Magistretti et al. [42]
		Digital Network	Strategic alliances, industry-academia-research (IAR) collaborations, and resource and technology cooperation
	Organizational Resilience	Digital Empowerment for Organizational Collaboration	Using digital technology to enhance organizational collaborative work capacity	He et al. [40]
		Digital Empowerment for Standardized Processes	Using digital technology for the standardization of plans, work, and workflow processes
		Digital Empowerment for Incentive and Assessment	Using digital technology for work motivation and performance evaluation
	Organizational Culture	Digital Cultural Atmosphere	Cultivating digital cultural concepts and fostering digital cultural atmospheres	Pradana et al. [41]
		Digital Cultural Activities	Conducting cultural activities for the popularization and application of digital technology knowledge
	Entrepreneurial Spirit	Smart Manufacturing Acuity	The identification, learning of emerging digital technologies in globally advanced manufacturing enterprises, and guiding the smart manufacturing of enterprises	Putra et al. [20]
		Digital Leadership	In response to national policy calls and new situations, the deployment of smart manufacturing strategic layout and planning is led to guide the development of corporate smart manufacturing.

.

To ensure the robustness of the coding results, the latent Dirichlet allocation (LDA) topic model was utilized to analyze the number of topics and consistency in the smart manufacturing process data. Considering the limitation of the fsQCA method on the number of condition variables, the range of topic numbers was set from five to eight. The Jieba segmentation tool in the Python language and the “HIT Stop Word List” were used for word segmentation and stop word removal. The number of topics and coherence scores are shown in Figure 2. When the number of topics is 7, the topic consistency is the highest, which is consistent with the coding results.

Considering the variability in the volume of textual data obtained from case enterprises, this paper follows the approach of Albalawi et al. [54]; when extracting the frequency of each keyword in the text, the variables were measured by dividing the keyword frequency by the length of the text segment. For ease of presentation, each variable indicator was multiplied by 100 in this paper. In this investigation, by employing the direct calibration technique, the three reference points for the quintuple conditions and the singular outcome variable were delineated at the 95th percentile (total affiliation), the 50th percentile (confluence), and the 5th percentile (total disaffiliation) of the dataset [55].

The calibration anchors and descriptive statistics of each variable are shown in

Table 3. Fuzzy set membership calibrations and sample descriptive statistics.

Variable	Fuzzy Set Calibrations			Measure Description
	Fully in	Crossover	Fully out	Mean	SD	Max	Min
Smart Manufacturing	24.6	10.3	5.3	12.1	6.2	25.2	4.5
Digital Technology Breakthrough	27.6	18.9	8.5	17.7	6.9	31.8	8.0
Digital Infrastructure	21.2	14.8	10.5	15.1	3.5	22.1	8.2
Digital Talent	14.5	5.5	1.8	6.9	4.5	17.2	1.4
Digital Sharing	38.9	27.4	13.8	26.5	8.6	39.4	12.8
Organizational Resilience	27.4	16.2	8.3	17.6	6.1	29.6	7.0
Organizational Culture	9.8	6.3	3.2	6.2	2.3	11.3	2.4
Entrepreneurial Spirit	10.7	4.9	1.4	5.5	3.0	11.4	1.2

. Based on the TOE-I theoretical model and the aforementioned analysis of the connotative relationships of variables, a configurational perspective was adopted to explore the complex mechanisms of linkage and matching among multiple factors in smart manufacturing and construct a model (Figure 3).

4. Analysis of Results

4.1. Necessary Condition Analysis (NCA)

R software (version 4.1.0), developed by Ross Ihaka and Robert Gentleman in 1991, is widely used free and open-source statistical software across various research fields. This paper employs R software to conduct necessary condition analysis (NCA). NCA can be used to determine whether a specific factor constitutes a necessary condition for the occurrence of a particular outcome, and it can also be used to analyze the effect size of the necessary condition, which is also referred to as the “bottleneck level” in NCA, reflecting the minimum level of the necessary condition required to produce a certain outcome [56]. In this work, upper limit regression (CR) and upper limit envelope analysis (CE) were used to generate the upper limit function to obtain the effect size of the antecedent conditions [57]. As shown in

Table 4. NCA results.

Variable	Method	Accuracy	Ceiling Zone	Scope	Effect Size (d) a	p Value
Digital Technology Breakthrough	CR	85%	0.170	0.88	0.192	0.065
	CE	100%	0.122	0.88	0.138	0.063
Digital Infrastructure	CR	100%	0.059	0.89	0.066	0.537
	CE	100%	0.117	0.89	0.131	0.243
Digital Talent	CR	85%	0.069	0.88	0.078	0.409
	CE	100%	0.037	0.88	0.042	0.426
Digital Sharing	CR	75%	0.157	0.86	0.183	0.106
	CE	100%	0.119	0.86	0.139	0.100
Organizational Resilience	CR	90%	0.189	0.87	0.217	0.069
	CE	100%	0.236	0.87	0.270	0.017
Organizational Culture	CR	80%	0.245	0.90	0.271	0.004
	CE	100%	0.298	0.90	0.330	0.000
Entrepreneurial Spirit	CR	95%	0.065	0.86	0.075	0.458
	CE	100%	0.076	0.86	0.088	0.329

Note: ^a 0.0 ≤ d < 0.1: low level; 0.1 ≤ d: high level.

, the effect sizes of digital infrastructure, digital talent, and entrepreneurial spirit are all less than 0.1, and they cannot be considered necessary conditions. The Monte Carlo simulation permutation test results for digital technology breakthrough, digital sharing, and organizational resilience are not significant, and they are obviously not necessary conditions for smart manufacturing. The CR estimated effect size of organizational culture is 0.271, and the test result is significant, but its precision is less than 95%. According to the relevant standards proposed by Dul et al. [58], it also cannot be identified as a necessary condition. In summary, none of the seven antecedent conditions are necessary conditions for achieving smart manufacturing.

Table 5. NCA necessity bottleneck level (%) analysis results.

Smart Manufacturing	Digital Technology Breakthrough	Digital Infrastructure	Digital Talent	Digital Sharing	Organizational Resilience	Organizational Culture	Entrepreneurial Spirit
0	NN	NN	NN	NN	NN	NN	NN
10	NN	NN	NN	NN	NN	NN	NN
20	NN	NN	NN	NN	NN	NN	NN
30	1.7	1.0	NN	NN	NN	8.4	NN
40	9.1	3.4	NN	NN	1.1	16.9	NN
50	16.4	5.8	NN	NN	12.8	25.4	NN
60	23.8	8.2	NN	14.5	24.4	33.9	NN
70	31.1	10.5	NN	29.3	36.1	42.4	NN
80	38.5	12.9	16.9	44.1	47.7	51.0	16.7
90	45.8	15.3	34.9	58.9	59.4	59.5	33.6
100	53.2	17.7	52.8	73.7	71.0	68.0	50.6

further shows the results of the bottleneck level analysis via the NCA method. The bottleneck level indicates the level (%) that each antecedent condition must meet within its maximum observable range to achieve a certain level of the outcome variable’s maximum observable range. As shown in Table 5, to achieve the 60% smart manufacturing level, 23.8% digital technology breakthrough, 8.2% digital infrastructure, 14.5% digital sharing, 24.4% organizational resilience, and 33.9% organizational culture are needed, and the remaining two conditions do not have a bottleneck.

This paper further employs fsQCA4.0 software, developed by Charles C. Ragin et al., to test the necessary conditions. In the fsQCA method, the necessity level of an antecedent condition is measured by consistency. When the consistency is greater than 0.9, the condition is considered necessary for the outcome to occur. As shown in

Table 6. Necessity test for a single condition.

Conditional Variable	High Smart Manufacturing	Nonhigh Smart Manufacturing
Digital Technology Breakthrough	0.690	0.477
~Digital Technology Breakthrough	0.552	0.731
Digital Infrastructure	0.679	0.617
~Digital Infrastructure	0.667	0.679
Digital Talent	0.624	0.599
~Digital Talent	0.637	0.626
Digital Sharing	0.680	0.586
~Digital Sharing	0.607	0.661
Organizational Resilience	0.847	0.487
~Organizational Resilience	0.434	0.755
Organizational Culture	0.812	0.465
~Organizational Culture	0.511	0.812
Entrepreneurial Spirit	0.722	0.531
~Entrepreneurial Spirit	0.544	0.698

Note: ~ indicates the negation of the condition.

, the consistency thresholds for the necessity of individual conditions are all less than 0.9, which is consistent with the results of the NCA method analysis; in other words, there are no necessary conditions for smart manufacturing.

4.2. Sufficiency Analysis

In this work, the fsQCA method was utilized to analyze the configuration of conditions that advance smart manufacturing, where different configurations represent various element-matching pathways leading to the same outcome. Following existing studies, the raw consistency threshold is set at 0.8, the proportional reduction in inconsistency (PRI) consistency threshold is set at 0.7, and the case frequency threshold is established at 1 [59]. By comparing the nested relationships between intermediate and parsimonious solutions, we identify the core and peripheral conditions in each configuration, leading to the final fsQCA analysis results (see

Table 7. Configuration paths of high smart manufacturing.

Configurations	High Smart Manufacturing
	S1		S2	S3	S4
	S1a	S1b
Digital Technology Breakthrough	•	ⓧ	ⓧ	●	●
Digital Infrastructure	ⓧ	•	ⓧ
Digital Talent	•	•	ⓧ	ⓧ	•
Digital Sharing		ⓧ	•	●	●
Organizational Resilience	●	●	●	•	ⓧ
Organizational Culture	●	●	●	•	ⓧ
Entrepreneurial Spirit	●	●	●	ⓧ	ⓧ
Consistency	0.99	0.96	0.93	0.99	0.90
Raw coverage	0.27	0.29	0.25	0.33	0.20
Unique coverage	0.06	0.13	0.04	0.09	0.07
Solution consistency	0.90
Solution coverage	0.66

Note: “●” indicates the presence of a core condition, “” indicates the absence of a core condition, “•” indicates the presence of a peripheral condition, “ⓧ” indicates the absence of a peripheral condition, and a blank indicates that the condition is indifferent.

).

4.3. Configuration Analysis

Table 7 shows five configurations (S1a, S1b, S2, S3, and S4) that promote high levels of effectiveness in smart manufacturing. The core conditions for S1a, S1b, and S2 are the same, as are the core conditions for S3 and S4. The overall solution consistency is 0.90, with a coverage degree of 0.66, and the consistency levels of all configurational solutions exceed the threshold standard of 0.8. By analyzing these four configuration models, we can further identify the differential combination relationships of different condition variables promoting smart manufacturing.

(1) Configuration S1: “Technology + talent” empowerment organizational model. It encompasses two configuration pathways (S1a and S1b). The core conditions of these paths, which are organizational resilience, organizational culture and entrepreneurial spirit, are uniform, but the edge conditions differ. Configuration S1a is characterized by supplemental conditions of digital technology breakthroughs, digital talent, and non-digital infrastructure, with a consistency of 0.99, a raw coverage of 0.27, and a unique coverage of 0.06. Configuration S1b is defined by supplemental conditions of nondigital technology breakthroughs, digital talent, digital infrastructure, and nondigital sharing, with a consistency of 0.96, a raw coverage of 0.29, and a unique coverage of 0.13. For both configurations, digital technological breakthroughs and digital infrastructure are considered interchangeable auxiliary variables. In other words, under conditions of high organizational resilience, a strong organizational culture, a robust entrepreneurial spirit, and proficient digital talent, either a strong capacity for digital technological breakthroughs or an ample supply of digital infrastructure, can promote the advancement of smart manufacturing.

Configuration S1a represents case enterprises such as AVIC Chengdu Aircraft Industry Group and Changkai. The specific implementation processes of these companies include the establishment of a smart control centre for centralizing the processing and analysis of data from various departments to support decision-making management; the introduction of a digitalization expert team for training to enhance digital practice capabilities; the integration of real-time management systems to improve efficiency and response speed at each management stage; and the consolidation of quality management systems and the regular evaluation and update of systems, introducing new technologies. The advantages lie in the ability to strengthen internal management and technological R&D capabilities, improve employees’ digital skills, and increase the enterprise’s response speed. The disadvantages are the reliance on technology and talent iteration, high initial investment, and the pressure of continuous innovation.

Configuration S1b represents case enterprises such as SAIC Volkswagen, China Southern Industries and Changan Industries. The specific implementation processes of these companies include the construction of a data intelligence management platform to achieve intelligent management and monitoring of production; implementing an intelligent production system, integrating automation technology and intelligent equipment into production lines to increase efficiency and accuracy; encouraging employee exchange and learning; publishing case studies on digital platforms to support brainstorming and mutual learning to enhance digital capabilities; comprehensively promoting the implementation of digital concepts at all levels; and establishing standardized processes within the enterprise to ensure consistency and sustainability in production and management. The advantages lie in increased production efficiency and quality through standardized and systematic methods, reduced error rates and costs, and the promotion of smart manufacturing. The disadvantages include potential overreliance on technology, necessitating continuous investment in facility maintenance and upgrades.

(2) Configuration S2: The digital environment fosters the organizational model. In this path, we found that organizational resilience, organizational culture and the entrepreneurial spirit were the core conditions, and nondigital technology breakthroughs, nondigital talent, non-digital infrastructure, and digital sharing were the supplemental conditions. Configuration S2 has a consistency of 0.93, with a raw coverage of 0.25 and a unique coverage of 0.04. This finding indicates that in enterprises with weaker digital technological breakthrough capabilities, relatively insufficient digital infrastructure, and lower scales and levels of digital talent, leaders with an entrepreneurial spirit can foster a digital sharing environment to enhance organizational resilience and implement digital cultural concepts, thereby promoting the advance of smart manufacturing within the enterprise. This configuration path emphasizes the synergistic effect of the digital environment and organizational culture. In the digital age, which is fraught with uncertainty, leaders with a higher level of digital literacy can accurately grasp the opportunities and key points of enterprise digital integration and development, externally explore potential resource and technology cooperation, and internally enhance the effect of implementing digital cultural concepts in complex digital environments. By integrating the essence of digital integration into the transformation of enterprise production, management, and services, they can empower and enhance smart manufacturing throughout the business processes of various enterprise flows.

Configuration S2 represents case enterprises such as Aeolus Tyre and Ningbo Oriental. The specific implementation processes of these companies include the proposal of a “human defence + technical prevention” dual digital system, deploying manual monitoring and technical automated safety systems to ensure data transparency and production safety; a real-time data sharing platform is constructed, covering standardized production processes and exception alerts, to increase transparency and response speed; the entrepreneurial spirit and digital sharing environment are strengthened, emphasizing innovation and encouraging cross-departmental collaboration; and a monitoring and evaluation mechanism is established to track the effects of smart manufacturing continuously and adjust strategies and solve problems in a timely manner. The advantages lie in promoting information sharing and transparency, improving production efficiency and employee engagement, and enhancing organizational resilience and adaptability. The disadvantage is that implementation is challenging in the absence of technology and talent. The benefits of smart manufacturing may take a considerable amount of time.

(3) Configuration S3: Organization-driven technological breakthrough model. We found that digital technology breakthroughs, non-digital infrastructure and digital sharing were the core conditions and that nondigital talent, organizational resilience, organizational culture and nonentrepreneurial spirit were the supplemental conditions. Configuration S3 has a consistency of 0.99, with a raw coverage of 0.33 and a unique coverage of 0.09. This configuration type emphasizes the significant role of organizational resilience and organizational culture within the enterprise in the context of smart manufacturing. Even in enterprises lacking digital infrastructure, digital talent, and an entrepreneurial spirit, encouraging digital technological breakthroughs and constructing a digital sharing environment through organizational resilience and culture can lead to smart manufacturing. To a certain extent, organizational resilience can alleviate the resistance and impact of smart manufacturing evolution on an enterprise, neutralize unknown risks at the micro level, and encourage the enterprise to explore more diverse fields and cross-disciplinary innovation. In this process, the organizational digital cultural culture and the digital sharing environment contain directions and opportunities for innovative technology and a variety of advanced product innovations. Under the synergistic effect of these four elements, the process of smart manufacturing will further advance.

Configuration S3 represents case enterprises such as the Zhengzhou Coal Mining Machinery Group, Qiqihar Equipment Company, Hangzhou Cigarette Factory, Jiuzhou Group, Maanshan Iron and Steel, 14th Research Institute and Shanghai Electric. The specific implementation processes of these companies include the establishment of innovation incentives and continuous improvement mechanisms to promote employee participation in innovation and process optimization; the promotion of the transformation and upgrading of the equipment manufacturing industry, with a focus on the development of high-end intelligent environmental protection equipment, and the advancement of production automation to reduce energy consumption and emissions; the realization of comprehensive integration and collaboration of digital systems within the organization, developing a digital system that covers the entire organization to ensure transparent information and seamless business processes; strengthening cooperation of state-owned capital in the field of shared energy storage technology research and development, utilizing state-owned platform resources to cooperatively develop shared energy storage solutions, and promoting technical exchanges; and innovating the supply chain management culture by establishing a unified supply chain value culture to enhance the collaborative efficiency of the entire supply chain. The advantages include enhancing innovation and adaptability, increasing organizational collaborative efficiency, facilitating information sharing, accelerating the flow of information, and expediting decision-making. The disadvantages are that, in the absence of talent and infrastructure, it is difficult to drive technological breakthroughs through organizational culture; strong leadership and a solid cultural foundation are necessary, and there may be significant resistance to transformative change; and the sustainability and depth of technology innovation and sharing are constrained.

(4) Configuration S4: Talent-driven digital intelligence dominance model. In this path, we found that digital technology breakthroughs, non-digital infrastructure and digital sharing were the core conditions, and digital talent, nonorganizational resilience, nonorganizational culture and nonentrepreneurial spirit were the supplemental conditions. Configuration S4 has a consistency of 0.90, with a raw coverage of 0.20 and a unique coverage of 0.07. This configuration indicates that when an enterprise has sufficiently strong digital technological breakthrough capabilities, a sufficiently good digital sharing environment, and a sufficiently high level of digital talent, it can still achieve smart manufacturing even if the digital infrastructure is imperfect, organizational resilience is low, and the organizational culture and entrepreneurial spirit are not pronounced. Talented individuals who have digital literacy and capabilities can effectively promote enterprise digital technological breakthroughs and innovation through their strong professional abilities. By leveraging emerging technologies such as big data, they can extract core elements beneficial to the enterprise’s smart manufacturing development from the shared digital technology and resource environment, meeting the enterprise’s own transformational personalized needs and making the process of smart manufacturing evolution more rapid and stable.

Configuration S4 represents case enterprises such as Hangzhou Gear Group, Hebei Iron and Steel, Tangshan Iron and Steel, Hangzhou Oxygen, Wuhan Shipbuilding Industry and Changfei Group. The specific implementation processes of these companies include the construction of a multilevel technology R&D platform to form a core support base for innovation; deepening industry-academia-research collaboration by establishing strategic partnerships to jointly tackle key technologies; relying on project cooperation to build laboratories and accelerate technology transfer; promoting the sharing of digital resources, including platforms for sharing data, tools, and best practises, to foster knowledge sharing and rapid technological iteration; strengthening the cultivation and attraction of digital talent by designing and implementing targeted talent development programmes to attract top talent; and facilitating the commercialization of R&D outcomes and accelerating the marketization of results, transforming scientific and technological innovations into economic value and competitiveness. The advantages lie in the high level of technical and talent advantages that help enterprises maintain a leading position in technological innovation; a digital sharing culture promotes the flow of knowledge and improves R&D efficiency. The disadvantages include insufficient organizational resilience and entrepreneurial spirit, which can affect the implementation of long-term strategies and responses to external changes; overreliance on technology and talent can lead to the neglect of the optimization of organizational culture and structure.

4.4. Sensitivity Analysis

A sensitivity analysis of the antecedent conditions of smart manufacturing was performed in this study. When the consistency threshold was increased from 0.80 to 0.85, the configurational outcomes remained consistent [60]. When the PRI was increased from 0.70 to 0.8 [61], the resulting configurational outcomes provided a subset of the original outcomes (

Table 8. Sensitivity analysis.

Configurations	R1	R2	R3
Digital Technology Breakthrough	•	ⓧ	ⓧ
Digital Infrastructure	ⓧ	•	ⓧ
Digital Talent	•	•	ⓧ
Digital Sharing		ⓧ	•
Organizational Resilience	●	●	●
Organizational Culture	●	●	●
Entrepreneurial Spirit	●	●	●
Consistency	0.99	0.96	0.93
Raw coverage	0.27	0.29	0.25
Unique coverage	0.09	0.13	0.09
Solution consistency	0.95
Solution coverage	0.50

Note: “●” indicates the presence of a core condition, “•” indicates the presence of a peripheral condition, “ⓧ” indicates the absence of a peripheral condition, and a blank indicates that the condition is indifferent.

). Consequently, the research findings of this paper exhibit a high degree of robustness.

5. Discussion

5.1. Theoretical Implications

First, this study, which is based on the TOE-I framework, deeply explores the core factors influencing smart manufacturing from the microlevel perspective of manufacturing enterprises, refines the element research of the TOE-I theoretical framework, and simultaneously enriches the academic horizon of smart manufacturing. Existing research indicates that digital technology and talent [7], along with a resilient organizational system [21,22], are significant drivers of smart manufacturing. However, few studies have conducted an in-depth exploration of multidimensional factors and their interplay at the enterprise level. This paper explores the core factors influencing manufacturing enterprises in promoting smart manufacturing under the interaction of the technology, organization, environment, and individual dimensions; refines the linkage effects between the elements; and provides ideas and insights for the in-depth integration of each element.

Second, this paper reveals the pathways leading to smart manufacturing under the dimensions of technology, organization, environment, and individuals, whereby multiple factors work together synergistically, thereby expanding the depth and breadth of smart manufacturing strategy research through its causal concurrency and dynamic matching approach. The existing research has focused on the influence of single factors on corporate transformation and development [20,21], whereas studies on the joint effects of multidimensional factors have focused mainly on the construction of smart manufacturing business models [24]. Drawing on existing research, this paper, starting from the enterprise level, deeply discusses the construction of smart manufacturing systems and proposes feasible strategies for manufacturing enterprises to advance smart manufacturing. The strategy not only focuses on the deployment of digital hardware facilities and the breakthrough and application of digital soft power, such as data analysis and artificial intelligence technology, but also emphasizes the improvement of organizational resilience and the implementation of digital organizational culture during the process of the industrial integration of digital and manufacturing, enriching the research on the driving paths of smart manufacturing.

Third, unlike previous research, which has relied primarily on paradigm analysis or single-dimensional empirical testing [23], this paper employs a mixed method approach, measuring variables on the basis of factual case study coding and exploring the multilinkage pathways of how technology, organization, environment, and individual dimensions drive the integration of manufacturing enterprise elements to promote smart manufacturing from a configurational perspective via NCA and fsQCA. This study enriches the research on fsQCA methods and leverages the advantages of NCA in the quantitative analysis of necessary conditions and fsQCA in sufficient causal analysis.

5.2. Practical Implications

First, top-level leaders should actively cultivate their entrepreneurial spirit and enhance their perception and precise understanding of the complex external digital environment to provide more accurate decision-making in the process of smart manufacturing. At the same time, they should overcome existing cognitive limitations and construct a digital strategic mindset from multiple dimensions, including technology, organizations, the environment, and individuals. They should clarify the direction, goals, key points, and pathways of corporate smart manufacturing to guide the transition from traditional production models to smart manufacturing, and promote the sustainable development of the enterprise.

Second, enterprises should review their resource endowments and allocate resources purposefully to identify drivers of smart manufacturing that match their situation. Enterprises with strong organizational resilience and effective implementation of organizational culture should coordinate the analysis of digital resource conditions related to technology, the environment, and individuals. They should scientifically integrate and synergize digital elements according to their own situation, build organizational resilience for rapid response to crises, and empower a digital organizational culture, continuously leveraging the advantages of internal organizational entities. Enterprises with obvious competitive advantages in digital areas can, on the one hand, actively build digital talent service centres, improve training systems for digital talent, increase efforts to attract talent, and create a composite digital talent team, thereby laying a solid talent foundation for the sustainable development of the enterprise. On the other hand, they can build a digital concept compatible with the enterprise and develop organizational resilience, thereby strengthening the ability to acquire digital technology capabilities. By relying on the driving force of organizational and individual dimensions, they can strengthen industry–academia–research cooperation, government, strategic alliances, and external resource allocation to accelerate digital technology innovation and the transformation of achievements, aiding in the autonomy of key core technologies and the development of “bottleneck” technology. Digital sharing across the entire business process of the enterprise can be promoted to provide opportunities and protection for cross-departmental collaboration.

Third, the government should strengthen local policy guidance and financial support to encourage enterprises in the R&D of emerging digital technologies and cooperate with externally related enterprises, promote the development of local digital industry clusters, and provide a good sharing environment for the advancement of smart manufacturing. Enterprises should be guided to carry out organizational changes and use digital technology to increase organizational resilience and implement organizational culture in ways that fit their interests and goals, improving the enterprise’s ability to resist risks in the face of a complex environment during the process of smart manufacturing.

6. Conclusions

6.1. The Main Conclusions

This paper, grounded in the TOE-I analytical framework and utilizing 20 advanced manufacturing enterprises as case samples, employs case study NCA and fsQCA to explore the antecedent conditions and pathway mechanisms of the “technology-organization-environment-individual” dimensions used by manufacturing enterprises to advance smart manufacturing from a configurational perspective. The conclusions of the study are as follows:

(1)

Digital technology breakthroughs, digital infrastructure, digital talent, digital sharing, organizational resilience, organizational culture, and the entrepreneurial spirit are the core influential factors in the pursuit of smart manufacturing by manufacturing enterprises.

(2)

None of the seven elements, including digital technology breakthroughs, can independently become a necessary condition for smart manufacturing. However, organizational resilience and organizational culture play more universal roles as key factors.

(3)

There are configuration models leading to smart manufacturing: the “technology + talent” empowerment organizational model, the digital environment fosters the organizational model, the organization-driven technological breakthrough model, and the talent-driven digital intelligence dominance model. These four models represent different configurational scenarios of technology, organization, environment, and individuals, and the differences in pathways depend on the multiple scheme choices for smart manufacturing by different enterprises.

(4)

There is a potential substitutive relationship between digital technology breakthroughs and digital infrastructure in the “technology + talent” empowerment organizational model, and organizational resilience, organizational culture, and the entrepreneurial spirit are crucial safeguards for the smooth implementation of pathways to smart manufacturing. In contrast, digital infrastructure plays a more indirect supportive role in the driving pathways that include factors such as digital talent and digital sharing.

The main contribution of this study is to reveal the factors and configuration paths that drive smart manufacturing at the microlevel of enterprises, explain the complex interplay among multidimensional factors within the TOE-I framework, and address the deficiencies in existing research that explain the multi-element drivers of smart manufacturing. This research can assist manufacturing enterprises in their transition to digital intelligence and in establishing sustainable competitive advantages.

6.2. Limitations and Future Directions

This paper has the following limitations. First, there is a certain degree of limitation in the selection of case studies. For this paper, case enterprises were selected from typical manufacturing enterprises in China that have integrated informatization and industrialization into smart manufacturing, with constraints on both the number of case companies and the level of detail in the interview materials. Future research can conduct more in-depth analyses of smart manufacturing cases in specific industries, such as the automotive industry, which is highly data-dependent, to enrich and improve existing theories.

In addition, in terms of the selection of antecedent conditions, this paper is based on the TOE-I framework, and a limited number of factors are selected to explore the driving paths of smart manufacturing. However, smart manufacturing is influenced by a multitude of factors, including government policies and the characteristics of industry competitors. Therefore, future studies could adopt alternative theoretical perspectives to conduct a more comprehensive investigation into the driving factors of smart manufacturing, thereby revealing more complex causal relationships.