Research on the Impact of University–Industry Collaboration on Green Innovation of Logistics Enterprises in China

Abstract

Green innovation has emerged as a key catalyst for the sustainable growth of logistics enterprises. Green innovation not only helps logistics enterprises reduce operating costs but also enhances their competitiveness and promotes the entire industry’s transformation towards environmental protection and efficiency. However, logistics enterprises encounter technical bottlenecks, capital shortages, and insufficient talent and infrastructure when implementing green innovation. Collaboration between universities and industries serves as a crucial method for logistics companies to access external resources and plays a significant role in promoting technological progress, knowledge transfer, and innovation capability enhancement of enterprises. This research, grounded in the theories of social capital and dynamic capabilities, explores the mechanism from the perspective of resources and capabilities, and examines how university–industry collaboration affects green innovation. This research employs a hierarchical regression model to evaluate the proposed hypotheses. The research results show that university–industry collaboration has a positive impact on social capital, slack resources, and dynamic capabilities, and social capital, slack resources, and dynamic capabilities positively influence green innovation. The research results have certain reference value for logistics enterprises to promote green innovation.

1. Introduction

Against the backdrop of increasingly severe global environmental issues, tackling climate change, lowering carbon emissions, and attaining sustainable development have become key strategic priorities for businesses [1]. As a vital component of economic activities, the logistics sector is crucial in driving global trade and economic growth. However, traditional transportation methods, packaging, and transportation of goods have made the logistics industry one of the primary sources of environmental pollution [2]. The current energy consumption and carbon emissions in the logistics industry are too high to support its sustainable development [3].

With the increasingly strict environmental policies and the growing market demand for green and sustainable products and services, green innovation has emerged as a key strategic tool for businesses to secure sustainable competitive advantages [2]. The implementation of green innovation not only reduces resource waste but also builds a green image for enterprises and enhances their core competitiveness [4]. Green innovation in logistics companies involves a range of innovative activities in technology, business processes, service models, and environmental management aimed at achieving energy savings, reducing emissions, and preventing pollution [5,6]. Green innovation in logistics enterprises covers not only new energy technologies, intelligent logistics, green packaging, and green transportation, but also management measures such as process optimization and waste recycling and reuse [7]. To reduce environmental pollution and gain sustainable competitiveness, green innovation in logistics enterprises is particularly important [8].

In green innovation research, many studies concentrate on the manufacturing sector, while the logistics industry remains an underexplored area, with limited theoretical research specifically focused on logistics companies [9]. However, green innovation in different industries presents unique characteristics [10]. In the traditional manufacturing supply chain, green innovation is primarily led by manufacturing companies and is focused on procurement, production, and transportation processes. However, retailers at the downstream end of the supply chain have a relatively low level of involvement in green innovation. In contrast, the logistics industry belongs to the service sector and is a typical industry with numerous divisions of labor, collaboration, and innovation subjects. Customer demand at the downstream end of the supply chain is often more important for the innovation of logistics enterprises than for manufacturing enterprises [11]. Since the service quality of logistics is reflected in multiple aspects such as delivery speed, reliability, response ability, and cost-effectiveness, the improvement of service quality in logistics enterprises is closely related to the demands of downstream customers [12,13]. Therefore, green innovation in the logistics enterprises is more susceptible to external influences compared with other industries, and research on green innovation in other industries offers limited practical insights for the logistics sector [11]. Currently, logistics companies face considerable fluctuations in demand within the green logistics service market due to the absence of historical data when implementing green innovation [14]. Furthermore, green innovation in logistics enterprises is a long-term process. Logistics enterprises encounter multiple challenges when engaging in green innovation, such as high costs, immature technologies, a shortage of talents and infrastructure. The intangible and unstable nature of services also reduces the motivation of logistics companies to engage in green innovation. The investment pressure and the uncertainty of returns seriously impede the motivation of logistics enterprises to carry out green innovation [6,14,15]. As a result, studying ways to foster green innovation in logistics companies is highly important.

With the growing emphasis on green innovation, considerable research has been dedicated to identifying its driving factors, such as pressure from stakeholders [16], resources and capabilities [17,18], strategic orientation [19], and executives’ personality traits [20]. However, previous research on the drivers of green innovation in logistics enterprises has largely overlooked the role of university–industry collaboration. University–industry collaboration takes various forms, including the employment of graduates, joint research, academic exchanges, employee training, technical consulting, shared facilities, and patent licensing [21]. In the present era of open innovation, collaboration between universities and industries plays a crucial role in enhancing national innovation systems and promoting economic growth. It is not only a critical driving force for corporate development but also a significant potential driver for green innovation [22]. As a means for enterprises to access external resources, university–industry collaboration helps enterprises rapidly acquire the necessary resources, providing access to cutting-edge research and enhancing their capacity for developing new technologies [23,24]. Universities, with their faculty, academic resources, and research conditions, can provide strong support for logistics enterprises in areas such as technological upgrades and product development. For instance, logistics enterprises such as China Ocean Shipping Enterprise Logistics have established enterprises–academia–research collaboration alliances with the Logistics Research Center of Shanghai Maritime University, resulting in numerous achievements in logistics planning, supply chain optimization, intelligent logistics information technology, and Internet of Things engineering for logistics equipment, which have been applied and transformed within the logistics enterprises.

Despite extensive discussions in prior research on the driving factors of green innovation, few scholars have investigated the impact of university–industry collaboration on green innovation within logistics enterprises. This study aims to explore the influence of university–industry collaboration on green innovation in logistics enterprises and elucidate its underlying mechanisms, thereby analyzing how university–industry collaboration facilitates green innovation in this sector. The primary objective of this research is to enhance logistics enterprises’ understanding of the significance of university–industry collaboration in driving their green innovation endeavors, as well as the mechanisms through which university–industry collaboration influences their green innovation processes. By identifying the key factors that enable university–industry collaboration to promote green innovation in logistics enterprises, this study seeks to assist these firms in optimizing the content and modalities of their collaborative efforts. This, in turn, will enable them to strategically acquire the resources necessary for green innovation through university–industry collaboration, thereby bolstering their sustainable innovation capabilities.

From the perspectives of resources and capabilities, and grounded in social capital theory and dynamic capability theory, this study empirically examines the relationship between university–industry collaboration and green innovation in logistics enterprises. Additionally, it investigates the mediating roles of social capital, slack resources, and dynamic capabilities in the relationship between university–industry collaboration and green innovation within logistics firms.

The remaining structure of this study is as follows: Section 2 reviews the relevant literature and proposes the research hypotheses. Section 3 outlines the research methodology and the data collection process. Section 4 presents the empirical test results. Finally, Section 5 discusses the findings and their implications, summarizes the study, and outlines potential avenues for future research. The analytical process is illustrated in Figure 1.

2. Theoretical Basis and Hypothesis Development

2.1. Theoretical Basis

2.1.1. Green Innovation

Green innovation refers to the innovative activities undertaken by enterprises with the goal of developing green technologies, such as green production, energy saving, emission reduction, and pollution control [25]. Green innovation is a crucial strategy for companies to achieve sustainable development and significantly contributes to enhancing their competitive advantage [26]. To foster green innovation within businesses, numerous scholars have conducted in-depth studies and surveys to identify the factors influencing green innovation [27,28,29]. Existing research can be categorized into two main perspectives when examining the drivers of corporate green innovation: the external environment and the internal environment of companies.

Research from the external environment perspective primarily explores the factors influencing green innovation through the lenses of stakeholder theory and institutional theory. Buysse et al. [16], based on stakeholder management, revealed that stakeholders such as employees, shareholders, and economic institutions play a vital role in the adoption of leadership-driven environmental strategies by companies. Delgado-Ceballos et al. [30] expanded upon this research, suggesting that companies with greater stakeholder integration capabilities are better equipped to develop proactive environmental strategies. Lin et al. [31] found that environmental regulations and various stakeholders, including customers, suppliers, and competitors, have differing effects on corporate green innovation activities. Institutional theory emphasizes that corporate behavior is shaped and constrained by the institutional environment in which it operates. Chen et al. [32] synthesized institutional theory with the resource-based view to investigate the mechanisms through which external institutional pressures influence corporate green innovation. The study revealed that both coercive and normative pressures significantly and positively impact corporate green innovation. Coercive pressure, manifesting through the formulation of explicit environmental standards and regulatory frameworks, obliges enterprises to implement green innovation initiatives to ensure regulatory compliance. Conversely, normative pressure guides enterprises to proactively pursue green innovation to enhance their competitiveness, driven by industry consensus and expectations.

On the other hand, research on the internal environment of companies focuses on how resources and capabilities shape corporate green innovation, often using the natural resource-based view and resource-based theory [33,34,35,36]. Ketata et al. [34] discussed that internal absorptive capacity significantly drives sustainable innovation within companies. Lin et al. [35] examined internal knowledge resources and capabilities, discovering that the sharing of green knowledge enhances dynamic green capabilities, which, in turn, fosters green service innovation and provides a green competitive edge. Lastly, Chen et al. [36] compared reactive and proactive green innovation, concluding that only internal factors, such as environmental leadership, culture, and environmental capabilities, can promote proactive green innovation strategies. Sharma et al. [37] showed that the interpretation of environmental issues by managers as opportunities or threats significantly influences the selection of corporate green innovation strategies.

With the increasing prominence of global environmental issues, the logistics industry is actively exploring green technological innovation to achieve sustainable development [38]. Green innovation in logistics enterprises refers to innovative activities focused on energy saving, consumption reduction, and pollution prevention, applied to technologies, business processes, service models, and environmental management [6]. Green innovation in logistics enterprises often involves green transportation, green packaging, green warehousing, innovation in green supply chain management, and green logistics models [15].

In studies examining the driving forces behind green innovation in logistics companies, most scholars conduct case studies of specific logistics enterprises, concluding that factors influencing the implementation of green innovation by logistics suppliers include customer pressure, organizational incentives, technological resources, human resources, and policy support. For example, Jazairy et al. [39] used case study methods to conclude that market, regulatory, competitive, enterprise size, and environmental pressures drive shippers and logistics suppliers to implement green practices. Lieb et al. [40] discovered that the main factors driving third-party logistics providers to adopt environmentally sustainable practices are customer demands and the pursuit of competitive advantage. Rossi et al. [41] conducted a case analysis of six leading third-party logistics suppliers in Europe, identifying customers, policies and regulations, marketing techniques, internal factors, and competitors as the main driving forces, as well as enterprise-specific barriers and investment costs, and it was found that the absence of legitimacy posed major barriers to green innovation. Evangelista et al. [15] conducted a case study of ten medium-sized third-party logistics suppliers in Italy and the United Kingdom, concluding that policy support, optimizing customer relationships, reducing costs, and improving corporate image are key drivers of green innovation, while barriers to green innovation include a lack of funding, economic incentives, and clear regulations. Several researchers have also carried out empirical research on green innovation in third-party logistics enterprises. For example, Chu et al. [11] conducted a questionnaire survey on third-party logistics suppliers in China, and it was discovered that both customer and competitive pressures play a significant role in driving third-party logistics providers to adopt green innovation.

Within the existing studies on the drivers of green innovation, the majority of these studies have predominantly focused on the manufacturing sector. However, different industries exhibit markedly distinct characteristics in their green innovation practices, which renders the current research findings based on manufacturing green innovation inadequate in fully meeting the practical needs of logistics enterprises exploring green innovation. In the context of the logistics industry, green innovation remains in its nascent stages of theoretical formulation and empirical exploration, with its conceptual connotations and practical pathways remaining relatively ambiguous. Currently, there is a dearth of specialized theoretical research on green innovation within logistics enterprises, with existing studies primarily relying on case analyses and lacking robust support from systematic empirical research. This deficiency hinders the comprehensive elucidation of the inherent patterns and driving mechanisms underlying green innovation in logistics enterprises. In light of this, the present study takes logistics enterprises and their associated logistics management departments as the core subjects of investigation. By constructing an empirical research framework, it delves deeply into the actual impact of university–industry collaboration on green innovation within logistics enterprises and its underlying operational mechanisms, with the aim of providing a scientific basis for the theoretical advancement and practical promotion of green innovation in the logistics sector.

2.1.2. University–Industry Collaboration

Due to the increasing complexity of emerging multidisciplinary technologies and the accelerating pace of technological updates, the traditional closed innovation model has become increasingly difficult to adapt and implement. It is becoming more important for enterprises to rely on external resources [42]. External knowledge, whether through knowledge transfer or spillover, has been shown to have a positive impact on both innovation and productivity [43,44]. When engaging in green innovation, enterprises cannot solely depend on internal resources; acquiring external knowledge is a necessary condition for the success of green innovation [45]. Research has shown that open innovation is an advanced innovation mechanism that can generate valuable ideas through knowledge exchange between different enterprises, which typically do not possess all the capabilities and knowledge required to develop new technologies [46]. External knowledge, via knowledge transfer or spillover, positively influences innovation and productivity. It refers to the collaboration between enterprises, universities, and research institutions through various means, aimed at achieving resource sharing, complementary advantages, and mutual development. Forms of collaboration include, but are not limited to, joint research, technology development, talent cultivation, intellectual property sharing, internships, and corporate training [42]. University–industry partnerships are among the most widely used open innovation models today to speed up technological progress. By establishing partnerships with academic departments, enterprises can integrate internal and external innovation resources and benefit from spillover effects [47,48]. Yang et al. [22] concluded that university–industry collaboration can alleviate the uncertainty pressures arising from cost investment, research and development risks, and economic effects associated with green innovation. It can also stimulate the driving force of manufacturing enterprises to participate in green innovation.

University–industry collaboration provides significant external support for green innovation in enterprises and is an effective pathway to promote sustainable development. Through this collaboration model, enterprises can not only speed up the development and implementation of green technologies but also gain technological advantages and innovative momentum in an increasingly competitive market. In research on the driving factors of green innovation in logistics companies, the connection between university–industry collaboration and green innovation in these companies has often been overlooked. Moreover, while many studies identify the factors influencing green innovation in logistics enterprises through case studies, few have explored the actual impact of these factors on green innovation [22,49,50]. Within the current research domain concerning the relationship between university–industry collaboration and green innovation, a notable deficiency lies in the fact that existing studies have not adequately examined the specific impact of university–industry collaboration on green innovation within logistics enterprises, nor have they fully elucidated the underlying mechanisms at play between the two. Given the diversity and complexity of interaction modes within university–industry collaboration, this study concentrates on university–industry collaboration established through both formal agreements and informal relationships. In response to the limitations of previous research, this study employs an empirical research methodology to investigate the effect of university–industry collaboration on green innovation within logistics enterprises. Furthermore, from a qualitative analytical perspective, it delves deeply into the influence of university–industry collaboration on green innovation in logistics enterprises and the mechanisms through which this influence operates.

2.1.3. Social Capital Theory

In the context of social capital theory, corporate social capital refers to the resources embedded in the relationships between enterprises and other organizations, as well as the ability to acquire these resources [51,52]. Social capital originates from social networks, with its core elements being the mutual trust and shared values formed among network members [52]. These connections, built on trust, respect, effective communication, and reciprocity, can provide competitive advantages for businesses by promoting collaboration and knowledge exchange [53]. Trust between enterprises and other organizations can strengthen communication and collaboration and promote resource exchange and combination, thus having a positive impact on product innovation [54]. Social capital theory is often used to analyze the positive factors through which the social network relationships possessed by enterprises influence innovation [55]. Business innovation often necessitates the combination of knowledge from various fields, and social capital facilitates this integration [56]. Lyu et al. [57] concluded that even during the Coronavirus Disease 2019 (COVID-19) pandemic, the social capital of digital enterprises remained significantly positively correlated with innovation performance. Furthermore, digital enterprises with higher social capital are more likely to achieve superior innovation performance. Social capital also promotes collaboration between enterprises and other organizations, fostering a shared sense of responsibility towards sustainable practices with their partners [58]. Companies with strong green social capital can improve their capacity to exchange internal information and disseminate new ideas, promoting the knowledge required for green innovation [59]. Huang et al. [18] viewed social reciprocity as the social capital that enterprises possess within a social network, equipping them with a wider range of information and greater knowledge. This enables companies to generate more ideas for spotting environmental trends and opportunities, helping them implement advanced manufacturing practices needed to create new green products or processes.

Social capital theory provides a highly explanatory analytical framework for elucidating the relationship between university–industry collaboration and green innovation in logistics enterprises. This theory focuses on how immaterial capital elements, such as social relations, networks, trust, and cooperation, influence firms’ innovative activities and competitiveness enhancement. In the context of university–industry collaboration and green innovation in logistics enterprises, the applicability of social capital theory is manifested in multiple dimensions. First, it underscores the pivotal value of relational networks in corporate innovation. University–industry collaboration inherently involves the construction of a social network, where logistics enterprises and universities establish long-term, stable cooperative relationships to facilitate the sharing of resources, technologies, and knowledge. During this process, logistics enterprises can fully leverage the research outcomes, technological advantages, and talent resources of universities and research institutions. These cooperative relationships themselves constitute significant social capital reserves for enterprises. By strengthening their relational networks, university–industry collaboration can provide logistics enterprises with broader information sources and innovative inspirations for their green innovation activities, thereby promoting the enhancement of their green innovation capabilities. Second, social capital theory emphasizes trust as a core element of social capital, which plays a crucial role in university–industry collaboration. The success of university–industry collaboration hinges not only on the exchange of material resources but also on the trust established between the two parties. Close cooperation between logistics enterprises and universities facilitates the smooth sharing of technologies and knowledge during the green innovation process, reduces information asymmetry and cooperation frictions, and thus enhances the efficiency and effectiveness of green innovation. As a form of social capital, trust can effectively reduce transaction costs in university–industry collaboration, improve the stability of innovative cooperation, and consequently promote the transformation and application of green innovation outcomes.

When analyzing the relationship between university–industry collaboration and green innovation in logistics enterprises, social capital theory not only uncovers the underlying social network mechanisms of university–industry collaboration but also provides robust theoretical support for understanding how logistics enterprises can drive green innovation through the accumulation and application of social capital.

This study investigates the impact of university–industry collaboration on green innovation in logistics enterprises and its underlying mechanisms, grounded in social capital theory and dynamic capability theory. The aim is to further enrich the application of social capital theory in the field of corporate green innovation and offer valuable insights and guidance for logistics enterprises’ green innovation practices.

2.1.4. Dynamic Capability

When investigating the influencing factors of corporate green innovation, the majority of scholars have adopted a resource-based theory or resource-based perspective, analyzing the impact of resources, capabilities, and other factors on firms’ green innovation [33,34,35,36]. These existing studies primarily focus on the resource requirements for firm development within a specific context, yet they fall short in adequately explaining how enterprises can acquire competitiveness in a changing environment.

The dynamic capabilities theory focuses on the capacity of businesses to coordinate, configure, and integrate both internal and external capabilities and resources in environments characterized by drastic changes. Specifically, it can be divided into three dimensions: sensing capability, seizing capability, and integrating capability [60]. In recent years, as governments place greater emphasis on green and low-carbon economic development, green innovation has become a strategy embraced by numerous enterprises. Businesses must undergo essential green transformations driven by market demand for green products and the competitive performance of other companies in the green market [35]. Owing to the ongoing shifts in the market landscape and technological unpredictability, enterprises must possess dynamic capabilities, proactively seeking effective resource acquisition channels to acquire the latest information on government policies, industry dynamics, and consumer demand for green products [61]. Dynamic capabilities help reduce the negative effects of environmental changes, enabling businesses to spot opportunities for green innovation in the market [62]. Many scholars have applied the dynamic capabilities theory to research on corporate green innovation. Huang et al. [18] argue that dynamic capabilities are a key driver of green innovation. Other research indicates that dynamic capabilities serve as a mediator between green innovation and the enhancement of a company’s competitive advantage [63].

Logistics enterprises face various factors when implementing green innovation, with policy being one of the primary influences. Green innovation policies influence how enterprises view green innovation, thus changing the connection between their behaviors and perceptions [64]. Moktadir et al. [65] indicate that logistics enterprises face obstacles in implementing green innovation due to the lack of government support and policies. Since policymakers and government agencies have not disclosed a clear road map for achieving sustainable development, transportation enterprises, including those in Nigeria, have not received guidance or direction from the government regarding the implementation of green innovation, which consequently affects their motivation to pursue green innovation [6]. Moreover, green innovation requires enterprises to continuously acquire external green knowledge and technology [66]. The logistics industry has a broad supply chain network, including transportation, warehousing, and distribution, meaning that the success of sustainable innovation in logistics companies depends not only on their own resources but also on establishing green partnerships with multiple collaborators to jointly optimize resource utilization [4]. Therefore, logistics enterprises need to have certain capabilities in perceiving policies, acquiring knowledge, and integrating resources for green innovation. This study uses Teece’s definition of dynamic capabilities to summarize the dynamic capabilities required for green innovation in logistics enterprises, focusing on policy awareness, knowledge acquisition, and knowledge integration capabilities [67]. Policy awareness refers to a company’s ability to detect, comprehend, and interpret relevant policy documents [27]. Knowledge acquisition capability refers to the degree to which an enterprise acquires knowledge resources from its partners, including knowledge in areas such as technology, management, products, and markets [68]. Knowledge integration capability refers to a company’s ability to systematically combine and process existing knowledge, and it is the key to innovation [69]. In this study, dynamic capabilities refer to the ability of logistics companies to recognize shifting green policy trends in a changing environment, gather the required resources for green innovation, and integrate them efficiently to strengthen their capacity to handle uncertain risks.

Dynamic capability theory emphasizes that enterprises need to possess the ability to continuously adjust and innovate in response to dynamic changes in the external environment, which is crucial for optimizing resource allocation and enhancing core competitiveness. The process of green innovation in logistics enterprises is characterized by both dynamism and uncertainty, requiring enterprises to flexibly adjust their resource bases and operational strategies in accordance with policy orientations, shifts in market demand, and technological advancements [70]. Green innovation practices in the logistics industry are significantly influenced by multiple factors, including the policy environment, market dynamics, and technological innovations [11,64]. University–industry collaboration serves as a vital channel for logistics enterprises to acquire external resources and knowledge, providing them with a platform for in-depth collaboration with academia and research institutions. This collaboration facilitates the continuous absorption of new knowledge, the enhancement of technological capabilities, and the optimization of management practices [47,48]. This process not only strengthens the ability of logistics enterprises to respond to external changes but also provides a sustained impetus for their green innovation activities.

Dynamic capability theory further posits that enterprises need to possess the capability to effectively transform external knowledge into internal capabilities, a process that significantly contributes to the enhancement of their green innovation capabilities. As an input source of external resources, university–industry collaboration provides logistics enterprises with a wealth of innovative elements. By promoting knowledge flows in areas such as technological innovation and management innovation, university–industry collaboration assists logistics enterprises in enhancing their dynamic adaptability to green innovation, thereby fostering the advancement of green innovation activities. Based on dynamic capability theory, this study delves into the mediating mechanism between university–industry collaboration and green innovation in logistics enterprises and aims to offer robust theoretical support and practical guidance for the green transformation and sustainable development of the logistics industry.

In conclusion, this study investigates the effect and mechanism of university–industry collaboration on the green innovation of logistics firms, using social capital theory and dynamic capabilities theory. It introduces social capital, slack resources, and dynamic capabilities as mediating variables from the resource and capability perspective to explore how university–industry collaboration affects green innovation in logistics enterprises.

2.2. Hypothesis Development

2.2.1. University–Industry Collaboration and Social Capital

Recent studies indicate that patent output, academic background, and alliance behavior serve as positive quality signals that are conveyed to an enterprise’s stakeholders [71]. For example, Bruton et al. [72] found that listed companies with an academic background send positive signals to the market, such as the high reputation of entrepreneurs and the strong decision-making ability of the management team. Investors who receive these positive signals are likely to reduce concerns about the enterprise’s operational status and prospects, which in turn positively affects the enterprise’s initial public offering performance. Czarnitzki et al. [73] further demonstrated that cross-organizational collaboration can alleviate financial resource constraints by functioning as a quality signal that attracts external resources. Collaboration between enterprises and universities offers opportunities for enterprises to access external resources, helping them compensate for deficiencies in specialized knowledge when developing new technologies [42]. Building on existing research, this study posits that university–industry collaboration may serve as a positive quality signal. When enterprises collaborate with universities, they transmit quality signals to other related organizations, which can attract more potential collaborators such as universities, enterprises, and customers; enhance trust between enterprises and their collaborators; and strengthen their relationships.

Social capital refers to the resources obtained by enterprises through external connections and the ability to access those resources, which largely relies on trust and shared values among members of the social network [74,75,76]. The social network is primarily made up of connections between enterprises and various social entities, including government departments, customers, suppliers, research institutions, and other enterprises. These inter-organizational connections can help enterprises acquire resources [57]. Logistics enterprises, being part of the service enterprises, often find that customer demand has a more important role in innovation than in manufacturing enterprises [11]. Logistics enterprises typically explore customers’ potential needs through collaborative alliances, which help them better acquire external resources and information [77]. Collaboration helps build high-quality, enduring relationships between logistics enterprises and their customers, increasing their social capital [13]. Therefore, for logistics enterprises, collaborative relationships with other enterprises serve as a key type of social capital for innovation. University–industry collaboration fosters long-lasting and trustworthy relationships between logistics enterprises and universities, increasing the social capital of logistics enterprises. Moreover, the academic background provided by such collaboration acts as a positive quality signal that is transmitted to existing partners, enhancing trust with these partners. This positive quality signal also attracts more universities, enterprises, and customers to collaborate with logistics enterprises, expanding their resource acquisition channels. Based on this, this study proposes the following hypothesis:

H1. 

University–industry collaboration has a positive impact on social capital.

2.2.2. Social Capital and Green Innovation in Logistics Enterprises

Social capital denotes the resources and the capacity to access resources that a business acquires by forming social connections with external entities. It facilitates the exchange of knowledge, information, and value among stakeholders during the innovation process, creating more possibilities for innovation [57,78]. In the logistics industry, the social capital required for enterprise innovation primarily manifests in the establishment of high-quality and long-lasting relationships with external partners [13,77].

Green innovation in logistics enterprises encompasses innovative improvements in product design, manufacturing, process optimization, management, marketing, and service assurance, all aimed at achieving sustainable and environmentally friendly development [6]. The complexity, systematization, and novelty of green innovation highlight its demand for knowledge resources, which necessitates continuous acquisition of green knowledge and technology from external sources [66]. The logistics industry has a broad supply chain network, including transportation, warehousing, and distribution; this indicates that achieving successful green innovation in logistics companies relies not just on internal resources but also on forming green partnerships with various collaborators to optimize resource utilization together [4]. For example, enterprises that emphasize environmental protection in the supply chain network can assist other enterprises in engaging in green innovation [32,79,80]. Similarly, a lack of environmental awareness among downstream organizations can hinder the green innovation of logistics enterprises [15,81]. Many logistics enterprises promote green innovation through establishing close strategic relationships [4]. Enterprises with a high level of trust in collaboration aim for long-term collaboration, sharing innovation experiences and costs, which can reduce the marginal innovation costs of the enterprises [82]. Suppliers who maintain good relationships with enterprises provide more environmentally friendly raw materials and green production technologies, helping enterprises conduct environmental protection activities and accelerating the pace of green innovation [33]. In contrast, enterprises with weak relationships have infrequent interactions and find it difficult to engage in deep collaboration [83]. Studies have highlighted that cooperation among governments, industry players, and academic institutions is essential for exchanging knowledge, carrying out research, and applying best practices in sustainable logistics [2]. The more social capital logistics enterprises have, the more they collaborate with external partners, and the stronger and more trusting their relationships become. Thus, higher social capital not only expands the resources and avenues for green innovation in enterprises but also encourages their motivation to pursue green innovation. High social capital also generates more pressure from stakeholders, prompting enterprises to pursue green innovation. Based on this, the following hypothesis is proposed:

H2. 

Social capital has a positive impact on green innovation in logistics enterprises.

2.2.3. University–Industry Collaboration and Slack Resources

Universities, as key players in knowledge innovation, possess significant advantages in terms of high-level talent, advanced equipment, cutting-edge technologies, and knowledge [42]. Enterprises can obtain the transfer and application of explicit knowledge, mainly in the form of patents [84]. Furthermore, enterprises can directly achieve talent transfer through mechanisms such as academic personnel exchanges, internships for university students, and university–enterprise targeted recruitment and employment [42]. For example, Shentong Express has collaborated with Shanghai Jiao Tong University on multiple technological studies, making progress in areas such as intelligent warehousing, automated distribution, and unmanned vehicle delivery technologies. Shanghai Jiao Tong University has provided Shentong Express with advanced research equipment, helping the enterprise with experimental validation and technology trials. Meanwhile, Shentong Express has acquired logistics-related technology patents, such as AI-based delivery route optimization, through its collaboration with universities, thereby strengthening its ability to attract talent and innovate technologically. Similarly, SF Express has established the “SF Logistics and Transportation Research Center” in collaboration with Beijing Jiaotong University, regularly training professionals in logistics management, transportation technology, and data analysis. It is evident that universities can provide logistics enterprises with resources such as technology patents, talent, and equipment.

Slack resources refer to surplus resources that exceed the actual needs of an enterprise [85]. Slack resources in logistics enterprises are typically related to services such as transportation, warehousing, and distribution, including labor, time, materials, machinery, external services, and necessary new technologies, processes, transportation, and information systems [86]. In joint talent development between universities and enterprises, logistics enterprises, together with universities, establish talent training bases and develop specialized training courses to cultivate talent based on industry needs. This customized education enhances the employability of graduates, ensuring that enterprises have sufficient technical and management personnel during peak hiring periods, forming a talent reserve and strengthening surplus human resources. For instance, JD Logistics has successively established industrial colleges, productive training bases, and campus supply chain ecosystems in collaboration with several universities, including Shenzhen Technology University, with the aim of nurturing high-quality talent that meets industry demands. In addition, logistics enterprises can improve resource utilization by acquiring technological resources or equipment from universities, such as information systems, patent technologies, and specialized knowledge. The application of these technologies can enhance operational efficiency, thereby releasing more surplus resources. Furthermore, governments play a key role in facilitating university–industry collaboration. Governments across the globe have introduced policies and initiatives to foster collaboration, including funding joint research projects, offering tax breaks for research and development, and setting up innovation hubs and technology parks. These measures ease financial strain on businesses and enhance surplus funds [21]. Based on this, the following hypothesis is proposed in this study:

H3. 

University–industry collaboration has a positive impact on slack resources.

2.2.4. Slack Resources and Green Innovation in Logistics Enterprises

The “dual externality problem” of green innovation indicates that businesses encounter significant costs and risks in green technological innovation, leading to a lack of incentive to pursue green process innovation [87]. Green innovation requires high investment costs, which may lead to additional expenses for the enterprise [88]. Moreover, there are several potential risks, such as technological immaturity and insufficient government support, that exist during the process of green innovation [89,90,91]. Research indicates that the foundation of green innovation in enterprises lies in the resources and capabilities they currently possess [33]. Some scholars have approached the issue from a resource perspective and concluded that slack resources can cushion the adverse effects of green innovation on financial performance [92]. Slack resources refer to the resources that exceed the minimum level required to achieve a certain output level within an organization [93]. In logistics enterprises, slack resources include available cash, underutilized technological facilities, and labor, such as research scientists with specialized knowledge [86,94]. When enterprises engage in green innovation, adequate slack resources can effectively reduce the risk of innovation failure, reduce the risks of green innovation, and minimize the performance decline at the initial stage of green strategies [95,96,97,98]. As an organizational resource that plays a buffering role internally and enhances organizational flexibility, slack resources can effectively alleviate the additional costs associated with green innovation activities [92]. In the process of green innovation, logistics enterprises can retrofit idle transportation tools (such as electric and energy-efficient modifications) rather than directly investing substantial funds in purchasing new equipment. In this way, slack resources can effectively share the financial burden of green innovation and reduce the investment risks. Moreover, slack resources can often be used as additional resources to help enterprises attempt new strategies and engage in more breakthrough innovative activities [96]. Logistics enterprises can use slack warehouse facilities and transportation tools to test and validate green technologies, thus reducing the risks of technological failure or poor market response during the innovation process. If the innovation outcome is not as expected, slack resources can be repurposed for other business areas, minimizing the losses incurred from failed innovation. Decision makers can also more easily allocate resources, effectively mitigating the threat of innovation failure and increasing executives’ tolerance for innovation risks [97,98].

Thus, slack resources offer the essential support for green innovation in logistics companies and minimize the related risks. Based on this, this study proposes the following hypothesis:

H4. 

Slack resources have a positive impact on green innovation in logistics enterprises.

2.2.5. University–Industry Collaboration and Dynamic Capability

The dynamic capabilities theory suggests that, for enterprises, the inability to adapt to external environmental changes cannot be improved solely by maintaining strategic resources or expanding competitive advantages [60]. Because of ongoing shifts in the market environment and technological unpredictability, enterprises need dynamic capabilities to timely grasp the latest developments in government policies, industry trends, and consumer demand for green products [61]. In this article, dynamic capabilities refer to the capacity of logistics companies to sense green policy information in a shifting environment, acquire the resources required for green innovation, and integrate and utilize the acquired resources to enhance their ability to cope with uncertain risks. This includes policy perception capability, knowledge acquisition capability, and knowledge integration capability [99].

University–industry collaboration allows both parties to benefit from each other’s knowledge, professional skills, and technology [100]. Universities can apply their research findings to practical business problems, improving the practical relevance and applicability of the research and making academic achievements more targeted and meaningful, while also securing project funding from enterprises. Enterprises can enhance their innovation capabilities and technical levels through collaboration, better responding to market changes and technological challenges. Specifically, in university–industry collaboration, both parties can organize policy seminars, training courses, and other activities to deepen the enterprise’s understanding of relevant policies. Such activities provide a platform for enterprises to interact with professionals and scholars, thereby enhancing their understanding and perception of policies. For example, Zhejiang University provided Cainiao Network Technology with interpretations of tax preferential policies for the e-commerce logistics industry. Expert teams helped logistics enterprises understand tax reduction policies, how to avoid taxes reasonably, and how to legally utilize policy support, further enhancing the enterprise’s competitiveness and profitability. In addition, logistics enterprises generally use a variety of digital technology facilities to support various business activities. Common digital infrastructure includes the internet and various software systems like transportation management systems, enterprise resource planning systems, and supply chain management systems [101]. University–industry collaboration fosters frequent interaction and communication, enabling universities to leverage their expertise and resources to provide cutting-edge research and development capabilities to enterprises [24]. The technological resources provided by universities can strengthen the digital capabilities of logistics enterprises, enhancing their ability to acquire and integrate knowledge. Meanwhile, talent joint training, as the core content of university–industry collaboration [42], allows enterprises to absorb many information technology talents, which strengthens the use of information technology within the enterprise. Studies indicate that digital technology resources can promote social interaction and knowledge sharing among small and medium-sized enterprises, reduce the cost of information search, and promote the scope and effectiveness of knowledge integration, thus increasing the knowledge acquisition and integration capabilities of enterprises [102]. In addition to digital resources, the increase in social capital within logistics enterprises also enhances their policy perception, knowledge acquisition, and integration capabilities. As mentioned earlier, university–industry collaboration and the potential collaborators it brings enrich the enterprise’s social network, directly increasing the social capital of logistics enterprises. Social capital refers to the amount and strength of relationships between organizations, indicating the channels through which they perceive and acquire information [103]. Regular and close interactions between organizations enhance expectations for collaborative growth among different enterprises and facilitate the integration of business resources [104]. Based on this, this study proposes the following hypothesis:

H5. 

University–industry collaboration has a positive impact on dynamic capabilities.

2.2.6. Dynamic Capabilities and Green Innovation in Logistics Enterprises

Studies have revealed that logistics companies’ limited access to and understanding of green policies can affect their green innovation initiatives. Due to the lack of clear guidance for attaining sustainable development provided by policy analysts and government agencies, transportation enterprises, including those in Nigeria, have not received government directions and guidelines on implementing green innovation, which, in turn, affects their motivation to engage in green innovation [6]. Lin and Ho et al. [105] used survey data collected from Chinese logistics enterprises and confirmed that policy information has an impact on promoting green innovation in enterprises. A strong policy perception capability indicates that an enterprise is highly sensitive to government policies and can quickly filter through different policy information in the external environment to identify the information beneficial to the enterprise’s development, thus leading to a higher response rate to policies [106]. An enterprise with good policy perception ability can help logistics enterprises understand market and government trends. By gaining insights into policy movements, enterprises can anticipate green technological and management innovations, aligning with future environmental standards and improving the likelihood of receiving policy rewards and subsidies. This helps reduce innovation costs for the enterprise. Additionally, due to the high cost and technical complexity of green innovation, which requires more knowledge resources than traditional innovation, enterprises need to continuously acquire green knowledge and technologies from external sources [66]. The stronger the enterprise’s ability to acquire knowledge, the more external knowledge sources and channels it can access, thus enabling the enterprise to acquire knowledge from various technological fields. Green innovation in logistics companies covers key activities throughout the entire supply chain, such as ordering, packaging, transportation, warehousing, and recycling [7]. Within the service supply chain, logistics companies, as key partners, are tasked with resource integration, developing comprehensive logistics service plans and leveraging their market knowledge and strategic decision-making skills to lead the overall operation of the supply chain [107]. Therefore, logistics enterprises need to have strong integration capabilities to mobilize and distribute resources such as funds, human resources, and knowledge required for green innovation, improving the availability of resources and facilitating the promotion of green innovation across the entire supply chain. By integrating green knowledge from suppliers, partners, and the enterprise itself, logistics enterprises can optimize every stage of the supply chain, reducing carbon footprints. For instance, integrating multiple knowledge points such as the selection of green packaging materials, optimization of energy efficiency, and environmentally friendly transportation methods ensures that green innovation covers the entire supply chain, thereby achieving the collaborative effects of green logistics. Research has shown that the availability of high-quality data can effectively help the Nigerian transportation sector adopt green innovation, while the lack of knowledge integration tools could pose significant barriers to achieving sustainability in freight logistics [88]. Gauthier et al. [108] suggested that the key to achieving green innovation is a company’s capacity to successfully incorporate environmentally friendly technologies obtained from external sources into its existing production and manufacturing systems. This requires blending external, diverse knowledge with the company’s pre-existing knowledge base.

Under the strategic objectives of “reaching peak carbon emissions and achieving carbon neutrality”, an enterprise’s higher dynamic capability for green development is beneficial for timely understanding government policies, quickly capturing changes in market demand, and promoting green technological innovation [109]. An enterprise with higher policy perception ability, resource acquisition capability, and integration ability will help better understand policy information, enrich the enterprise’s knowledge resources, and promote the reorganization of enterprise knowledge, thus securing the funds required for green innovation and mitigating the risks linked to it. Based on this, the following hypothesis is proposed in this study:

H6. 

Dynamic capabilities have a positive impact on green innovation in logistics enterprises.

Based on the above hypothesis, the research model of this study is shown in Figure 2.

3. Data and Methods

3.1. Preliminary Test

This study collected research data through a questionnaire survey. The questionnaires were distributed via multiple online and offline channels. To ensure the reliability and validity of the research process and outcomes, this study, during the scale construction phase, comprehensively referenced research scales that have been widely acknowledged in both domestic and international literature. Concurrently, it closely integrated the actual operational contexts of logistics enterprises and fully incorporated the professional insights of logistics practitioners, thereby conducting the preliminary design and pre-survey of the questionnaire.

In the process of developing scale items, this study systematically reviewed the literature in relevant fields to construct the measurement scale content for core variables. On this basis, and in light of the specific background and requirements of this study, some linguistic expressions in the original scale were adjusted and refined. Furthermore, to further enhance the accuracy and rationality of the scale items’ expressions, this study actively sought the opinions of logistics practitioners and professionals in related fields, and accordingly made targeted revisions and improvements to the scale, ultimately forming the survey questionnaire. In this study, DeepSeek R1 was solely utilized for fundamental tasks such as language and grammar checking, as well as rectifying spelling errors.

During the pretest phase, this study initially distributed the questionnaire to a small subset of logistics enterprises managers in China to validate the applicability and accuracy of the constructed scale, ensuring that the questionnaire could truthfully and accurately reflect the opinions and perspectives of the survey respondents. Specifically, a total of 167 questionnaires were distributed during the pretest phase. The research team rigorously screened the returned questionnaires, excluding those that were incompletely filled out, were answered in a non-standardized manner, contained incorrect responses to test questions, exhibited logical inconsistencies, or had abnormal response times, such as excessively short or long completion times. Ultimately, this study obtained 110 valid questionnaires from the pretest. This study conducted reliability and validity tests on the data collected during the pretest. The test results indicated a high level of reliability and validity in the data, providing robust support for the in-depth continuation of subsequent research. The specific test results are detailed in

Table 1. Pretest reliability and validity results.

Variable	Item Coding	Cronbach’s α	KMO
University–industry collaboration	UIC1	0.820	0.859
	UIC2
	UIC3
	UIC4
	UIC5
	UIC6
Social capital	SC1	0.779	0.787
	SC2
	SC3
	SC4
Slack resources	SR1	0.709	0.792
	SR2
	SR3
	SR4
Dynamic capabilities	DC1	0.740	0.718
	DC2
	DC3
	DC4
Logistics enterprises green innovation	LEGI1	0.779	0.743
	LEGI2
	LEGI3
	LEGI4

.

3.2. Data Collection

To examine the effect of university–industry collaboration on the green innovation of logistics companies, and to investigate the mediating roles of social capital, slack resources, and dynamic capabilities in the relationship between university–industry collaboration and green innovation within logistics enterprises, this study focuses on logistics enterprises or departments related to logistics in enterprises. Questionnaires were distributed to the managers of logistics enterprises or related departments in China. To guarantee the feasibility of the questionnaire, the design process was carried out in three stages. First, the measurement indicators were modified and improved through consultations and interviews with relevant experts from academia and industry. Second, a preliminary test was conducted by collecting 110 questionnaires for a small-scale pretest, followed by a second round of revisions and improvements to the overall content of the questionnaire. Finally, the formal survey was conducted. A total of 500 questionnaires were distributed, with 398 returned, yielding a return rate of 79.6%. After excluding invalid questionnaires, 280 valid responses were obtained, resulting in an effective response rate of 56.0%.

In the process of conducting exploratory factor analysis, the selection of an appropriate sample size is crucial for ensuring the reliability of the analytical outcomes. The determination of sample size necessitates a flexible adjustment that closely aligns with the research objectives and the number of factors under investigation. According to statistical principles, when performing exploratory factor analysis, each factor should ideally correspond to a minimum of 5 to 10 samples to safeguard the robustness of the analysis [110]. In the present study, a total of 22 factors were involved in the analysis. Ultimately, through empirical investigation, an effective sample size of 280 was obtained. Consequently, the effective sample size in this study adequately fulfills the requirements for conducting exploratory factor analysis.

This study primarily employs procedural control and statistical control methods to reduce the potential impact of common method variance [111]. First, a t-test was conducted on the valid and invalid questionnaires, and the results showed no significant differences, indicating that there was no significant non-response bias. Second, during the survey process, participants were informed that the survey was solely for academic research purposes, that complete anonymity was ensured, and that there were no right or wrong answers. This helped reduce the issue of common method bias. Third, the independent and dependent variables were placed in different sections of the questionnaire scale to reduce response bias from participants. Fourth, the language of the questionnaire was kept concise, clear, and easy to understand to improve the accuracy of the responses. Fifth, during the statistical analysis, Harman’s single-factor test was performed on the variables, and the first factor explained 37.65% of the variance. This suggests that the impact of common method bias on the research results is minimal.

3.3. Descriptive Statistics

The survey participants in this study are employees from logistics enterprises or logistics departments that have collaboration with universities. The forms of collaboration between logistics enterprises and universities include both formal agreements and informal relationships established through university–industry collaboration. These forms specifically include, but are not limited to, the employment of graduates, joint research, academic exchanges, employee training, technical consulting, shared facilities, and patent licensing. As shown in

Table 2. Overview of respondents (n = 280).

Variables	Options	Percentage
Gender	Men	73.6
	Female	26.4
Age	18–30	19.6
	30–40	68.9
	40–50	8.6
	50–69	2.9
Education	High school	2.9
	Junior college	8.2
	Undergraduate	77.1
	Master’s degree or above	11.8
Position	Senior managers	5.7
	Middle managers	27.9
	Grassroots managers	65.0
	Grassroots personnel	1.4
Enterprise age	Less than 5 years	8.2
	5–10 years	20.4
	10–15 years	28.2
	15–20 years	20.0
	More than 20	23.2
Enterprise scale	Less than 20 people	6.8
	20–100 persons	20.0
	100–500 persons	30.7
	500–1000 people	12.1
	More than 1000	30.4
Enterprise nature	State-owned	11.4
	Privately owned	85.7
	Foreign investment	2.1
	Sino-foreign joint venture	0.7

, the survey primarily targeted middle and lower-level managers within the enterprises, most of whom hold a bachelor’s degree or higher. These individuals possess a deeper understanding of professional knowledge and the implementation of green activities within the enterprise, thus enhancing the authenticity of the survey responses. Over 70% of the enterprises have been operating for more than 10 years, and the types of logistics industries are primarily focused on general transportation, cold chain transportation, general warehousing, and cold chain warehousing, which are the main types of logistics industries facing environmental challenges. In terms of years of operation, the largest proportion of enterprises (28.2%) have been operating for 10 to 15 years. Regarding the size of the enterprises, the largest proportion (30.7%) consists of enterprises with 100 to 500 employees, indicating that most of the surveyed enterprises are medium-sized. In terms of ownership, the largest proportion (85.7%) consists of privately owned enterprises.

3.4. Variable Measurement

To guarantee the consistency and accuracy of the scale, the variables in this study primarily refer to mature scales and related literature from both domestic and international sources. Adjustments were made based on the actual conditions of green innovation in logistics enterprises to ensure their rationality and effectiveness. Prior to finalizing the official questionnaire, a pretest was conducted based on field interviews with a portion of the enterprises, and the questionnaire was revised according to feedback from the interviews and pretest. The questionnaire uses a 7-point Likert scale, with responses ranging from “strongly disagree” to “strongly agree”, assigned scores from 1 to 7. The questionnaire is divided into two major modules: First, the control variable items refer to research on corporate green innovation and incorporate theoretical foundations and practical backgrounds. It is believed that enterprise size, enterprise nature, and years of operation may have certain effects on the research path [112,113]. Therefore, this study includes enterprise size, enterprise nature, and years of operation as control variables, with a total of three items, to minimize the impact of individual enterprise factors on the research outcomes and further improve the accuracy and validity of the empirical analysis. The second module contains the items for the research variables, including five variables: green innovation in logistics enterprises, university–industry collaboration, social capital, slack resources, and dynamic capabilities, with a total of 22 items. Specifically, the green innovation scale for logistics enterprises contains 4 items, the university–industry collaboration scale contains 6 items, the social capital scale contains 4 items, the slack resources scale contains 4 items, and the dynamic capabilities scale contains 4 items. The specific items for each variable are shown in

Table 3. Measurement indicators and basis.

Variables	Code	Item	Source
Logistics enterprises green innovation	LEGI	LEGI1. Our enterprise creates and builds new services based on our focus on the environment.	[114]
		LEGI2. Our enterprise reduces resource and energy consumption in the service process.
		LEGI3. We minimize the release of harmful substances during the service process of our enterprises.
		LEGI4. We use clean technology to treat and recycle waste.
University–industry collaboration	UIC	UIC1. There are many transfers of scientific and technological achievements between our enterprises and universities.	[115,116]
		UIC2. There is a large flow of talents between our enterprises and universities.
		UIC3. The forms of collaboration between our enterprises and universities are diversified.
		UIC4. Our enterprise cooperates closely with a university (research institute, research institute) to carry out R&D activities of a project.
		UIC5. Our enterprises and universities build a joint laboratory and school-enterprise communication platform.
		UIC6. Our enterprise invites university professionals to guide the problems encountered in the work of employees, or to hire technical consultants for the enterprise.
Social capital	SC	SC1. The relationship between our enterprise and the enterprise that establishes the connection is trustworthy.	[74,117]
		SC2. Enterprises that establish contact with our enterprise are willing to support the work of our enterprise.
		SC3. Our enterprise is the same or similar in the rules of conduct with the enterprise that has established contact with our enterprise.
		SC4. Enterprises that establish contact with our enterprise can communicate sincerely and effectively with our enterprise.
Slack resources	SR	SR1. We have enough financial resources within the enterprise that can be used for free control.	[118]
		SR2. Our enterprise ‘s retained earnings (such as undistributed profits) are sufficient to support market expansion.
		SR3. Our enterprise has more potential relationship resources to use.
		SR4. We enterprises can obtain bank loans or other financial institutions’ funding when needed.
Dynamic capabilities	DC	DC1. Our enterprise can understand government policy in many ways.	[60]
		DC2. Our enterprise has more advantages in obtaining government policies than other enterprises in the industry.
		DC3. Our enterprise can transfer external knowledge to internal applications.
		DC4. Our enterprise often communicates with stakeholders (such as distributors and retailers).

.

4. Empirical Results

4.1. Reliability and Validity Analysis

In this study, data analysis was conducted using a hierarchical regression model. To ensure the applicability of the research data, normality tests were first performed on five key variables: university–industry collaboration (UIC), social capital (SC), slack resources (SR), dynamic capabilities (DC), and logistics enterprises’ green innovation (LEGI).

In the normality tests, skewness and kurtosis serve as core indicators for measuring the shape of data distribution and are widely employed to assess whether the data conform to a normal distribution. The closer the values of skewness and kurtosis are to zero, the more the data distribution approximates a normal distribution. The detailed results of the normality tests in this study are presented in

Table 4. The normality test results.

Variable	Item Coding	Mean	Skewness	Kurtosis
University–industry collaboration	UIC1	4.95	−0.464	0.009
	UIC2	5.21	−0.847	0.36
	UIC3	4.98	−0.643	−0.114
	UIC4	5.29	−0.965	1.192
	UIC5	5.36	−1.107	1.078
	UIC6	4.86	−0.652	−0.152
Social capital	SC1	5.62	−1.25	2.559
	SC2	5.35	−1.109	1.94
	SC3	5.31	−1.034	1.338
	SC4	5.51	−1.13	2.098
Slack resources	SR1	5.45	−0.926	1.446
	SR2	5.18	−0.902	0.964
	SR3	5.39	−1.238	2.864
	SR4	5.21	−0.906	1.171
Dynamic capabilities	DC1	5.65	−0.994	1.631
	DC2	5.50	−1.211	2.118
	DC3	5.52	−1.172	2.106
	DC4	5.61	−0.91	1.171
Logistics enterprises green innovation	LEGI1	5.44	−1.048	1.759
	LEGI2	5.55	−1.091	1.684
	LEGI3	5.66	−1.107	1.724
	LEGI4	5.52	−1.236	2.318

. The findings reveal that the absolute values of skewness for all measured items fall within the range of 0.464 to 1.238, below the threshold value of 3. Meanwhile, the absolute values of kurtosis range from 0.152 to 2.864, below the threshold value of 10 [119]. Based on these test results, it can be concluded that the sample data in this study generally conform to the characteristics of a normal distribution, thereby providing a solid data foundation for the subsequent parametric statistical analyses in this research.

This study employed SPSS 25.0 for data validation. As demonstrated in

Table 5. Scale reliability analysis.

Variable	Item Coding	Cronbach’s α	CR
University–industry collaboration	UIC1	0.820	0.797
	UIC2
	UIC3
	UIC4
	UIC5
	UIC6
Social capital	SC1	0.779	0.742
	SC2
	SC3
	SC4
Slack resources	SR1	0.709	0.643
	SR2
	SR3
	SR4
Dynamic capabilities	DC1	0.740	0.738
	DC2
	DC3
	DC4
Logistics enterprises green innovation	LEGI1	0.779	0.780
	LEGI2
	LEGI3
	LEGI4

, the Cronbach’s α values and composite reliability (CR) values for all variables in this study were found to exceed 0.6, surpassing the standard threshold [25,120]. Consequently, the scale indicators in this study exhibit satisfactory consistency, and the data quality is deemed reliable.

As shown in

Table 6. Scale validity analysis.

Variables	Number of Items	KMO	Bartlett Test of Sphericity
			Approximate Chi-Square	DF	Sig.
Logistics enterprises green innovation	4	0.735	303.818	6	0.000
University–industry collaboration	6	0.865	496.655	15	0.000
Social capital	4	0.770	295.836	6	0.000
Slack resources	4	0.739	192.683	6	0.000
Dynamic capabilities	4	0.734	234.941	6	0.000

, the KMO values for the scales of green innovation, university– industry collaboration, social capital, slack resources, and dynamic capabilities are all greater than 0.7, and Bartlett’s test of sphericity is significant at the 95% level, indicating that the scales possess high structural validity, and that factor analysis can proceed to the next step [121].

This study conducted exploratory factor analysis using SPSS version 25.0. As shown in

Table 7. Composition matrix after rotation.

Measure Item	LEGI	UIC	SC	SR	DC
LEGI1	0.749
LEGI2	0.743
LEGI3	0.627
LEGI4	0.620
UIC1		0.731
UIC2		0.420
UIC3		0.655
UIC4		0.789
UIC5		0.693
UIC6		0.450
SC1			0.594
SC2			0.751
SC3			0.621
SC4			0.619
SR1				0.737
SR2				0.480
SR3				0.406
SR4				0.591
DC1					0.611
DC2					0.698
DC3					0.693
DC4					0.567

, the results reveal that all items fall within the expected five dimensions. The results of the principal component analysis align with the scale’s structural dimensions, confirming that the scale possesses strong structural validity. The factor loadings range from 0.406 to 0.789, all of which are greater than 0.4, indicating that the factor loading is relatively high and that the items for each variable have a high explanatory power for the respective variables [121].

4.2. Correlation Analysis

Correlation analysis refers to the examination of the degree of association between two variables to clarify the relationship between them, typically represented by the Pearson correlation coefficient.

Table 8. Pearson correlation analysis.

Variable	Mean	Standard Deviation	1	2	3	4	5
Logistics enterprises green innovation	5.541	0.921	1
University–industry collaboration	5.109	1.012	0.542 ***	1
Social capital	5.449	0.941	0.613 ***	0.615 ***	1
Slack resources	5.310	0.893	0.549 ***	0.656 ***	0.631 ***	1
Dynamic capabilities	5.572	0.881	0.526 ***	0.505 ***	0.572 ***	0.622 ***	1

Notes: *** indicates significance at the 0.01 level.

presents the means, variances, and correlation coefficients of the main variables. The results show that the variables of green innovation, university–industry collaboration, social capital, slack resources, and dynamic capabilities in logistics enterprises are significantly correlated with each other.

4.3. Common Method Bias Analysis

This study employed a dual strategy combining procedural and statistical controls to mitigate common method bias [111]. First, the study explicitly stated the academic nature of the survey and the principle of anonymity in the questionnaire instructions to reduce respondents’ defensive tendencies. Additionally, randomization of item sequencing was implemented, and the measurement modules for independent and dependent variables were spatially segregated to effectively minimize measurement bias arising from order effects. Second, through expert consultations and other means, the study refined the phrasing of items in the questionnaire to ensure semantic clarity. Furthermore, t-tests were conducted to compare invalid and valid questionnaires, revealing that none of the t-values for the measured items reached statistical significance, thereby indicating the absence of systematic non-response bias. Harman’s single-factor test was also conducted during the analysis. As demonstrated in

Table 9. Common method bias analysis.

Component	Unrotated Eigenvalues
	Total	Variance (%)	Cumulative (%)
1	8.282	37.646	37.646
2	1.515	6.889	44.535
3	1.234	5.611	50.146
4	0.987	4.486	54.632
5	0.930	4.225	58.857

, the first factor extracted under unrotated conditions explained 37.65% of the variance, which is below the 40% critical threshold [122]. This suggests that common method bias does not pose a substantive threat to the validity of the study’s conclusions.

4.4. Regression Results Analysis

This study employs the testing steps proposed by Baron and Kenny et al. [123] to validate the theoretical hypotheses: (1) Examine the impact of the independent variable on the mediator variable, taking into account control variables (such as the type of enterprise, enterprise size, and enterprise age); the independent variable, university–industry collaboration, is entered into the regression equation to analyze its impact on the green innovation of logistics enterprises. (2) Examine the effect of the independent variable on the dependent variable. By including control variables, the independent variable is incorporated into the regression equation to assess how university–industry collaboration influences green innovation in logistics enterprises. (3) Test the mediation effect. With the independent variable and control variables included, the mediator variables are added to the regression equation to evaluate the combined impact of university–industry collaboration, social capital, slack resources, and dynamic capabilities on the green innovation of logistics enterprises.

This study employed a hierarchical regression model to test the proposed research hypotheses. In the data processing stage, mean-centering was conducted on the variables involved in the interaction terms to enhance the accuracy and stability of model estimation [124]. This study calculated and examined the variance inflation factor (VIF) values. The detailed results of the regression analysis are presented in

Table 10. Model details.

Model	Relationship
1	Control Variable -> Social Capital
2	University–Industry Collaboration -> Social Capital
3	Control Variable -> Slack resources
4	University–Industry Collaboration -> Slack Resources
5	Control Variable -> Dynamic Capabilities
6	University–Industry Collaboration -> Dynamic Capabilities
7	Control Variable -> Green Innovation
8	University–Industry Collaboration -> Green Innovation
9	Social Capital -> Green Innovation
10	Control Variable and Social Capital -> Green Innovation
11	Slack Resources -> Green Innovation
12	Control Variable and Slack Resources -> Green Innovation
13	Dynamic Capabilities -> Green Innovation
14	Control Variable and Dynamic Capabilities -> Green Innovation

and

Table 11. Regression analysis results.

Variable	Social Capital		Slack Resources		Dynamic Capabilities		Green Innovation
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9	Model 10	Model 11	Model 12	Model 13	Model 14
Firm age	0.085	0.000	0.141	0.043	0.139	0.078	0.068	−0.003	0.018	0.004	−0.010	−0.023	−0.011	−0.033
Firm scale	−0.066	−0.061	−0.036	0.034	−0.079	−0.075	0.023	−0.027	0.061	0.052	−0.043	0.038	0.067	0.056
Firm nature	0.322	0.256	0.143	−0.018	0.143	0.095	0.109	−0.053	−0.076	−0.050	0.030	0.027	0.028	0.017
University–industry collaboration		0.609 ***		0.603 ***		0.440 ***		0.509 ***		0.263**		0.301 ***		0.341 ***
Social capital									0.576 ***	0.404 ***
Slack resources											0.554 ***	0.345 ***
Dynamic capabili													0.565 ***	0.383 ***
R2	0.029	0.414	0.036	0.439	0.036	0.263	0.018	0.359	0.425	0.480	0.369	0.438	0.363	0.480
ΔR2	0.029	0.385	0.036	0.403	0.036	0.227	0.018	0.341	0.407	0.121	0.351	0.079	0.351	0.121
F	2.543	45.093	3.146	49.807	3.205	22.716	1.529	35.636	47.059	46.816	37.204	39.577	37.204	46.897
ΔF	2.543	167.775	3.146	183.078	3.205	78.344	1.529	135.549	180.435	59.074	141.710	35.854	141.710	59.332
Max VIF	1.268	1.298	1.268	1.298	1.268	1.298	1.268	1.298	1.287	1.648	1.304	1.770	1.299	1.364

Notes: *** indicates significance at the 0.01.

. The findings reveal that the VIF values for all variables range from 1.268 to 1.770, which are lower than the commonly used threshold of 10. This outcome indicates that, in the context of this study, the issue of multicollinearity is not prominent and does not pose a substantial threat to the reliability and validity of the regression analysis results [125].

Models 1 to 6 in Table 7 assess the influence of control variables on each mediator variable, as well as the effect of the independent variable on the mediator variables after accounting for control variables. In Model 1, social capital is set as the dependent variable, with control variables as the predictors in the regression equation. The findings reveal that no linear regression relationship exists between the control variables and social capital. Model 2 builds on Model 1, incorporating university–industry collaboration as an independent variable. The results demonstrate a significant positive relationship between university–industry collaboration and social capital (β = 0.609, p < 0.001), indicating a strong influence of university–industry collaboration on social capital. Thus, Hypothesis 1 is supported.

In Model 3, slack resources are the dependent variable, and control variables are the predictors in the regression model. The results show no linear regression relationship between control variables and slack resources. Model 4, an extension of Model 3, introduces university–industry collaboration as the independent variable. The analysis reveals a strong positive correlation between university–industry collaboration and slack resources (β = 0.603, p < 0.001), demonstrating a substantial impact of university–industry collaboration on slack resources. Hypothesis 3 is thus supported.

Model 5 uses dynamic capabilities as the dependent variable, with control variables as independent variables in the regression equation. The results suggest no significant linear regression relationship between the control variables and dynamic capabilities. Model 6, based on Model 5, includes university–industry collaboration as the independent variable. The findings show a significant positive relationship between university–industry collaboration and dynamic capabilities (β = 0.440, p < 0.001), indicating a strong effect of university–industry collaboration on dynamic capabilities. Hypothesis 5 is thus supported.

Models 7 to 8 analyze the impact of control variables on the dependent variable, and the effect of the independent variable on the dependent variable after incorporating the control variables. In Model 7, green innovation of logistics enterprises serves as the dependent variable, with control variables as the independent variables in the regression equation. The results show no linear regression relationship between the control variables and green innovation of logistics enterprises. Model 8, building upon Model 7, includes university–industry collaboration as the independent variable. The results reveal a significant positive correlation between university–industry collaboration and green innovation (β = 0.509, p < 0.001), indicating a strong positive effect of university–industry collaboration on green innovation of logistics enterprises.

Models 9 to 14 examine the effect of each mediator variable on the dependent variable after including the control variables, and the combined effect of the control variables and independent variables on the dependent variable when each mediator is introduced. Model 9, building on Model 7, includes social capital as the independent variable. The results show a significant positive impact of social capital on the green innovation of logistics enterprises (β = 0.576, p < 0.001), confirming a strong positive influence of social capital on green innovation. Therefore, Hypothesis 2 is supported.

In Model 11, based on Model 7, slack resources are introduced as the independent variable. The results show that slack resources significantly enhance the green innovation of logistics enterprises (β = 0.554, p < 0.001), indicating a strong positive effect of slack resources on green innovation. Hence, Hypothesis 4 is supported.

Model 13, based on Model 7, includes dynamic capabilities as the independent variable. The results show that dynamic capabilities have a significant positive impact on the green innovation of logistics enterprises (β = 0.565, p < 0.001), supporting the conclusion that dynamic capabilities strongly influence green innovation. Therefore, Hypothesis 6 is supported.

Based on the findings of Hypotheses 1 and 2, it is concluded that university–industry collaboration positively affects social capital, and social capital in turn influences the green innovation of logistics enterprises. Therefore, social capital mediates the relationship between university–industry collaboration and green innovation. Similarly, based on Hypotheses 3 and 4, slack resources also mediate the effect of university–industry collaboration on green innovation. Finally, as established by Hypotheses 5 and 6, dynamic capabilities mediate the relationship between university–industry collaboration and green innovation.

In Model 10, extending Model 9, university–industry collaboration is added as the independent variable. The results indicate that social capital significantly influences the green innovation of logistics enterprises (β = 0.404, p < 0.001), and university–industry collaboration also has a notable impact on green innovation (β = 0.263, p < 0.05). Compared with Model 8, where university–industry collaboration has a stronger effect on green innovation (β = 0.509, p < 0.001), the presence of social capital weakens the effect of university–industry collaboration on green innovation, confirming that social capital partially mediates the relationship between university–industry collaboration and green innovation.

Model 12, building on Model 11, includes university–industry collaboration as the independent variable. The results show that slack resources significantly affect green innovation (β = 0.345, p < 0.001), and university–industry collaboration also impacts green innovation (β = 0.301, p < 0.05). Compared with Model 8, where university–industry collaboration had a stronger effect (β = 0.509, p < 0.001), the effect of university–industry collaboration on green innovation is reduced by the inclusion of slack resources, supporting the claim that slack resources partially mediate the relationship between university–industry collaboration and green innovation.

Model 14, extending Model 13, includes university–industry collaboration as the independent variable. The findings reveal that dynamic capabilities significantly affect green innovation (β = 0.383, p < 0.001), and university–industry collaboration also influences green innovation (β = 0.341, p < 0.05). Compared with Model 8, where university–industry collaboration had a stronger effect (β = 0.509, p < 0.001), the presence of dynamic capabilities reduces the effect of university–industry collaboration on green innovation, confirming that dynamic capabilities partially mediate the relationship between university–industry collaboration and green innovation.

4.5. Robustness Test

This study uses the approach of “introducing control variables” to assess the robustness of the empirical findings and further validate the research conclusions. The process of implementing green innovation usually demands an extended period of accumulation and learning. When enterprises start green innovation, they usually undergo a process of trial and error, adjustment, and improvement. Enterprises that have implemented green innovation for a longer time can leverage the advantage of time to accumulate experience in continuously optimizing technology, management, and market operations, thereby enhancing their innovation capabilities. Conversely, companies that adopt green innovation at a later stage can leverage the experiences and insights of early innovators, yet they still encounter considerable challenges, especially in the areas of technology integration and management innovation. Exploring green innovation takes longer for these enterprises, which could influence the outcomes. As such, the time of implementing green innovation is included as a control variable. The robustness test is conducted to confirm the stability of the regression results from the earlier sections, maintaining the same testing method and omitting the model-building process here. Only the regression results from the robustness test are shown. The analysis reveals the following: collaboration between universities and enterprises positively affects social capital (β = 0.615, p < 0.001), social capital positively influences green innovation in logistics companies (β = 0.576, p < 0.001), collaboration between universities and enterprises positively impacts slack resources (β = 0.606, p < 0.001), slack resources positively affect green innovation in logistics companies (β = 0.556, p < 0.001), collaboration between universities and enterprises positively influences dynamic capabilities (β = 0.447, p < 0.001), and dynamic capabilities positively impact green innovation in logistics companies (β = 0.564, p < 0.001). Consequently, all hypotheses (H1–H6) are supported. In conclusion, the robustness test results align with the regression analysis, confirming the robustness of the research findings.

5. Discussion and Conclusions

5.1. Discussion

This study employs logistics enterprises as the empirical research context, constructing a theoretical analytical framework grounded in social capital theory and dynamic capabilities theory. Adopting a resource-based view and dynamic capabilities perspective, it introduces social capital, slack resources, and dynamic capabilities as mediating variables to investigate the driving effects and underlying mechanisms of university–industry collaboration on green innovation in logistics enterprises. Through empirical research and data analysis, the following key conclusions are drawn:

First, university–industry collaboration exerts a significant and positive promotive effect on the social capital of logistics enterprises, evidenced by the effective expansion of their external relational network scope and intensity through industry–academia–research synergistic mechanisms. Second, university–industry collaboration demonstrates a substantial positive influence on the slack resources of logistics enterprises, manifesting as incremental resource redundancy generated by knowledge spillover effects and technology-sharing mechanisms. Third, university–industry collaboration significantly and positively drives the dynamic capabilities of logistics enterprises, reflected in enhanced organizational learning and knowledge integration capabilities. Fourth, social capital exerts a significant and positive impact on green innovation in logistics enterprises. Fifth, slack resources exhibit a notable and positive effect on green innovation in logistics enterprises, providing strategic buffer space for innovation trial-and-error and resource reconfiguration. Sixth, dynamic capabilities significantly and positively contribute to green innovation in logistics enterprises. Seventh, social capital, slack resources, and dynamic capabilities partially mediate the relationship between university–industry collaboration and green innovation in logistics enterprises, forming a causal chain of “‘university–industry collaboration’–resource and capability development–green innovation in logistics enterprises”.

All hypotheses proposed in this study have been empirically validated, with the results summarized in

Table 12. Hypothesis test results.

Model	Hypotheses	Results
1	H1. University–industry collaboration has a positive impact on social capital.	Supported
2	H2: Social capital has a positive impact on green innovation in logistics enterprises.	Supported
3	H3: University–industry collaboration has a positive impact on slack resources.	Supported
4	H4: Slack resources have a positive impact on green innovation in logistics enterprises.	Supported
5	H5: University–industry collaboration has a positive impact on dynamic capabilities.	Supported
6	H6: Dynamic capabilities have a positive impact on green innovation in logistics enterprises.	Supported

. These findings elucidate the intrinsic mechanisms through which university–industry collaboration drives green innovation in logistics enterprises via resource integration and capability development pathways, thereby offering new empirical evidence for the application of industry–academia–research collaboration in the logistics sector.

(1) Empirical studies have demonstrated that university–industry collaboration significantly enhances the social capital of logistics companies (β = 0.471, p < 0.001). This finding suggests that engaging in such collaboration can foster the growth of social capital within logistics enterprises. When logistics enterprises collaborate with more universities, they not only gain more channels for acquiring resources and expand their knowledge search scope, but also help to build a broader social network, including connections with other enterprises, the government, and community organizations. This network not only facilitates the flow of information but also provides logistics enterprises with more opportunities for collaboration. Logistics enterprises that actively engage in university–industry collaboration are often able to enhance their sense of social responsibility and improve their brand image. This enhancement in brand image helps logistics enterprises attract more customers and investors, thereby increasing the enterprise’s market value. In addition, university–industry collaboration establishes a trust relationship between the enterprise and educational institutions, promoting long-term collaboration between both parties. This trust is not limited to the two cooperating entities but can also influence the trust between the logistics enterprise and other stakeholders.

(2) Social capital positively influences green innovation in logistics companies, with a significant effect (β = 0.774, p < 0.001). This result indicates that the social capital of logistics enterprises promotes their green innovation activities. Social capital not only increases trust and collaboration between organizations, promotes knowledge sharing among enterprises, and helps to obtain resources from partners, but also encourages green behavior within logistics enterprises. Through establishing trust relationships between enterprises and customers, suppliers, communities, and the government, social capital fosters cross-sector collaboration. This trust makes enterprises more willing to share resources and knowledge, thereby driving the implementation of green innovation projects. The knowledge-sharing platform provided by social capital enables enterprises to learn from and adopt successful green innovation experiences.

(3) University–industry collaboration significantly boosts the slack resources of logistics enterprises (β = 0.572, p < 0.001). On the one hand, universities cultivate professional talents for enterprises, particularly in fields such as logistics management and supply chain optimization, thereby enhancing the company’s talent pool. Conversely, university–industry collaboration promotes the sharing of equipment and facilities between the two parties, and the application of these technologies can improve operational efficiency, thereby releasing more surplus resources. Furthermore, university– industry collaboration is often supported by government funding, which alleviates the financial pressure on enterprises and increases surplus capital.

(4) The increase in slack resources within logistics companies significantly boosts their green innovation (β = 0.737, p < 0.001). This indicates that the slacker the resources a logistics enterprise possesses, the stronger its willingness to engage in green innovation. Slack resources are often used as additional resources to help enterprises experiment with new strategies and undertake more breakthrough innovation activities. The use of slack resources can reduce the financial investment required for green innovation, such as using idle equipment and materials for experiments and research and development, thereby reducing the risk of initial investment. Logistics companies can rapidly modify their production processes and product lines with slack resources to respond to shifts in market demand or regulations enforcing green standards. This flexibility can reduce risks associated with failing to adapt to market changes.

(5) University–industry collaboration has a significant positive impact on the dynamic capabilities of logistics enterprises (β = 0.487, p < 0.001). University–industry collaboration provides enterprises with opportunities to acquire the latest logistics technologies and management concepts. Through collaboration with universities, companies can quickly acquire advanced logistics knowledge and innovative techniques, improving their ability to adapt to market changes. Participation in university–industry collaboration helps cultivate talents that meet the needs of logistics enterprises. Logistics enterprises can directly participate in talent development through practical projects, internships, and other opportunities, ensuring that they acquire employees with relevant skills and knowledge, thereby improving organizational flexibility and adaptability. Additionally, university–industry collaboration promotes resource sharing, including information, technology, and market resources. This, in turn, enhances the operational efficiency and adaptability of logistics companies. Efficient coordination in logistics can reduce costs, shorten delivery times, and enhance the dynamic adaptability of enterprises.

(6) Dynamic capabilities have a strong positive effect on green innovation in logistics companies (β = 0.743, p < 0.001). The stronger the dynamic capabilities of logistics enterprises, the stronger their ability to perceive, acquire, and integrate resources, which in turn promotes green innovation. The higher the dynamic capabilities of logistics enterprises, the better they are at quickly identifying and responding to market demand changes for green logistics services. This flexibility allows enterprises to adjust their operating models, quickly introduce environmentally friendly technologies and processes, and meet customers’ expectations regarding sustainability and environmental protection. Dynamic capabilities enable logistics enterprises to effectively integrate internal and external resources, including collaboration with universities, research institutions, and other enterprises, to obtain advanced green technologies and management knowledge, thus driving the implementation of green innovation.

This study also confirms the partial mediating roles of social capital, slack resources, and dynamic capabilities in how university–industry collaboration influences green innovation in logistics companies. The findings suggest that such collaboration fosters green innovation by strengthening social capital, slack resources, and dynamic capabilities, affecting innovation from both resource and capability angles, with social capital having the greatest impact.

As environmental pollution issues become increasingly severe, green innovation has emerged as a core driver for enterprises’ low-carbon transformation, with its associated driving factors becoming a focal point of academic inquiry. However, existing research on green innovation predominantly concentrates on the manufacturing sector [125,126], while systematic investigations into green innovation within logistics enterprises remain inadequate. The extant literature largely adopts a case study approach [6,15,39]. Given the marked differences in industry characteristics, operational models, and value-creation logics between manufacturing and logistics, theoretical insights derived from manufacturing-focused green innovation studies cannot be directly transferred to the logistics context. This study, by expanding the theoretical framework of green innovation in logistics enterprises, offers a more industry-specific theoretical foundation for research in this domain.

Previous studies on green innovation have elucidated the driving mechanisms of corporate green innovation from dimensions such as stakeholder pressure [16], resource and capability endowments [17,18], strategic orientation choices [19], and executive traits [20]. Nonetheless, existing research pays relatively limited attention to university–industry collaboration, a critical factor. Within the paradigm of open innovation, university–industry collaboration serves not only as a core mechanism for restructuring national innovation systems and invigorating economic momentum but also as a significant potential driver for enterprises’ green transformation [22]. This study focuses on the driving effects and pathways of university–industry collaboration on green innovation in logistics enterprises, aiming to provide empirical evidence for deepening the theoretical construction of the relationship between university–industry collaboration and green innovation, while simultaneously expanding the practical boundaries of university–industry collaboration in the green development of logistics enterprises.

By dissecting the internal mechanisms through which university–industry collaboration influences green innovation in logistics enterprises, this study reveals the mediating roles of social capital and dynamic capabilities. Drawing on social capital theory and dynamic capability theory, it elucidates how university–industry collaboration drives green innovation in logistics enterprises through resource integration and capability reconfiguration. This research not only deepens the understanding of the relationship between university–industry collaboration and green innovation but also extends the application of social capital theory and dynamic capability theory in the field of green innovation within logistics enterprises, thereby offering a novel theoretical framework for studying green innovation in this sector.

5.2. Conclusions

Based on research into the driving mechanisms of university–industry collaboration on green innovation in logistics enterprises, this study proposes the following managerial implications from a management practice perspective to assist logistics firms in constructing green competitive advantages.

In the process of pursuing green innovation, logistics enterprises should deepen university–industry collaboration mechanisms, transcending the traditional model of singular technology transfer in industry–academia–research cooperation and instead adopting a multi-dimensional collaborative framework encompassing joint technological research and development programmed, co-cultivation of talent, and co-creation of brand value. Regarding technological research and development programs, it is recommended that enterprises establish joint green logistics laboratories with universities to conduct collaborative research on key technological domains such as carbon footprint tracking and new energy transportation equipment, facilitating the sharing of intellectual property rights. In terms of talent development, logistics enterprises can collaborate with universities to explore a “dual-tutor system” for talent cultivation, integrating theoretical learning at universities with practical engagement in corporate green projects to nurture interdisciplinary professionals proficient in both low-carbon operational thinking and digital management capabilities. With respect to brand value co-creation, enterprises are advised to convert academic collaboration outcomes into brand value assets through initiatives such as sponsoring green logistics innovation competitions at universities and jointly developing industry white papers, thereby leveraging the academic credibility of universities to enhance their appeal within the ESG domain.

Concurrently, it is suggested that logistics enterprises integrate social capital management into their green innovation strategic frameworks and reconstruct their social capital networks. Collaborating with universities and research institutes, logistics enterprises should proactively participate in the formulation of industry standards such as the Evaluation Criteria for Green Logistics Enterprises, elevating their nodal status within green supply chains, accessing cutting-edge information related to green logistics, and attracting like-minded green partners. Facing the dual challenges of accelerated green technology iteration and heightened policy environment uncertainty, logistics enterprises can enhance their dynamic capabilities through strengthened collaboration with universities. For instance, they may establish agile development teams in collaboration with universities to achieve rapid technological adaptation.

Furthermore, given the heterogeneity of logistics enterprises in terms of their resource endowments, technological foundations, and market positions, it is advisable for them to dynamically adjust university–industry collaboration strategies. For enterprises in the “technology follower” stage, priority should be given to participating in university-led research and development programs, rapidly acquiring mature green technologies through technology licensing and equipment procurement. For those in the “technology catch-up” stage, it is recommended to establish green technology pilot bases in conjunction with universities, leveraging their practical conditions to validate laboratory outcomes through engineering processes. For enterprises in the “technology leadership” stage, exploration of joint green logistics talent cultivation programs with universities is encouraged, fostering specialized professionals through mechanisms such as establishing green innovation funds and providing entrepreneurial mentorship services.

While this study offers an initial investigation into the effect of university–industry collaboration, social capital, slack resources, and dynamic capabilities on green innovation in logistics enterprises, and verifies the proposed research hypotheses and relational model, there are still some limitations. First, this study uses the entire logistics industry as the sample, overlooking discussions that focus on specific business sectors within the logistics industry. Second, the study only counts the number of universities collaborating with logistics enterprises, without discussing how the type of universities involved in the collaboration influences green innovation in logistics enterprises. Lastly, in practical business management, various factors influence the green innovation of logistics companies. This study primarily focuses on university–industry collaboration, without addressing other influencing elements. Future research could examine logistics enterprises within specific sub-sectors to explore green innovation across different types of companies. It could also investigate how collaboration with various types of universities impacts green innovation or approach the study of green innovation drivers from multiple angles to deepen and broaden the research.