Research on Innovation in Enterprise Ecosystem Restructuring Driven by Big Data and Artificial Intelligence

다국어 초록 (Multilingual Abstract)

In the context of the digital economy, big data and artificial intelligence are transforming the operation mode and value creation logic of the enterprise ecosystem. Existing studies have discussed from perspectives such as digital technology, ecosystem reconfiguration, and value creation. However, there are still deficiencies in the measurement of key variables, cross-level empirical evidence, and the explanation of the mechanism of action. Specifically, the operationalization measurement of data capabilities and AI capabilities in existing studies is still not direct enough. The empirical evidence at the enterprise-level and the industry-level has not been fully connected. How does technological capability influence ecosystem reconfiguration through specific paths and further transform into value creation also needs to be further clarified. Based on ecosystem theory and international business theory, this paper constructs an integrated analysis framework of ＂technical capabilities - reconfiguration paths - ecosystem value creation＂, and focuses on answering three questions: How do data capabilities and AI capabilities affect the reconfiguration of the enterprise ecosystem, how does the reconfiguration of the ecosystem promote value creation, and how does the external context affect the above relationship.
This paper conducts an empirical analysis by combining data at the enterprise-level and the industry-level. At the enterprise-level, using 86 A-share listed companies in the Chinese digital economy sector from 2018 to 2023 as samples, 516 annual observations of enterprises were formed; at the industry-level, using 12 sub-sectors of the digital economy during the same period as samples, 72 annual observations of industries were formed. In terms of research methods, this paper comprehensively employs panel regression models, mediation effect tests, moderation effect tests, and instrumental variable methods to test the relevant research hypotheses.
The research results indicate that both data capabilities and AI capabilities have a significant positive impact on the reconfiguration of the enterprise ecosystem, with the influence of AI capabilities being more prominent. Decision-making intelligence, collaboration networking, and structural platformization play a mediating role in the process of technology capabilities influencing the reconfiguration of the ecosystem. Among them, the path of ＂AI capabilities - collaboration networking - ecosystem reconfiguration＂ has strong explanatory power. Further analysis reveals that ecosystem reconfiguration can significantly promote the creation of ecosystem value. The strengthening of external regulatory intensity strengthens the relationship between technology capabilities and ecosystem reconfiguration, while the internationalization level strengthens the relationship between ecosystem reconfiguration and value creation. Heterogeneity tests also show that these effects are more pronounced in platform-based enterprises and highly digitalized industries.
The main contributions of this study are reflected in three aspects: Firstly, it combines the ecosystem theory with the international business theory to explain the intrinsic relationship among technological capabilities, organizational restructuring and value creation; Secondly, it directly measures data capabilities and AI capabilities operationally, enhancing the consistency between variable measurement and the research topic; Thirdly, it adopts a dual-level empirical design combining enterprise-level mechanism analysis and industry-level result verification, providing empirical evidence for understanding the effects of technology-driven ecosystem reconfiguration and value creation. In practice, enterprises should give priority to improving AI capabilities and focus on collaborative networking. The industry needs to formulate collaboration standards differently, and the country should improve the digital infrastructure and regulatory framework. Due to the limitations of data coverage period and the measurement methods of some variables, future research can further focus on the impact of generative AI on the reconfiguration of enterprise ecosystems and expand it by introducing more abundant international context variables.

목차 (Table of Contents)

• I. Introduction 1
• 1. Research Background and Significance 1
• 1.1 Research Background 1
• 1.2 Research Significance 3
• 2. Current Research Status Domestically and Internationally 6
• 2.1 Existing Research Consensus 6
• 2.2 Preliminary Research Gap 7
• 3. Research Approach and Content 8
• 3.1 Overall Research Approach 8
• 3.2 Core Tasks of Each Chapter 9
• 4. Research Methods 11
• 4.1 Literature Review Method 11
• 4.2 Case Study Method 12
• 4.3 Secondary Data Analysis Method 12
• 5. Research Innovations 14
• II. Literature Review and Hypothesis Development 16
• 1. Big Data and AI-Driven Restructuring of Enterprise Ecosystems 16
• 1.1 Technical Perspective: Data/AI as the Core Driving Force for Reconstruction 16
• 1.2 Organizational-Level: Academic Consensus and Controversies on the Pathway for Restructuring 18
• 1.3 Environmental-Level: Constraints and Support from the External Environment for Reconstruction 21
• 1.4 Relevant Theoretical Foundations 24
• 2. Value Creation Effect of Ecosystem Restructuring 27
• 2.1 Enhancement of Ecosystem Performance: From "Single-Enterprise Profitability" to "Ecosystem Value Addition" 28
• 2.2 Acquisition of Competitive Advantage: Network Effects and Ecosystem Lock-in 30
• 2.3 Value distribution mechanism: from "zero-sum game" to "co-creation and sharing" 32
• 3. Research Gaps and Focus of Research Questions 35
• 3.1 Limitations of Existing Research 35
• 3.2 The research question is clearly defined 39
• 4. Theoretical Integration Logic 41
• 4.1 Ecosystem Theory: The Core Explanatory Framework That Reveals Structural Evolutionary Mechanisms 41
• 4.2 International Business Theory: A Situational Supplementary Framework that Defines the Boundaries between Transnational Situations and Value Creation 42
• 4.3 Construction and Theoretical Innovation of the Cross-Theoretical Transmission Model 44
• 5. Summary of This Chapter 46
• III. Hypothesis Formulation and Research Model 48
• 1. Mechanisms and Pathways for Enterprise Ecosystem Restructuring 48
• 1.1 Evolution of Enterprise Ecosystems and Intelligent Transformation 49
• 1.2 Big Data and AI-Driven Pathways for Enterprise Restructuring 54
• 1.3 Drivers and Conditions for Enterprise Ecosystem Restructuring 60
• 1.4 Organizational Adaptation Challenges During the Reconstruction Process 65
• 2. Research Hypotheses 70
• 2.1 Fundamental Assumptions: The Direct Impact of Data/AI Capabilities on Ecosystem Restructuring 70
• 2.2 Mediation Hypotheses: Restructuring Pathways as Mediators 71
• 2.3 Results and Moderation Hypotheses: The Value Creation Effect and Boundary Conditions 73
• 3. Research Model and Empirical Design 75
• 3.1 Conceptual Model 75
• 3.2 Research Hypothesis Framework Overview 77
• 3.3 Research Questions and Their Dual Empirical Correspondences 79
• IV. Research Design 81
• 1. Data Sources and Sample Selection 81
• 1.1 Empirical1 (Enterprise-Level) Data: Panel of Listed Companies in the Digital Economy 84
• 1.2 Empirical2 (Industry-level) Data: Digital Economy Subsector Panel 87
• 2. Variable Definitions and Measurement Methods 90
• 3. Identification Strategies and Endogeneity Issues 97
• 3.1 Identification of Endogeneity Sources 98
• 3.2 Selection of Instrumental Variables 99
• 3.3 Argumentation of Exclusivity Assumptions 101
• 3.4 Weak Instrumental Variable Test 102
• 4. Model Specification and Estimation Methods 103
• 4.1 Empirical Model One (Enterprise-Level): Testing H1, H2, H3a–H3c, and H5 104
• 4.2 Empirical Model Two (Industry-Level): Testing H4 and H6 108
• 4.3 Robustness Test Methods 109
• 5. Summary of This Chapter 110
• V. Empirical Study 1: Enterprise-level Technical Capability and Ecosystem Reconfiguration 113
• 1.Descriptive Statistics and Correlation Analysis 113
• 1.1 Descriptive Statistics (Enterprise-Level Sample) 113
• 1.2 Correlation Analysis 115
• 2. Regression Result Analysis 118
• 2.1 Main Effect Test (H1, H2) 118
• 2.2 Mediation Effect Test (H3a, H3b, H3c) 120
• 2.3 Moderating Effect Test (H5) 122
• 3. Robustness and Heterogeneity Tests 124
• 3.1 Robustness Test (Variable Replacement) 125
• 3.2 Heterogeneity Test (Grouped by Enterprise Type) 126
• 4. Summary of Empirical Research 127
• VI. Empirical Study II: Ecosystem Reconfiguration at the Industry-Level and Ecosystem Value Creation 130
• 1. Descriptive Statistics and Correlation Analysis 130
• 1.1 Descriptive Statistics (Industry-level Sample) 131
• 1.2 Correlation Analysis 132
• 2. Regression Result Analysis 134
• 2.1 Main Effect Test (H4) 135
• 2.2 Moderation Effect Test (H6) 136
• 3. Robustness and Heterogeneity Tests 138
• 3.1 Robustness Test (Instrumental Variable Method) 139
• 3.2 Heterogeneity Test (Grouped by Industry Digitalization Level) 140
• 4. Comparison and Discussion of Dual-level Empirical Results 142
• 5. Summary of Empirical Study Two 144
• VII. Discussion and Conclusion 146
• 1. Main Research Conclusions 146
• 2. Responses to Research Questions (RQ1 - RQ3) 148
• 2.1 Response to RQ1: How does the technological capabilities of enterprises drive ecosystem restructuring 148
• 2.2 Response to RQ2: How does the external environment moderate the relationships among technological capabilities, ecosystem restructuring, and value creation 149
• 2.3 Response to RQ3: What are the consistency and differences between the two empirical results 151
• 3. Theoretical Contributions 152
• 4. Practical Implications 153
• 4.1 Enterprise-Level: Strengthening AI Capability, Networked Collaboration, and Platform-Based Organization 153
• 4.2 Industry-Level: Improving Digital Standards, International Cooperation, and Basic Digital Conditions 154
• 4.3 Government-Level: Using Regulation to Support Data Governance, Algorithmic Compliance, and Trust-Based Ecosystem Restructuring 155
• VIII. Research Limitations and Future Research Directions 157
• 1. Research Limitations 157
• 2. Future Research Directions 158
• References 160