Research on the Impact of Digital-Real Integration on Logistics Industrial Transformation and Upgrading under Green Economy

Abstract

Digital-real integration and green economy have become an important driving force to propel logistics industrial transformation upgrading and sustainable development. This paper analyzed the impact of digital-real integration on the transformation and upgrading of the logistics industry and the role of green economy from the perspectives of endogenous growth theory, green development theory, and industrial organization theory. Utilizing the panel data spanning 2013 to 2022 from 11 cities in Hebei Province, China, this study established the measurement indicator systems of logistics industrial transformation and upgrading, digital-real integration, and green economy, and measured the integrated developmental level employing an entropy weight–TOPSIS–grey correlation model. Further, the benchmark regression model, moderation effect model, and threshold effect model were applied to delve into the influence mechanism of logistics industrial transformation and upgrading influenced by digital-real integration under the green economy. The research results showed that: Digital-real integration contributed favorably to the transformation and upgrading of the logistics industry. The green economy exerted a beneficial moderating influence on the process, where digital-real integration affected the transformation upgrading of the logistics industry, and the moderating effect of the innovation elements was significantly the strongest. Digital-real integration exhibited a single threshold effect, and the moderating impact of the green economy also had a single threshold condition, and once both thresholds were surpassed, digital-real integration significantly promoted the logistics industrial reformation and upgrading. Drawing from the aforementioned conclusions, this study advanced development recommendations in terms of strengthening regional cooperation, intensifying digital-real integration, exerting the effect of the green economy, and establishing dynamic monitoring and evaluation mechanisms.

1. Introduction

As a basic, strategic, and trailblazing sector of the national economy, the logistics industry performs pivotal functions of commodity, information, and capital flows, and its development level is not only directly related to the enhancement of the national economic operational efficiency, but also the key to building a modernized logistics system and realizing the strategy of a strong logistics country. In the era of digitalization, the current wave of technological revolution and industrial change is profoundly affecting the global economic pattern and industrial structure, changing the way of production, consumption, and communication, which brings new development opportunities for the logistics industry. The logistics industrial transformation and upgrading is the decisive measure to align with the economic evolution of the new era and enhance international competitiveness, which is crucial for promoting regional synergy, optimizing industrial layout, improving logistics efficiency, and building a world-class city cluster. However, with the constant changes in market demand and the intensification of environmental problems, the logistics industry is facing many problems in the process of development, such as insufficient innovation in mode, system, technology, and business format, and inadequate green and sustainable development [1]. In the face of these challenges, seeking the road of logistics industrial transformation and upgrading is an urgent need for the current pursuit of high-quality economic development.

As an important trend of industrial development in the new era, the convergence of digital technology with the real economy (known as “digital-real integration”) is a key component and practical application scenario of Industry 4.0 and big data, and is the profound application of digital technology’s innovative achievements into the real economy. This integration is characterized by industrial digital upgrading, enterprise digitization, and labor skill transformation, and can activate the data elements of the industry, thereby creating more value and forming a new economic growth point and competitive advantage [2]. Deep digital-real integration is gradually becoming a novel driver for logistics industrial transformation and upgrading and economic growth. While pursuing economic benefits, the logistics industry should also pay attention to the sustainability of industrial development. The green economy, as an important trend in current economic development, emphasizes the effective use of resources and is committed to promoting innovation and technological progress to minimize environmental harm and strike a balance between economic growth and ecological protection [3]. Its introduction provides a new theoretical guidance and practical path for the logistics industrial transformation and upgrading. Although digital-real integration is regarded as containing huge development momentum, its actual effect in the transformation and upgrading of the logistics industry, especially its specific impact under the guidance of green economy, needs to be fully empirically analyzed and clearly tested. Therefore, this paper investigates how digital-real integration affects logistics industrial transformation and upgrading, and the role mechanism of the green economy effect, which is of great significance for realizing the shift from traditional to emerging drivers in the logistics industry, as well as regional economic sustainable development.

The main contributions of this paper are as follows: First, this paper explores an important but little-known topic; that is, the impact of digital-real integration on the transformation and upgrading of the logistics industry under the green economy. Based on the theories of endogenous growth, green development, and industrial organization, new insights are put forward from the perspectives of technological integration and production mode reformation, green transformation, and stage effect changes. Secondly, compared with the previous limitation of development-level estimation [4,5], this paper innovatively adopts a multi-dimensional perspective to construct a more comprehensive measurement index system for logistics industrial transformation and upgrading, digital-real integration, and the green economy. At the same time, on the basis of the entropy weight and TOPSIS methods [6,7], the grey correlation theory is incorporated to construct an entropy weight–TOPSIS–grey correlation model for system-level estimation. This broadens the evaluation horizon of the three aspects and enriches the measurement methodology in related fields. Thirdly, using panel data from 11 cities in Hebei Province, China, the facilitating role of digital-real integration and the regulating mechanism of green economy are tested, and the critical conditions of the two are clarified by the threshold effect model. The results not only deepen the understanding of the driving forces of logistics industrial transformation and upgrading, but also reveal the internal mechanism of green economy as a moderating factor, providing more detailed and insightful insights for subsequent policy formulation and practical guidance.

The remainder of the paper is organized as follows: Section 2 offers a review of relevant literature. Section 3 presents the theoretical analysis and the formulation of research hypotheses. Section 4 introduces the models required for level measurement and hypothesis testing and describes the relevant variables and data. Section 5 empirically analyzes the levels of logistics industrial transformation and upgrading, digital-real integration, and green economy, verifying the research hypotheses and conducting the robustness tests. Section 6 discusses the validity of the hypotheses in light of empirical results and theoretical analysis, indicating the limitations of this study and the future research directions. Section 7 shows the research conclusions and development recommendations.

2. Literature Review

2.1. Logistics Industrial Transformation and Upgrading

Industrial transformation and upgrading encompass both the evolution of industries and their enhancement. Gereffi [8] divided industrial upgrading into intra-industrial upgrading, domestic inter-industrial upgrading, and international industrial upgrading based on resource allocation. Jin [9] analyzed industrial transformation and upgrading, which involve the transformation and advancement of industries from the perspective of the mission and value of industry. Pipkin [10] believed that the core features of industrial structure upgrading are reflected in the reconfiguration and upgrading of production factors. Lin [11] pointed out that the industrial structure is unimodal, and there are dominant industries at any time in the process of economic development. Elberry [12] argued that the drivers of industrial transformation are not limited to technological advances, but are intertwined effects between economic, social, and environmental factors. As the logistics industry flourishes, scholars have increasingly focused on its current development status and potential avenues for transformation and upgrading. Dubey [13] found that carbon emission reduction is a key enabler of green logistics development. Wang [14] explored the application of big data analytics in logistics and supply chain management. Zaman [15] illuminated that logistics greening transformation is reflected in the health of the environment and economies of scale. Liu [16] stated that a core concern of logistics industrial transformation and upgrading tends to focus on efficiency improvement. Winkelhaus [17] defined the term “Logistics 4.0” and developed its comprehensive framework, while Belmonte [18] described the emergence of Logistics 4.0 under Industry 4.0. Ar [19] quantitatively investigated the feasibility of blockchain technology in the logistics industry, based on which Görçün [20] proposed a robust and efficient decision-making tool that addressed the problem of blockchain platform selection for the logistics industry. Samadhiya [1] discussed the critical role of factors such as the Internet of Things (IoT) in driving a harmonious symbiosis between logistics operations and economic, environmental, and social sustainability. Banerjee [21], based on a yearly sample of logistics firms, discovered that higher-capacity managers are adept at finding the right balance between immediate liquidity needs and long-range capital allocations, which drives significant improvements in firm performance. Wang [22] suggested that under the new development pattern, the logistics industry shoulders the crucial mission of achieving the “dual carbon” goal and must accelerate the green and low-carbon transformation. Hamidi [5] concluded that the digital transition of the maritime logistics industry is strongly driven by the extent of enterprises’ readiness in digital technologies.

2.2. Digital-Real Integration

The sustainable, healthy, and efficient development of the industrial economy requires digital-real integration. Eller [23] analyzed the motivations, expected outcomes, and challenges behind the digitization of enterprises. Elia [24] provided an autonomous definition of digital entrepreneurial ecosystems and focused on the convergence of digital technologies in reshaping enterprises. Tian [4] found that the deep integration of digital technologies in the real business and financial sectors can accelerate the kinetic transformation of economic growth. Apart from emphasizing the formation of digital-real integration, scholars have increasingly studied its application in industrial transformation and upgrading, innovation process, sustainable development, etc. Urbinati [25] suggested that enterprise digital transformation is gradually becoming the core driver of innovation, and Blichfeldt [2] also argued that the use of digital technology promotes product and service innovation, and provides powerful technological and instrumental support for enterprise innovation. Llopis-Albert [26] analyzed the benefits of integrating the automotive industry with the digital economy from the perspectives of manufacturers and consumers, respectively. Gao [27] found that this convergence has evolved into a favorable catalyst for economic development, and it serves as a significant impetus for achieving economic growth of a superior quality. Bendig [28] regarded digital-real integration as a core strategy for business development to improve international competitiveness and enhance environmental advantages. Meng [29] discovered that the fusion of digital and real facilitates industrial green transformation. Khayyat [30] argued that the adoption of green technologies is essential for green logistics practices and sustainability initiatives. Yu [31] pointed out that the deep integration of digital technologies and the logistics industry can drive green innovation and performance growth of enterprises, mitigating the negative impact of their rapid expansion on the environment and society. Chu [32] emphasized that leveraging digital technologies to improve all elements and processes within the logistics system, with the aim of vigorously developing digital logistics, is conducive to realizing the transformation and upgrading of the logistics industry. Wagner and Avelar [33,34] examined the relationship between digitization and sustainability based on the fashion industry and SMEs. Bagherian and Lei [35,36], additionally, analyzed how digital advantages can be used to address the challenges of energy sustainability and promote the growth of green economy. Tolentino-Zondervan [37] suggested that textile and apparel firms utilize digital technologies for information traceability to improve transparency and enhance accountability. Alsakka [38] found, through a case study, that the convergent application of digital twin technology in off-site construction plant operations can reduce deviations between planned and actual production.

2.3. Green Economy

The green economy is based on the traditional industrial economy. Kosoy [39] stated that the green economy aims to achieve low-carbon development, improve resource utilization efficiency, and demonstrate strong social inclusion. Zeb [3] mentioned that the concept of green economy has become a new buzzword in the discussion of the world’s response to economic crises, and Zhou [40] also believed greening fits the world economic development trend and is an intrinsic motivation of economic transformation. Siyal [41] illuminated that implementing a corporate green development strategy is crucial under the challenge of global competition and institutional pressure. As the green economy proliferates, how to measure its development level, investigate the underlying factors that influence it, and analyze its driving effects have also become the focus of research. Xu [6] constructed an index system from five aspects of social progress, economic effectiveness, innovation momentum, and fairness for the people, then measured the development level of the green economy by using entropy thought. The green economy is an economic form supported by innovation, and Walsh [42] argued that innovation is the key driver for industrial structural upgrading, Chaudhuri [43] analyzed the advantages of green innovation applications in improving the environmental performance of companies, and Shou [44] quantitatively confirmed the inverted U-shape relationship between green technological innovation and the market value of logistics. Aisbett [45] concluded that green industrial policy is an important driver of international green economy cooperation, and the formation of cooperative organizations helps foster a mutually beneficial outcome for both the environment and industry, while Akalibey [46] qualitatively analyzed the differences and correlations of green economy drivers in five environmentally sensitive industries. In addition, the root of the green economy lies in green sustainability. Ning [47] suggested that the dynamic capabilities of firms are an important source of sustainable competitive advantage, while Yadav [48] found that there is a positive impact of the integration of digital technology and green elements on the sustainable performance of supply chains. Soto [49] surveyed that foreign direct investment can accelerate the global transition to a green economic model.

In summary, research on logistics industrial transformation and upgrading, digital-real integration, and the green economy has increased in recent years, and the existing results have more often discussed logistics industrial transformation and upgrading under the promotion of the green economy or digital-real integration, but the literature on the relationship between the three is very sparce, especially the scarcity of quantitative examination of the impact mechanism. Furthermore, logistics industrial transformation and upgrading are mostly measured from a single dimension, so the construction of the index system is not perfect enough. The level assessments mostly used the entropy weighting approach and TOPSIS methodology, and the evaluation method lacks the consideration of index correlation analysis. Therefore, this paper takes 11 cities in Hebei Province as examples, analyzes the underlying dynamics of digital-real integration in promoting logistics industrial transformation and upgrading from the theoretical and practical levels, and constructs relevant measurement index systems and entropy weight–TOPSIS–grey correlation evaluation model to assess the generalized development of the transformation and upgrading of the logistics industry, the digital-real integration, and the green economy. On this basis, the direct, indirect, and nonlinear effects of logistics industrial transformation and upgrading affected by digital-real integration under the green economy are examined, which also provides theoretical and methodological references for the sustainable development of similar industries.

3. Theoretical Analysis and Research Hypothesis

3.1. The Direct Impact of Logistics Industrial Transformation and Upgrading Affected by Digital-Real Integration

As digital technology progresses rapidly, digital-real convergence is gradually becoming a key driving force for the logistics industrial transformation and upgrading. The endogenous growth theory emphasizes that technological progress is the endogenous driving force for industrial structure transformation and economic growth [4]. Digital-real integration integrates digital technology into all aspects of the logistics industry, helping to break down traditional logistics industry barriers and explore new business areas, service models, and management methods, thus promoting industrial transformation and upgrading. Through the introduction of IoT [1], big data [15], blockchain [20], and other information technologies, logistics enterprises can dig deeper into the consumption habits and demand preferences of customers, provide more personalized and customized services, flexibly adjust service strategies with the help of customer feedback and data analysis, and continuously optimize the service process, to improve the service quality and efficiency of logistics enterprises while enhancing customer satisfaction and trust and loyalty to logistics enterprises [2]. This technology-driven model enables enterprises to not only promote the sharing and application of logistics information in the supply chain, realize real-time collection, processing, and analysis of logistics information, and effectively improve the efficiency of information flow, but also predict market supply and demand, optimize inventory need, formulate more reasonable transportation routes and distribution plans, and achieve accurate matching and dynamic adjustment of resources, thus improving resource utilization and production efficiency, reducing operating costs, shortening the logistics cycle, and injecting new vitality into the logistics industry [14,19]. In addition, through the application of intelligent robots, automated warehousing, driverless vehicles, and other equipment technologies, logistics enterprises can realize the automated operation of warehousing, distribution, and other links, reduce labor costs, improve the efficiency and accuracy of logistics operations, and then drive the logistics industry toward a more intelligent and efficient direction.

Hypothesis H1:

Digital-real integration can favorably facilitate logistics industrial transformation and upgrading.

3.2. The Indirect Impact of Logistics Industrial Transformation and Upgrading Affected by Digital-Real Integration

The theory of green development indicates that economic growth, ecological environment, and social progress are interconnected and mutually reinforcing [50]. The in-depth combination of digital technology and the logistics industry has promoted the logistics industrial development toward intelligence, informatization, and digitalization, and the introduction of the green economy, with innovation as the driving force, has further accelerated this trend, which is both an important influencing factor for the logistics industrial transformation and upgrading and exerts an influential function in digital-real integration. For one thing, the continuous advancement of digital-real integration has provided strong technical support and innovation power for the development of the green economy [25,46], and the new business models and formats it has spawned have also expanded the market space for green innovation. This data-driven innovation approach helps logistics enterprises achieve technological revolutionary breakthroughs, innovative allocation of production factors, collaborative cooperation in supply chains, and value co-creation [43], and promotes the greening and sustainable development of the logistics industry, which in turn promotes the upgrading and optimization of the entire industrial ecosystem, meets the needs of contemporary consumers for green products and services, and enhances core competitiveness [29,36]. For another, with the successive emergence of green technologies and the increased demand for green products, the mode and mechanism of digital-real integration are also evolving and improving, and the demand for digital technology in various fields of the logistics industry continues to increase, and the application scenarios of digital-real integration are also expanding. This deep integration driven by the green economy [28] helps logistics enterprises expand their business areas under the guidance of innovative thinking, reduce waste generation and emissions, increase revenue sources, and realize the shift from traditional logistics to modern, smart, and green logistics, while advancing industrial structural optimization and economic benefit augmentation [48]. This underscores the synergistic and complementary interplay between the green economy and digital-real integration.

Hypothesis H2:

Green economy positively moderates the impact of logistics industrial transformation and upgrading affected by digital-real integration.

3.3. The Nonlinear Impact of Logistics Industrial Transformation and Upgrading Affected by Digital-Real Integration

During the logistics industry’s transformation and upgrading, the application degree of digital and real integration is different, and its impact will also show differentiation [2]. Moreover, according to the industrial organization theory, the market structure of a firm has an impact on its market behavior and market performance [51]. The initial stage of digital-real integration is often accompanied by low capital investment, and the management ability of enterprises has not yet kept pace with technological development and may only apply some simple automation equipment or information systems [23]. Although these technologies can improve operational efficiency to a certain extent, the technical level is not high enough to create economies of scale. However, with the deepening of digital-real integration, digital technology has gained increasing prevalence in the logistics industry, and the management mode and organizational structure of enterprises have been gradually optimized to form a modern management system that adapts to digital operations, promoting more efficient resource allocation and more accurate decision support [32], thus significantly improving operational efficiency and service quality. In addition, the green economy is the power source for digital-real integration to promote logistics industrial transformation and upgrading. Insufficient development of green economy often reflects that regional green economic development policies have not been fully implemented and regulated [40], which also means that enterprises have weak adaptability in the face of challenges, such as digital transformation and environmental pollution [52], lack green core technology and management innovation ability, fail to make full use of the technological advantages brought by digital-real integration, and are unable to adjust and optimize their own operating mode and management process promptly, on time, resulting in limited effects of digital-real integration and difficulty in fully realizing its potential [44]. When the green economy continues to improve to a certain level, logistics enterprises have higher technological application and green innovation capabilities and can more effectively integrate digital resources and innovative service modes, thereby greatly enhancing the market competitiveness of enterprises.

Hypothesis H3:

The impact of logistics industrial transformation and upgrading affected by digital-real integration has a nonlinear characteristic, and there is also a threshold condition for the moderating effect of the green economy.

4. Research Methods and Data Sources

4.1. Model Construction

This paper constructed an entropy weight–TOPSIS–grey correlation model to measure the comprehensive development situation of the logistics industry’s transformation and upgrading, digital-real integration, and the green economy, and the mechanism of logistics industrial transformation and upgrading affected by digital-real integration was tested and analyzed through a benchmark regression model, moderation effect model, and threshold effect model. We defined the composite system composed of the three as U = {L, D, P}, the subsystem of the transformation and upgrading of the logistics industry as L, the subsystem of the integration of digital-real as D, and the subsystem of the green economy as P.

4.1.1. Entropy Weight–TOPSIS–Grey Correlation Model

As a highly integrated evaluation approach, the entropy weight–TOPSIS–grey correlation model introduces information entropy to objectively assign weights to evaluation indexes, and skillfully integrates the evaluation logic of the TOPSIS method and the theoretical essence of grey correlation analysis. This model not only evaluates each object from the perspective of static distance measurement, but also takes into account the deep-seated correlation and change trends of data, which significantly enhances the credibility of the evaluation results and effectively avoids subjective assumptions. To scientifically measure the generalized development level of these subsystems and obtain the variable data required for hypothesis testing, drawing on the research of Oztaysi and Alao [53,54], this study assigned weights using the entropy weight method in light of the degree of change in various indicator data, and the grey correlation method and TOPSIS method were combined based on the weighting to measure the system development level from the two aspects of static distance and dynamic trend, to improve the accuracy and reliability of the evaluation results. The following outlines the construction of the entropy weight–TOPSIS–grey correlation model:

(1)

Calculation of Indicator Weights

After normalizing the values of each measurement index to remove the scale and positive and negative influences, the information entropy (${\mathrm{E}}_{\mathrm{j}}$) and weights (${\mathsf{\Omega }}_{\mathrm{j}}$) of each index were calculated:

$${\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{{\mathrm{X}}_{\mathrm{i}\mathrm{j}}-\mathrm{m}\mathrm{i}\mathrm{n}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)}{\mathrm{m}\mathrm{a}\mathrm{x}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)}}$$

(1)

$${\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)-{\mathrm{X}}_{\mathrm{i}\mathrm{j}}}{\mathrm{m}\mathrm{a}\mathrm{x}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left({\mathrm{X}}_{\mathrm{i}\mathrm{j}}\right)}}$$

(2)

$${\mathrm{E}}_{\mathrm{j}}=-{\displaystyle \frac{1}{\ln \mathrm{n}}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}\left({\displaystyle \frac{{\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}}}\ln \left({\displaystyle \frac{{\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}}}\right)\right)$$

(3)

$${\mathsf{\Omega }}_{\mathrm{j}}={\displaystyle \frac{1-{\mathrm{E}}_{\mathrm{j}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{m}}\left(1-{\mathrm{E}}_{\mathrm{j}}\right)}}$$

(4)

Among them, Equation (1) was used to normalize positive indicators and Equation (2) was used to normalize negative indicators, where $\mathrm{i}$ represents the evaluated object, $\mathrm{j}$ is the measurement index, ${\mathrm{X}}_{\mathrm{i}\mathrm{j}}$ and ${\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}$ each represent the initial and standardized index values, $\mathrm{n}$ signifies the amount of evaluated objects, and $\mathrm{m}$ denotes the amount of measurement index. Equations (3) and (4) were applied with non-negative shifting plus 0.01 for ${\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}$ values.

(2)

Calculation of Euclidean Distance

Combining weights (${\mathsf{\Omega }}_{\mathrm{j}}$) and standardized data (${\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}$) to obtain the weighted normalization matrix (${\left({\mathrm{N}}_{\mathrm{i}\mathrm{j}}\right)}_{\mathrm{n}\times \mathrm{m}}$), the positive and negative ideal solutions (${\mathrm{N}}_{\mathrm{j}}^{+}$, ${\mathrm{N}}_{\mathrm{j}}^{-}$) of the matrix were determined, and the Euclidean distances (${\mathrm{g}}_{\mathrm{i}}^{+}$, ${\mathrm{g}}_{\mathrm{i}}^{-}$) were calculated from the evaluated object to the positive and negative ideal solutions:

$${\mathrm{N}}_{\mathrm{i}\mathrm{j}}={\mathrm{X}\prime }_{\mathrm{i}\mathrm{j}}\times {\mathsf{\Omega }}_{\mathrm{j}}$$

(5)

$${\mathrm{N}}_{\mathrm{j}}^{+}=\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{N}}_{\mathrm{i}1},\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{N}}_{\mathrm{i}2},\cdots ,\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{N}}_{\mathrm{i}\mathrm{m}}$$

(6)

$${\mathrm{N}}_{\mathrm{j}}^{-}=\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{N}}_{\mathrm{i}1},\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{N}}_{\mathrm{i}2},\cdots ,\mathrm{m}\mathrm{i}\mathrm{n}{\mathrm{N}}_{\mathrm{i}\mathrm{m}}$$

(7)

$${\mathrm{g}}_{\mathrm{i}}^{+}=\sqrt{{\sum }_{\mathrm{j}=1}^{\mathrm{m}}{\left({\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{+}\right)}^2},{\mathrm{g}}_{\mathrm{i}}^{-}=\sqrt{{\sum }_{\mathrm{j}=1}^{\mathrm{m}}{\left({\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{-}\right)}^2}$$

(8)

Therein, the smaller the ${\mathrm{g}}_{\mathrm{i}}^{+}$ and the greater the ${\mathrm{g}}_{\mathrm{i}}^{-}$, the greater the development level of the evaluated object, and vice versa, the lower the level.

(3)

Calculation of Grey Correlation

Once the grey correlation coefficients (${\xi }_{\mathrm{i}\mathrm{j}}^{+}$, ${\xi }_{\mathrm{i}\mathrm{j}}^{-}$) of each index value were calculated based on the positive and negative ideal solutions according to the weighted normalization matrix, the grey correlations (${\mathrm{r}}_{\mathrm{i}}^{+}$, ${\mathrm{r}}_{\mathrm{i}}^{-}$) of the evaluated object were calculated:

$${\xi }_{\mathrm{i}\mathrm{j}}^{+}={\displaystyle \frac{\underset{\mathrm{i}}{\mathrm{m}\mathrm{i}\mathrm{n}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{i}\mathrm{n}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{+}\right|+\mathsf{\rho }\underset{\mathrm{i}}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{+}\right|}{\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{+}\right|+\mathsf{\rho }\underset{\mathrm{i}}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{+}\right|}}$$

(9)

$${\xi }_{\mathrm{i}\mathrm{j}}^{-}={\displaystyle \frac{\underset{\mathrm{i}}{\mathrm{m}\mathrm{i}\mathrm{n}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{i}\mathrm{n}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{-}\right|+\mathsf{\rho }\underset{\mathrm{i}}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{-}\right|}{\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{-}\right|+\mathsf{\rho }\underset{\mathrm{i}}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{\mathrm{j}}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{\mathrm{N}}_{\mathrm{i}\mathrm{j}}-{\mathrm{N}}_{\mathrm{j}}^{-}\right|}}$$

(10)

$${\mathrm{r}}_{\mathrm{i}}^{+}={\displaystyle \frac{1}{\mathrm{m}}}\sum _{\mathrm{j}=1}^{\mathrm{m}}{\xi }_{\mathrm{i}\mathrm{j}}^{+},{\mathrm{r}}_{\mathrm{i}}^{-}={\displaystyle \frac{1}{\mathrm{m}}}\sum _{\mathrm{j}=1}^{\mathrm{m}}{\xi }_{\mathrm{i}\mathrm{j}}^{-}$$

(11)

Here, $\mathsf{\rho }$ expresses the resolution coefficient, typically set to 0.5. The larger the ${\mathrm{r}}_{\mathrm{i}}^{+}$ and the smaller the ${\mathrm{r}}_{\mathrm{i}}^{-}$, the higher the development level of the evaluated object, and vice versa, the lower it is.

(4)

Calculation of Comprehensive Closeness

Following the dimensionless processing of ${\mathrm{g}}_{\mathrm{i}}^{+}$, ${\mathrm{g}}_{\mathrm{i}}^{-}$ and ${\mathrm{r}}_{\mathrm{i}}^{+}$, ${\mathrm{r}}_{\mathrm{i}}^{-}$, respectively, they were combined to calculate the comprehensive closeness (${\mathrm{Z}}_{\mathrm{i}}$) of the evaluated object:

$${\mathrm{G}}_{\mathrm{i}}^{+}={\displaystyle \frac{{\mathrm{g}}_{\mathrm{i}}^{+}}{\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{g}}_{\mathrm{i}}^{+}}},{\mathrm{G}}_{\mathrm{i}}^{-}={\displaystyle \frac{{\mathrm{g}}_{\mathrm{i}}^{-}}{\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{g}}_{\mathrm{i}}^{-}}}$$

(12)

$${\mathrm{R}}_{\mathrm{i}}^{+}={\displaystyle \frac{{\mathrm{r}}_{\mathrm{i}}^{+}}{\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{r}}_{\mathrm{i}}^{+}}},{\mathrm{R}}_{\mathrm{i}}^{-}={\displaystyle \frac{{\mathrm{r}}_{\mathrm{i}}^{-}}{\mathrm{m}\mathrm{a}\mathrm{x}{\mathrm{r}}_{\mathrm{i}}^{-}}}$$

(13)

$${\mathrm{Z}}_{\mathrm{i}}^{+}=0.5{\mathrm{G}}_{\mathrm{i}}^{-}+0.5{\mathrm{R}}^{+},{\mathrm{Z}}_{\mathrm{i}}^{-}=0.5{\mathrm{G}}_{\mathrm{i}}^{+}+0.5{\mathrm{R}}^{-}$$

(14)

$${\mathrm{Z}}_{\mathrm{i}}={\displaystyle \frac{{\mathrm{Z}}_{\mathrm{i}}^{+}}{{\mathrm{Z}}_{\mathrm{i}}^{+}+{\mathrm{Z}}_{\mathrm{i}}^{-}}}$$

(15)

Of which, ${\mathrm{G}}_{\mathrm{i}}^{+}$, ${\mathrm{G}}_{\mathrm{i}}^{-}$ and ${\mathrm{R}}_{\mathrm{i}}^{+}$, ${\mathrm{R}}_{\mathrm{i}}^{-}$ denote the Euclidean distance and grey correlation degree after dimensionless processing, ${\mathrm{Z}}_{\mathrm{i}}^{+}$ and ${\mathrm{Z}}_{\mathrm{i}}^{-}$ represent the positive and negative closeness of the evaluated object to the ideal value, and ${\mathrm{Z}}_{\mathrm{i}}$ shows the overall closeness, which also reflects the systematic comprehensive development level of the evaluated object, with the larger value being superior.

4.1.2. Benchmark Regression Model

The benchmark regression model provides a fundamental framework for analyzing the influence of the explanatory variable on the explained variable, helping to quantify the specific extent and direction of the impact. To probe the direct influence of logistics industrial transformation and upgrading affected by digital-real integration and verify whether hypothesis H1 is valid, the benchmark regression model was constructed as follows:

$${\mathrm{L}\mathrm{o}\mathrm{g}}_{\mathrm{i}\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\alpha }}_2{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\delta }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(16)

where ${\mathrm{L}\mathrm{o}\mathrm{g}}_{\mathrm{i}\mathrm{t}}$ is the explained variable representing the development situation of the logistics industry’s transformation and upgrading of region $\mathrm{i}$ in period $\mathrm{t}$, ${\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}$ denotes the explanatory variable to measure the development situation of digital-real integration of region $\mathrm{i}$ in period $\mathrm{t}$, and ${\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i}\mathrm{t}}$ shows a series of controlled variables: technology R&D capacity (Tech), openness to the outside world (Fdi), industrial structure (Indus), financial development scale (Fsc), and government fiscal expenditure (Gov) in year $\mathrm{t}$ of region $\mathrm{i}$. ${\mathsf{\mu }}_{\mathrm{i}}$ means the individual-fixed effects that do not depend on time, ${\mathsf{\delta }}_{\mathrm{t}}$ means the time-fixed effect, and on the side, ${\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$ expresses the random perturbation term.

4.1.3. Moderation Effect Model

The moderation effect model is frequently used to explore the interaction mechanism among variables, revealing how a moderator variable affects the direction and strength of the relationship between the explanatory variable and the explained variable. For testing whether there is an assertion, as in hypothesis H2, the green economy (Ged) was used as the moderating variable, and regarding the study of Jiang [55], the interaction term between digital-real integration and green economy after decentralization was added to the model to examine the moderating influence of the green economy. The following outlines the construction of the moderation effect model:

$${\mathrm{L}\mathrm{o}\mathrm{g}}_{\mathrm{i}\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_2{\mathrm{G}\mathrm{e}\mathrm{d}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_3{\mathrm{D}\mathrm{r}\mathrm{f}\ast \mathrm{G}\mathrm{e}\mathrm{d}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_4{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\delta }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(17)

Among them, ${\mathrm{D}\mathrm{r}\mathrm{f}\ast \mathrm{G}\mathrm{e}\mathrm{d}}_{\mathrm{i}\mathrm{t}}$ is the interaction term between digital-real integration and green economy after decentralization, and its regression coefficient, ${\mathsf{\beta }}_4$, is used to judge whether the moderating effect exists or not. If ${\mathsf{\beta }}_4$ is dramatically positive, the green economy positively moderates the influence of the logistics industry’s transformation and upgrading affected by digital-real integration; if ${\mathsf{\beta }}_4$ is significantly negative, the green economy has a negative moderating effect.

4.1.4. Threshold Effect Model

The threshold effect model can keenly capture how the change direction and intensity of the explained variable will alter abruptly when the threshold variable reaches a certain threshold, which is suitable for testing the nonlinear relationship between the variables. To study whether digital-real integration and green economy had nonlinear characteristics and threshold conditions, the model was built according to the introduction of the panel threshold model by Hansen and Che [56,57] to verify hypothesis H3. The threshold effect model was constructed as follows:

$${\mathrm{L}\mathrm{o}\mathrm{g}}_{\mathrm{i}\mathrm{t}}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}·1({\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}\le \mathsf{\lambda })+{\mathsf{\gamma }}_2{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}·1({\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}>\mathsf{\lambda })+{\mathsf{\gamma }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(18)

$${\mathrm{L}\mathrm{o}\mathrm{g}}_{\mathrm{i}\mathrm{t}}={\mathsf{\phi }}_0+{\mathsf{\phi }}_1{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}·1({\mathrm{G}\mathrm{e}\mathrm{d}}_{\mathrm{i}\mathrm{t}}\le \mathsf{\vartheta })+{\mathsf{\phi }}_2{\mathrm{D}\mathrm{r}\mathrm{f}}_{\mathrm{i}\mathrm{t}}·1({\mathrm{G}\mathrm{e}\mathrm{d}}_{\mathrm{i}\mathrm{t}}>\mathsf{\vartheta })+{\mathsf{\phi }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(19)

Therein, 1(·) is the indicator function, with the value of 1 if the bracketed condition is satisfied and 0 otherwise. $\mathsf{\lambda }$ and $\mathsf{\vartheta }$ are the thresholds that allow the sample to be divided into different intervals for testing.

4.2. Variable Definitions

4.2.1. Explained Variable

Logistics industrial transformation and upgrading was the explained variable. We referred to existing studies to construct a measurement index system (

Table 1. Measurement index system of logistics industrial transformation and upgrading.

System Layer	Dimension Layer	Index Layer	Attribute	Unit
Logistics Industrial Transformation and Upgrading (L)	Development Condition (L1)	Road Mileage (L11)	+	km
		Percentage of Fiscal Expenditure on Transportation (L12)	+	%
	Industrial Scale (L2)	Gross Regional Product—Transportation, Storage, and Postal Services (L21)	+	100 million yuan
		Logistics Value Added as Share of GDP (L22)	+	%
		Investment in Fixed Assets for Transportation, Storage, and Postal Services (L23)	+	10,000 yuan
	Supply Quality (L3)	Freight Turnover (L31)	+	100 million tons-km
		Freight Volume (L32)	+	10,000 tons
		Total Imports of Goods (L33)	+	10,000 dollars
		Total Exports of Goods (L34)	+	10,000 dollars
	Green Environmental Protection (L4)	Electricity Consumption in Transportation, Storage, and Postal Services (L41)	−	kWh/10,000 yuan
		Sulfur Dioxide Emissions from Logistics Industrial Output Value Per 10,000 Yuan (L42)	−	ton

Note: “+” indicates the positive indicator; “−” indicates the negative indicator.

) of logistics industrial transformation and upgrading from four dimensions: development condition, industrial scale, supply quality, and green environmental protection [1,12]. The comprehensive closeness was calculated based on the entropy weight–TOPSIS–grey correlation model for reflecting the development level of logistics industrial transformation and upgrading, which was recorded as Log.

4.2.2. Explanatory Variable

Digital-real integration was the explanatory variable. Drawing from existing literature and considering data accessibility, this study combined the advancement situation of the integration of digital and real to construct a measurement index system (

Table 2. Measurement index system of digital-real integration.

System Layer	Dimension Layer	Index Layer	Attribute	Unit
Digital-Real Integration (D)	Integration Foundation (D1)	Internet Broadband Access Users (D11)	+	10,000 households
		Mobile Phone Penetration Rate (D12)	+	phone/100 people
		Percentage of Persons Employed in the Information Technology Industry (D13)	+	%
	Integration Input (D2)	Local Fiscal Expenditure on Science and Technology (D21)	+	10,000 yuan
		Expenditures on New Product Development of Industrial Enterprises Above Scale (D22)	+	10,000 yuan
		Gross Power of Agricultural Machinery (D23)	+	10,000 kW
		Fixed-Asset Investment Growth in Information Technology Industry (D24)	+	%
	Integration Utility (D3)	Sales Revenue on New Products of Industrial Enterprises Above Scale (D31)	+	10,000 yuan
		Patents Granted Per 10,000 People (D32)	+	item
		Profit-to-Cost Ratio of Industrial Enterprises Above Scale (D33)	+	%
		Electricity Consumption Per 10,000 GDP (D34)	−	kWh

Note: “+” indicates the positive indicator; “−” indicates the negative indicator.

) from three dimensions: integration foundation, integration input, and integration utility [4,23]. Furthermore, this study employed the entropy weight–TOPSIS–grey correlation model to survey the comprehensive closeness to reflect the level of digital-real integration, which was denoted as Drf.

4.2.3. Moderator Variable

The green economy was the moderator variable. Based on existing research and considering data availability, this paper built the measurement index system (

Table 3. Measurement index system of green economy.

System Layer	Dimension Layer	Index Layer	Attribute	Unit
Green Economy (G)	Economic Support (G1)	Per Capita GDP (G11)	+	yuan/person
		The Ratio of Retail Sales of Consumer Goods to Total Population (G12)	+	%
		Total Labor Productivity (G13)	+	yuan/person
	Innovation Elements (G2)	R&D Spending as a Scale of GDP (G21)	+	%
		Percentage of Education Employees (G22)	+	%
		Telecommunication Services Revenue (G23)	+	10,000 yuan
		Green Invention Patent Application (G24)	+	item
	Green Development (G3)	Greening Coverage in Built-Up Areas (G31)	+	%
		Green Invention Patent Authorization (G32)	+	item
		Percentage of Expenditure on Environmental Protection (G33)	+	%
		Harmless Treatment Rate of Life Waste (G34)	+	%

Note: “+” indicates the positive indicator.

) from three dimensions: economic support, innovation elements, and green development [6,44]. Meanwhile, the comprehensive closeness was obtained through the entropy weight–TOPSIS–grey correlation model to reflect the development level of the green economy, which was recorded as Ged.

4.2.4. Controlled Variable

A set of controlled variables was introduced to more accurately test the mechanism of logistics industrial transformation and upgrading affected by digital-real integration, including technology R&D capacity (Tech): the logarithm of the full-time equivalent of R&D personnel in industrial enterprises above scale expresses technological R&D capability. Openness to the outside world (Fdi): the logarithm of foreign direct investment amount captures openness to the outside world. Industrial structure (Indus): the logarithmic share of tertiary industrial added value in GDP shows the industrial structure. Financial development scale (Fsc): the logarithm of deposit and loan balances of financial institutions as a proportion of GDP characterizes the financial development scale. Government fiscal expenditure (Gov): the logarithmic proportion of general public budget expenditure to GDP indicates the government fiscal expenditure.

4.3. Data Sources and Descriptive Statistics

4.3.1. Data Sources

This study took the panel data spanning from 2013 to 2022 of 11 cities in Hebei Province of China as an example to verify the research hypotheses. The data sources mainly included the China Urban Statistical Yearbook, China Regional Economic Statistical Yearbook, China Urban Construction Statistical Yearbook, China Statistical Yearbook, Hebei Statistical Yearbook, Hebei Economic Yearbook, China Research Data Service Platform (CNRDS), and provincial (municipal) statistical bulletins, etc., using the interpolation method to fill in individual missing values.

4.3.2. Descriptive Statistics

The final dataset of panel data used in this paper contained 110 valid observations from 11 cities in Hebei Province, China, spanning from 2013 to 2022.

Table 4. Descriptive statistics.

Variable Type	Variable	Obs.	Mean	Std. Dev.	Min.	Max.	VIF
Explained Variable	Log	110	0.269	0.023	0.241	0.335	—
Explanatory Variable	Drf	110	0.300	0.025	0.263	0.389	4.09
Moderator Variable	Ged	110	0.335	0.025	0.298	0.413	5.91
Controlled Variable	Tech	110	8.534	0.894	6.547	10.078	3.64
	Fdi	110	10.991	0.894	7.870	12.252	2.58
	Indus	110	3.807	0.172	3.451	4.132	3.43
	Fsc	110	5.730	0.293	5.044	6.292	6.26
	Gov	110	2.934	0.330	2.095	3.694	5.01

provides a comprehensive statistical overview of these variables, including sample sizes, means, standard deviations, minimum values, and maximum values. The variance inflation factor (VIF) was calculated for determining the presence of multicollinearity among the variables. The VIF values of all variables were less than 10 (VIF mean was 4.42), demonstrating that there was no multicollinearity among the variables.

5. Results

Hebei Province has agriculture, industry, and services as its main industrial pillars, with the service sector experiencing rapid growth in recent years, particularly the commercial logistics industry, which has emerged as a vital engine of economic growth. By 2023, the added value of the commercial logistics industry reached 679.4 billion yuan, accounting for over 15% of the GDP, demonstrating a robust development momentum. As a significant economic province in China, Hebei boasts a solid foundation and immense potential in its logistics industry, yet it faces an urgent challenge of transitioning from high pollution and high consumption to green and low carbon. Furthermore, the rise of e-commerce, new retail, and other modes has also put forward higher requirements on the timeliness and intelligence of logistics services. Therefore, taking Hebei Province as the research object, this paper quantitatively analyzed its development level and explored the impact of digital-real integration on the transformation and upgrading of the logistics industry under the green economy, which is aimed at providing a useful reference for the future development of the logistics industry in Hebei Province and even the world under the background of green economy and digitalization.

5.1. Comprehensive Development Level

Utilizing panel data pertaining to Hebei Province, this paper determined the index weights of each systematic measurement index system via the entropy weight method and combined the TOPSIS method and grey correlation method to derive the comprehensive closeness of these systems, which was utilized to analyze the overall developmental status of logistics industrial transformation and upgrading, digital-real integration, and green economy in the period from 2013 to 2022.

5.1.1. Level of Logistics Industrial Transformation and Upgrading

Figure 1 shows the change trend in the development level according to the comprehensive closeness of the logistics industry’s transformation and upgrading. Overall, the progression of logistics industrial transformation and upgrading in Hebei Province had certain fluctuations, with a development level of 0.6518 in 2013, followed by year-on-year growth, and then a significant drop in 2015, which might be due to a combination of factors, such as industrial structure adjustment and transformation pressures, imperfections in infrastructure construction and logistics network, changes in policy and the market environment, shortages of talent and technology, and lagging regional economic integration processes. However, after a brief decline, it began to rebound and reached 0.7376 by 2022. This might be attributed to the promotion of the regional coordinated development strategy, resulting in the increasing improvement of logistics networks among cities and the gradual formation of new models of resource sharing and market interoperability, which has promoted the swift expansion of the logistics sector. Among them, Shijiazhuang, as a provincial capital, maintained a higher level of logistics industrial transformation and upgrading in most years, with the comprehensive closeness increasing from 0.2732 in 2013 to 0.3290 in 2022 at a relatively stable growth trend and growth rate of around 20.41%, which showed the leading role of the capital city in the transformation and upgrading of the logistics industry. Comprehensively, the general trajectory of logistics industry transformation and advancement was good in all cities of Hebei Province during this decade, but there were differences in the growth rate and fluctuation range, which were lower than the average level of Hebei Province, and further efforts are still needed to fit in economic development demands and continue to maintain a good trend.

5.1.2. Level of Digital-Real Integration

Figure 2 presents a trend chart of the development level based on the comprehensive closeness of digital-real integration. Over the selected period, the overall development level of digital-real integration in Hebei Province showed an accelerated growth trend, rising from 0.4770 in 2013 to 0.6870 in 2022, with an increase of about 44.03%, indicating that Hebei Province’s attention to the progression of digital-real fusion has continued to rise, and improvements, such as policy support and innovation capabilities, have driven the ever-increasing integration of digital technology with the real economy in the region. Shijiazhuang’s integration development level was leading in Hebei Province, generally higher than that of other regions, reaching 0.3891 by 2022, as well as growing at a faster rate, by about 27.21% from 2013 to 2022. This revealed Shijiazhuang’s strong momentum in the field of digital-real integration, and the obvious advantages in the policy environment, geographical location, technological resources, talent reserves, etc., have provided strong support for the in-depth digital-real integration. Overall, cities in Hebei Province have shown an upward trend in the development of digital-real integration, but there are differences in the internal development levels. Some cities, such as Shijiazhuang and Tangshan, have exhibited faster growth in recent years, while others have seen relatively slower growth and may need to further strengthen investment in the digital-real integration. With the further expansion of 5G, big data, artificial intelligence, and other technologies, the progression of digital-real integration in Hebei Province is expected to enter a new stage.

5.1.3. Level of Green Economy

Figure 3 presents the trend of the development level regarding the comprehensive closeness of green economy. From 2013 to 2022, the level of the green economy in Hebei Province as a whole region steadily rose, increasing from the initial 0.4881 to 0.6690, with a growth rate around 37.05%, indicating that the region’s green economy development was constantly increasing. Especially in 2021, there was a significant jump from 0.5947 in 2020 to 0.6531, which was closely related to the introduction of relevant policies and initiatives, the ongoing evolution and utilization of emerging industries and technologies, and the gathering of innovative talents. The level of green economy in Shijiazhuang was higher than that of other regions, and the growth trend was remarkable, from 0.3424 in 2013 to 0.4130 in 2022, showing the powerful development of Shijiazhuang in green economy, which might be because Shijiazhuang, as the capital city of the province, has policy inclination, resource concentration, and talent attraction, thus enhancing the level of its green economy. To summarize, various cities in Hebei Province have displayed a certain synergistic development trend in green economy, among which Shijiazhuang has played an important role as a leader, while other regions are also constantly improving their green economy level and promoting regional coordinated development.

5.2. Benchmark Regression Analysis

5.2.1. Model Estimation

Table 5. Benchmark regression results.

Variable	Model (1)	Model (2)	Model (3)
Drf	0.2854 **	0.3250 **	0.2988 **
	(2.42)	(2.04)	(2.16)
Tech	−0.0072	−0.0068	−0.0077
	(−1.16)	(−1.20)	(−1.17)
Fdi	0.0050 *	0.0073 *	0.0061 **
	(1.91)	(1.95)	(2.00)
Indus	0.0273 *	0.0347 **	0.0357
	(1.83)	(1.96)	(1.47)
Fsc	−0.0131	−0.0154	−0.0146
	(−1.30)	(−1.27)	(−1.16)
Gov	−0.0004	−0.0147	−0.0041
	(−0.04)	(−1.15)	(−0.30)
_cons	0.1881 ***	0.1485	0.1632
	(4.13)	(1.55)	(1.49)
id	Yes	No	Yes
year	No	Yes	Yes
R2	0.4448	0.4695	0.4851
N	110	110	110

Note: z statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

displays the model regression results of digital-real integration directly affecting logistics industrial transformation and upgrading, in which the model (1) was a regression for individual fixation, and the coefficient of Drf was positive and passed the 5% significance test, whereas model (2), which employed a time-fixed effect model, exhibited that the coefficient of Drf was improved by 0.0396 compared to that of model (1) and it was significant. Using the time-individual fixed regression results of model (3) as a baseline, the regression coefficient of Drf fluctuated compared to models (1) and (2), but its coefficient value remained notably positive at the significance level of 5%. With every 1-unit increment in digital-real integration, the transformation and upgrading of the logistics industry will rise by 0.2988 units, indicating that the regression results of digital-real fusion for the transformation and upgrading of the logistics industry were still stable and significant even after considering the effects of time trends and regional differences, which intuitively reflects that digital-real integration, as an important influencing factor on logistics industrial transformation and upgrading, had an active direct effect on it. From all models, the consistency in sign direction of the regression coefficient related to digital-real integration, coupled with its significant validation at the 5% level, indicated the substantial role of enhanced digital-real integration in advancing the transformation and upgrading of the logistics industry. This discovery validated the viewpoint in hypothesis H1.

5.2.2. Robustness Test

The above regression results point to the fact that digital-real integration dramatically fosters logistics industrial transformation and upgrading, and next, a series of regression tests were used to intensify the reliability of the conclusion. According to

Table 6. Robustness test results of benchmark regression.

Variable	Model (4)	Model (5)	Model (6)	Model (7)	Model (8)
Drf		0.2690 *		0.2801 **	0.2882 **
		(1.94)		(2.10)	(2.00)
Drf_0	0.3947 ***
	(2.62)
L.Drf			0.3078 ***
			(2.98)
Urb		−0.0681 **
		(−2.49)
Tech	−0.0066	−0.0091	−0.0065	−0.0080	−0.0078
	(−1.16)	(−1.38)	(−1.06)	(−1.26)	(−1.13)
Fdi	0.0053 *	0.0057 *	0.0062 **	0.0059 **	0.0062 *
	(1.96)	(1.92)	(2.23)	(2.33)	(1.92)
Indus	0.0298	0.0441 *	0.0257	0.0341	0.0381
	(1.41)	(1.71)	(1.02)	(1.31)	(1.53)
Fsc	−0.0139	−0.0133	0.0009	−0.0051	−0.0149
	(−1.21)	(−0.97)	(0.07)	(−0.28)	(−1.19)
Gov	−0.0025	−0.0077	−0.0058	−0.0080	−0.0040
	(−0.19)	(−0.57)	(−0.50)	(−0.56)	(−0.28)
_cons	0.1204	0.4323 **	0.1047	0.1344	0.1597
	(1.50)	(2.41)	(0.92)	(1.22)	(1.39)
Id and year	Yes	Yes	Yes	Yes	Yes
R2	0.5134	0.5113	0.4998	0.4837	0.4762
N	110	110	99	88	110

Note: z statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

, model (4) changed the calculation method of the explanatory variable, following the measurement index system of this paper, and re-derived the digital-real integration variable through the entropy weight–grey correlation model, recorded as Drf_0. Further, benchmark regression was conducted again, and the result showed that digital-real integration on logistics industrial transformation and upgrading was also a marked positive correlation. Model (5) added the controlled variable by introducing the logarithmic end-of-year urban population share to judge the urbanization level, called Urb, to exclude bias caused by the randomness of controlled variable selection. The findings also indicated that the coefficient of Drf was positive and striking, which means that digital-real integration still had a pilot effect on logistics industrial transformation and upgrading. To avoid the endogeneity problem caused by reverse causation as much as possible, Model (6) was re-modeled using the lagged one-period digital-real integration, and it was found that the correspondence of the benchmark regression above still existed, which states that digital-real integration was the main cause in the causality. Model (7), which intercepted the sample data from 2014 to 2021 to reduce the subjective error, yielded an outcome that was consistent with model (3). Furthermore, the regression coefficients of Drf went through the significance test at the 1% level and were positive. Model (8) was a 1% bilateral shrinkage of each variable to prevent adverse effects caused by outliers, and the test results showed that the coefficient of digital-real fusion remained notable. All conducted tests confirmed the reliability of the above findings, reaffirming the conclusion that digital-real integration effectively fostered the logistics industry’s transformation and upgrading.

5.3. Moderation Effect Analysis

5.3.1. Model Estimation

The results of the benchmark regression and the robustness test showed that digital-real integration had an energetic effect on logistics industrial transformation and upgrading. Aiming to further analyze the influencing mechanism of digital-real integration on logistics industrial transformation and upgrading under the green economy, this paper added the interaction term (Drf*Ged) to the regression model to test whether the green economy had an impact on the logistics industry’s transformation and upgrading promoted by digital-real integration. As can be seen from the results of model (10) in

Table 7. Moderation effect regression results.

Variable	Model (9)	Model (10)
Drf	0.2272 **	0.1516
	(2.04)	(1.45)
Ged	0.2555	0.0569
	(1.33)	(0.29)
Drf*Ged		3.8317 ***
		(2.78)
Tech	−0.0084	−0.0029
	(−1.31)	(−0.41)
Fdi	0.0050 *	0.0052 *
	(1.66)	(1.65)
Indus	0.0289	0.0291
	(1.10)	(1.11)
Fsc	−0.0071	−0.0044
	(−0.48)	(−0.31)
Gov	−0.0135	−0.0138
	(−0.96)	(−1.14)
_cons	0.1199	0.1390
	(0.95)	(1.02)
Id and year	Yes	Yes
R2	0.5107	0.5495
N	110	110

Note: z statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

, at the 1% level of significance, the regression coefficient of the interaction term between green economy and digital-real integration was 3.8317, which indicated that the green economy did play a positive regulatory effect between digital-real integration and logistics industrial transformation and upgrading. The existence of a positive moderating effect also means that the positive correlation between digital-real integration and logistics industrial transformation and upgrading was enhanced with the green economic growth. That is to say, when the degree of green economy is superior, the promotion role of digital-real integration on logistics industrial transformation and upgrading will be more notable. The reason for this may be that the introduction of the green economy makes the logistics industry more able to adapt to and take advantage of the trend of digital-real integration, thus boosting the efficiency, innovation, and competitiveness of the logistics industry, and accelerating its transformation and upgrading. Thus, hypothesis H2 was validated.

5.3.2. Robustness Test

The moderation effect regression results verified the positive moderating role of the green economy in digital-real integration for logistics industrial transformation and upgrading. To ensure the solidity and dependability of the conclusion, the green economy was divided into three sub-samples of economic support (Eco), innovation elements (Innov), and green development (Green), according to the dimensional layer of the systematic measurement index system, to carry out the moderation effect regression again. From

Table 8. Robustness test results of the moderation effect.

Variable	Economic Support		Innovative Elements		Green Development
	Model (11)	Model (12)	Model (13)	Model (14)	Model (15)	Model (16)
Drf	0.2967 **	0.2540 *	0.2305 **	0.1293	0.2493 *	0.2204 *
	(2.38)	(1.96)	(2.56)	(1.46)	(1.88)	(1.73)
Eco	−0.0753 *	−0.1030 *
	(−1.68)	(−1.91)
Drf*Eco		0.6659
		(1.54)
Innov			0.2225	−0.1064
			(1.28)	(−0.48)
Drf*Innov				5.6139 ***
				(2.87)
Green					0.1363	0.0960
					(1.54)	(1.13)
Drf*Green						1.9166 ***
						(2.59)
Tech	−0.0067	−0.0050	−0.0069	0.0000	−0.0089	−0.0068
	(−1.10)	(−0.81)	(−1.19)	(0.00)	(−1.42)	(−1.04)
Fdi	0.0060 **	0.0061 **	0.0054 **	0.0042	0.0051 *	0.0052 *
	(2.19)	(2.09)	(2.04)	(1.58)	(1.65)	(1.69)
Indus	0.0433 *	0.0494 **	0.0366	0.0163	0.0318	0.0299
	(1.86)	(2.08)	(1.47)	(0.69)	(1.29)	(1.26)
Fsc	−0.0266 **	−0.0197	−0.0126	−0.0204	−0.0128	−0.0119
	(−2.44)	(−1.56)	(−1.00)	(−1.56)	(−0.91)	(−0.86)
Gov	−0.0037	−0.0086	−0.0105	−0.0057	−0.0131	−0.0131
	(−0.27)	(−0.61)	(−0.80)	(−0.50)	(−0.91)	(−0.98)
_cons	0.2242 ** (2.34)	0.1806 * (1.95)	0.1154 (0.94)	0.2960 * (1.83)	0.1806 (1.43)	0.1828 (1.47)
Id and year	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.4989	0.5139	0.5126	0.5851	0.5106	0.5257
N	110	110	110	110	110	110

Note: z statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

, it is clear that the test outcomes of the three sets of data presented a certain gap. From the perspective of economic support, the coefficient of the interaction term between economic support and digital-real integration in model (12) was positive but not significant, which may be related to the unstable policy environment and the uncertainty of market competition patterns and consumer demand in different regions. In terms of innovation elements, the coefficient of the interaction term between innovation elements and digital-real integration in model (14) appeared to be 5.6139, which passed the 1% level significance test. Namely, innovation elements had a forward moderating effect, and the optimization of innovation elements could strengthen the facilitating role of digital-real integration on logistics industrial transformation and upgrading. Meanwhile, for green development, the coefficient of the interaction term between green development and digital-real integration in model (16) was striking at 1.9166, which indicated that green development exhibited an active regulatory effect, and the intensity of its regulating effect was less than that of innovation elements. This means that among the three dimensions of the green economy, the innovation elements were more likely to affect the function of digital-real integration on logistics industrial transformation and upgrading, which overall also confirmed the reliability of the moderation effect of the green economy.

5.4. Threshold Effect Analysis

5.4.1. Model Estimation

Based on passing the above tests, digital-real integration and green economy were set as threshold variables, and the single threshold and multiple threshold tests were carried out by using the measurement software Stata 17.0 to repeat sampling 300 times, respectively, to derive the F statistic and its corresponding p-value, then to unpack the nonlinear influence of digital-real integration on the transformation and upgrading of the logistics industry. The findings of the threshold effect test in

Table 9. Threshold effect test results.

Threshold Variable	Threshold Number	F	p	Critical Value			Threshold Value	Confidence Interval of 95%
				10%	5%	1%
Drf	Single	37.41 ***	0.0033	14.0481	18.4518	27.7001	0.3541	(0.3483, 0.3554)
	Double	11.40	0.1233	12.1217	13.3977	16.9845	0.3181	(0.3140, 0.3183)
Ged	Single	30.60 **	0.0167	19.3161	22.2026	32.8191	0.3820	(0.3756, 0.3821)
	Double	−15.30	1.0000	29.3390	38.3409	64.0972	0.3827	(0.3756, 0.3852)

Note: ** p < 0.05, and *** p < 0.01.

revealed that digital-real integration passed the single threshold test at the 1% significance level, while the multiple threshold tests failed, and the corresponding threshold value of digital-real integration was 0.3541 according to the single threshold test results. Also, the F-statistic of the single threshold test for green economy was notable at the significance level of 5%, but it did not pass the multiple threshold test and, depending on the results of the single threshold test, the correlative threshold value of green economy was 0.3820. The above test preliminarily verified hypothesis H3, namely, the influence of digital-real integration on logistics industrial transformation and upgrading had the thresholds of digital-real integration and green economy.

Against Table 9, the likelihood ratio (LR) function plot of threshold estimate values at the 95% confidence interval was plotted. In Figure 4, for (a), the likelihood ratio function is graphed at the 95% confidence interval for the threshold value of digital-real integration (0.3541), and (b) is a LR plot corresponding to the threshold of the green economy (0.3820). The dotted line in the figure represents the critical value, which was 7.35, the lowest point of the LR statistic is the relevant real threshold value, and below the dotted lines refer to the threshold interval correspondent to the critical value of 7.35 at the LR statistic less than the 5% significant level. Since the critical value of 7.35 was significantly higher than the threshold values, marked by the lowest points of the solid lines, it can be determined that the aforementioned threshold values were real and valid.

With digital-real integration and green economy as the threshold variables,

Table 10. Threshold effect regression results.

Threshold Variable: Drf		Threshold Variable: Ged
Variable	Model (17)	Variable	Model (18)
Drf·1 (Drf ≤ 0.3541)	0.1101	Drf·1 (Ged ≤ 0.3820)	0.1297
	(1.25)		(1.41)
Drf·1 (Drf > 0.3541)	0.1814 **	Drf·1 (Ged > 0.3820)	0.1772 *
	(2.33)		(1.95)
Tech	−0.0007	Tech	−0.0026
	(−0.16)		(−0.58)
Fdi	0.0045 *	Fdi	0.0049 *
	(1.82)		(1.88)
Indus	0.0260 **	Indus	0.0306 **
	(2.26)		(2.76)
Fsc	−0.0132	Fsc	−0.0162
	(−1.42)		(−1.72)
Gov	−0.0012	Gov	−0.0008
	(−0.12)		(−0.10)
_cons	0.1715 ***	_cons	0.1757 ***
	(4.73)		(4.51)
R2	0.5960	R2	0.5255
N	110	N	110

Note: t statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

shows the regression results of the threshold effect under the condition of controlling the variables of technology R&D capacity, openness to the outside world, industrial structure, financial development scale, and government fiscal expenditure. It can be observed that the influence of digital-real integration on the transformation and upgrading of the logistics industry featured a nonlinear characteristic; that is, there existed threshold conditions of the green economy and digital-real integration. Under these conditions, hypothesis H3 was validated.

In Table 10, when the threshold variable was digital-real integration in model (17), its threshold value divided digital-real integration into two intervals, in which interval one was Drf ≤ 0.3541 and interval two was Drf > 0.3541. Specifically, when the value of digital-real integration was in the first interval, the regression coefficient of digital-real integration on logistics industrial transformation and upgrading was 0.1101, but it was not significant. When the digital-real integration value was located in interval two, the regression coefficient stood at 0.1814, exhibiting significant importance at the 5% level, which suggested a favorable contribution of digital-real integration in this interval to advancing logistics industrial transformation and upgrading. The reason for presenting the above threshold effect may be that in the early stage of digital-real integration, enterprises face the problems of insufficient technological accumulation and insignificant scale effect, which may lead to the limited application scope of digital technology in the field of logistics and fail to give full play to its potential to improve the logistics efficiency and reduce costs. However, with the role of technological accumulation and the scale effect gradually appearing, for digital-real integration to reach a high level, logistics enterprises can make full use of digital technology to form their competitive advantage, thereby driving logistics industrial transformation and upgrading.

The dynamic impact of digital-real integration on the logistics industry’s transformation and upgrading is not only restricted by its level but also may be limited by other regulatory variables. Model (18) in Table 10 took green economy as the threshold variable for regression, and the results expressed that when green economy was at a low level (Ged ≤ 0.3820), the regression coefficient was not marked but was positive. When green economy was at a high level (Ged > 0.3820), the regression coefficient was remarkable, and it increased from 0.1297 to 0.1772, indicating that after the level of green economy crossed the threshold, digital-real integration significantly promoted logistics industrial transformation and upgrading. For this reason, it may be that after the green economy attains a critical threshold, it can provide additional innovative resources and momentum, enabling a deeper integration between digital technology and the logistics industry, which promotes the digitalization process of the entire logistics industry and accelerates the pace of logistics industrial transformation and upgrading.

5.4.2. Robustness Test

From the previous regression results, it is evident that there were threshold effects of the green economy and digital-real integration on the logistics industry’s transformation and upgrading. For the sake of the reliability of the conclusion, we re-ran the threshold effect of adding the controlled variable in the benchmark regression robustness test to eliminate the impact of extremum, as listed in

Table 11. Robustness test results of the threshold effect.

Threshold Variable: Drf		Threshold Variable: Ged
Variable	Model (19)	Variable	Model (20)
Drf·1 (Drf ≤ 0.3541)	0.1238 (1.48)	Drf·1 (Ged ≤ 0.3820)	0.1466 (1.72)
Drf·1 (Drf > 0.3541)	0.1947 ** (2.68)	Drf·1 (Ged > 0.3820)	0.1922 ** (2.28)
Tech	−0.0002	Tech	−0.0023
	(−0.05)		(−0.53)
Fdi	0.0052 *	Fdi	0.0055 *
	(1.86)		(1.89)
Indus	0.0387 **	Indus	0.0413 **
	(2.52)		(2.63)
Fsc	−0.0046	Fsc	−0.0088
	(−0.49)		(−0.87)
Gov	−0.0050	Gov	−0.0041
	(−0.53)		(−0.49)
Urb	−0.0391	Urb	−0.0333
	(−1.59)		(−1.26)
_cons	0.2265 ***	_cons	0.2220 ***
	(4.43)		(4.30)
R2	0.6142	R2	0.5385
N	110	N	110

Note: t statistics in parentheses; * p < 0.1, ** p < 0.05, and *** p < 0.01.

. The regression findings of model (19), with digital-real integration as the threshold variable, displayed that after breaking through the threshold value, the regression coefficient of digital-real integration for logistics industrial transformation and upgrading was positive at the 5% level, which had the same change trend as model (17). According to the regression results of model (20), with green economy as the threshold variable, it is clear that the influence of digital-real integration in promoting logistics industrial transformation and upgrading with the growth of green economy still maintained the influence relationship corresponding to model (18). According to the analysis of the above test results, it shows that the core conclusion of hypothesis H3, that there exist threshold effects of digital-real integration on logistics industrial transformation and upgrading, still holds.

6. Discussion

Digital-real integration exerts a profound impact on the transformation and upgrading of the logistics industry. This study delved into the influence of this integration, unequivocally revealing a significant positive correlation between the two, and underscored its pivotal role in driving the logistics industry toward a more efficient and sustainable transformation. According to Saqib’s research [58], technology adoption can strengthen the sustainable operational capabilities of logistics companies, opening up a broad prospect for them to improve their sustainability performance through strategic technology investments, showing strong potential and vitality. Moreover, the integration of digital technologies into crucial aspects, such as warehouse management, transportation optimization, and customer service, has led to remarkable efficiency gains and cost savings [18]. It has also spurred the trend of personalization and customization in logistics services, accurately matching the diversified needs of the market. Based on the validation of the benchmark regression model, this study further corroborated the substantial facilitation effect of digital-real integration on the transformation and upgrading of the logistics industry, thus strongly supporting hypothesis 1.

As a global issue, the development of green economy aims to achieve a win–win situation between economic growth and environmental protection [3]. This study explored the positive impact of green economy in promoting logistics industrial transformation and upgrading through digital-real integration. Existing studies have emphasized the importance of the logistics industry’s shift from traditional models to digital and intelligent ones, and pointed out the high environmental costs associated with energy consumption and carbon emissions during this process [15]. The logistics industry’s reliance on inefficient modes of transportation and the overuse of non-environmentally friendly packaging materials have directly weakened its service quality, efficiency, and competitiveness. In the long run, the transformation, upgrading, and green development of the logistics industry exhibit obvious systemic and complex characteristics. Technological innovations, business model reconstructions, and management practice optimizations driven by the green economy have facilitated the enhancement of logistics efficiency and environmental benefits [44]. Hypothesis 2 posited that the green economy actively moderates the promotion of digital-real integration, and this hypothesis was verified by the moderation effect model.

This paper also analyzed the nonlinear impact of digital-real integration on the transformation and upgrading of the logistics industry. The research indicated that there was a significant positive correlation between digital-real integration and logistics industrial transformation and upgrading after surpassing the respective development threshold of digital-real integration and green economy, which highlighted the differentiated roles of the depth of application of digital technology and the stage of green economic growth in facilitating the development of the logistics industry. Notably, some enterprises demonstrate hesitation in funding investments to a certain extent during the process of transformation and upgrading, which is mainly due to the non-immediate manifestation of transformation benefits to increase the uncertainty and risk of capital investment [26]. However, with the increasing prevalence of digital technology and the full opening of a new era of green economy, the industry’s focus is increasingly converging on technological innovation and sustainable models. In particular, the construction of intelligent logistics systems and the deepening of green logistics practices can accelerate technological iteration, optimize resource allocation, and strengthen green supply chain management [43]. These cumulative positive effects spur enterprises to awaken, and timely undertake transformation and upgrading to help the digital and green development of logistics. Under the threshold effect model test, hypothesis 3’s assertion that the influence of digital-real integration on logistics industrial transformation and upgrading has the threshold conditions of digital-real integration and green economy held true.

Apart from that, there are some shortcomings in the research process. In terms of data collection, data from each region tended to be scattered in different institutions and units, with differences in data formats and statistical calibers, and some key information was difficult to obtain from public sources due to confidentiality and other reasons. Follow-up research can pay attention to the application of technologies such as crowdsourcing data collection in data mining. On the selection of indicators, although we selected relevant indices as much as possible from the perspectives of systematicity and operability, the indicators chosen cannot cover all elements. In the future, more comprehensive information can be collected through methods such as expert consultation to identify potential indicators. For model building, even if the model built in this paper enriched the method application of systematic level measurement, there were still constraints; for example, the resolution coefficient was set at 0.5 to simplify the calculation of correlation, which failed to fully consider the actual differences between data. In future works, the quantification of the resolution coefficient will be further studied to improve the evaluation method. In addition, the empirical research in this paper focused on Hebei Province in China, and the geographical coverage of the sample can be broadened in the future to explore similar or different phenomena and trends in a wider region.

7. Conclusions

Utilizing panel data spanning from 2013 to 2022 for 11 cities in Hebei Province of China, this study empirically examined the direct, indirect, and nonlinear impacts of digital-real integration on logistics industrial transformation and upgrading under the green economy, after comprehensively analyzing the advancement of logistics industrial transformation and upgrading, digital-real integration, and green economy. The key findings of the research are outlined as follows: (1) Digital-real integration exerted a notable positive influence on the transformation and upgrading of the logistics industry, reflecting that the in-depth fusion of digital technology and real economy has become a critical driver for advancing the modern logistics industry toward greater efficiency and intelligence. (2) The green economy positively regulated the promotion function of digital-real integration on logistics industrial transformation and upgrading. Namely, the growth of green economy will strengthen the positive effect of digital-real integration, which enriches the new way of digital-real integration, affecting the transformation and upgrading of the logistics industry. (3) There existed threshold conditions for the promotion of digital-real integration and the moderating effect of the green economy. After exceeding their respective thresholds, digital-real integration significantly facilitated logistics industrial transformation and upgrading, which extends a new research perspective for further understanding the mechanism of digital-real integration in the transformation and upgrading of the logistics industry.

Based on the aforementioned findings, the following suggestions are put forth: (1) Regions should improve transportation infrastructure to jointly promote the interconnection of global logistics networks. While formulating the development plan of the logistics industry according to its own resource endowment, they should reinforce international legal cooperation and exchanges to promote regional complementarity, thereby cultivating logistics enterprises with international competitiveness. (2) Each region should continue to popularize advanced technologies and create a global digital logistics platform to achieve real-time sharing and efficient processing of logistics information. At the same time, the introduction and cultivation of professional and technical talents in smart logistics should be enhanced, and the deep integration of government, industry, academia, research, and application should be promoted, which can offer a strong talent guarantee for the intelligent development of the logistics industry. (3) All regions should provide a policy environment conducive to the development of green economy, encourage logistics enterprises to increase R&D investment in green digital technologies, such as clean energy intelligent management systems, and innovate products, management, and service models to optimize resource utilization, thus fostering the green transformation and sustainable development of the logistics industry. (4) With the continuous growth of the green economy and the sustainable progress of digital-real integration, the competition pattern and market demand of the logistics industry will also be affected. Enterprises should establish sound dynamic monitoring and evaluation mechanisms to grasp the development pulse of the global logistics industry and timely innovate the direction of research and development, so as to ensure their continuous leading core competitiveness in the logistics industry.