2.1. Emergency industry innovation

The enhancement of national and regional emergency management capabilities must be supported by a robust emergency industry underpinned by a strong emergency technology system. In this regard, the academic community has conducted discussions on innovation in the emergency industry, including covering aspects such as the application of innovative technologies, the design of industrial policies and so on. Sandeep et al. studied the application of mobile communication technology in emergency disaster management, emphasizing its key role in helping disaster management personnel make decisions (2024) [14]. Martina et al.(2021) discussed the application of drone systems in emergency response scenarios, highlighting their efficiency and flexibility in disaster monitoring, material transportation, and search-and-rescue operations [15].In addition, digital twin technology (Enrico and Leonardo, 2024) [16]. IoT (Krichen,et al. 2024) [17]and the application of AI (Murthy et,al.,2025) [18] all contribute to the development of more effective and sustainable disaster management strategies. In summary, it is evident that the application of emergency innovation technologies is crucial for optimizing emergency safety management. In recent years, China has continuously increased its investment in emergency field research, and the industry as a whole has developed rapidly, with a wealth of research findings. In their studies, some scholars have pointed out that the development of China’s emergency industry should focus on emergency technology innovation (Guo Xiang, 2014) [19], as well as the construction of an emergency industry innovation system (Zhou, 2016; Xu et al., 2020) [20,21]. On the other hand, China’s industrial development is based on a combination of market mechanisms and government mechanisms, where the market mechanism plays a fundamental role and the government mechanism has a leading, regulatory, and supportive effect. Shen and Zhang (2024) discussed the transformation process of China’s emergency industry in depth, pointing out the cross-effect of government intervention, market dynamics, and social demand in the innovation and development process of the emergency industry; they suggested that the Chinese government should focus on social development needs and should guide the input of scientific and technological innovation in the emergency industry [22]. In addition, Yang and Jia (2023), based on the perspective of institutional grammar, used fuzzy set qualitative comparative analysis (fsQCA) to examine how policy design affects the innovation performance of China’s emergency industry; they found that policy design oriented by rules and standards could improve the innovation performance of China’s emergency industry, but that the lack of strategy-oriented policy design would lead to poor innovation performance [23].

2.2. Industrial collaborative innovation

The current scholarly research on industrial collaborative innovation mainly focuses on three aspects: connotation, impact effects, and mechanisms.

1. (1). The concept of industrial collaborative innovation can be analyzed from three dimensions: participants, process, and innovation network. In terms of participants, enterprises, universities, and research institutes are the core entities of industrial collaborative innovation. At the same time, other entities such as the government, financial institutions, and intermediary organizations may also be involved (Rajalo and Maaja, 2017) [10]. The government, as the policy maker and resource allocator, plays a crucial guiding and supporting role in collaborative innovation(Etzkowitz and Leydesdorff, 2000) [24]. From the process perspective, industrial collaborative innovation refers to a series of resource-sharing activities, such as technological cooperation, personnel exchange, information consultation, and achievement transformation, which are carried out through the division of labor and cooperation among related entities in the industrial chain (Jin et al., 2018) [25]. From the perspective of innovation networks, industry–university–research collaborative innovation is a collection of interactive relationships formed by multiple participating entities with networked and dynamic characteristics (Matteo et al., 2019) [26], and this networked structure facilitates the flow and diffusion of knowledge, thereby enhancing overall innovation efficiency (Powell et al., 1996) [27].
2. (2). Impact effects. Most studies affirm the positive impact of industrial collaborative cooperation on innovation performance. For example, Alessandra (2016) studied the impact of public-funded university collaborations with industry on corporate R&D efforts and found that participation in the project had a positive impact on R&D for each employee [28]. Using the triple helix theory, Liu et al. (2017) analyzed the positive effects of collaborative innovation carried out by governments, related enterprises, universities, and research institutions on product technology innovation and regional innovation [29]. Xie et al. (2023) explored the impact of two different types of collaborative innovation on innovation performance and found that supply chain (SC) collaborative innovation had a more significant effect on corporate innovation performance than innovation performance generated through industry–university–research cooperation (IUR) [30].
3. (3). The mechanisms of industrial collaborative innovation can be analyzed from three aspects: internal mechanisms, external mechanisms, and resource-sharing mechanisms. Zhou et al. (2013) proposed that internal incentives and profit drivers are the internal mechanisms of collaborative innovation, while government support is the external mechanism, and internal mechanisms emphasize mobilizing the motivation of all participating entities through benefit distribution and incentive mechanisms, while external mechanisms provide support for collaborative innovation through policy backing and resource investment [31]. Wang et al. (2021), using game theory, examined the optimal mechanism for profit distribution in the collaborative innovation process from a sustainability perspective, arguing that the optimal profit distribution for each enterprise should be directly proportional to its importance in the collaborative innovation project, which helps achieve the long-term stable development of collaborative innovation [32]. Yang et al. (2022) studied the impact of government subsidies on sustainable innovation in industry–university cooperation and found that subsidized industry–university collaborators generated more profit and greater social welfare [33]. In addition, knowledge and technology resource sharing is also a key mechanism for promoting collaborative innovation (Wu et al., 2019; Wang and Hu, 2020; Iqra et al., 2022) [34–36].

2.3. Evolutionary game models

John Maynard Smith and Price (1973) combined game theory with biological evolutionary theory to develop evolutionary game theory [37]. This theory breaks through the assumptions of perfect rationality and perfect information in traditional game theory, emphasizing the concept of bounded rationality, where participants typically cannot make optimal decisions at once but instead optimize their strategies through continuous trial and error and learning. It also emphasizes the long-term dynamic evolution process, analyzing how strategies spread, stabilize, or disappear within a group over time, thereby aligning more closely with the complexity and uncertainty of the real world [13]. Since then, evolutionary game theory has been applied to various disciplines, including economics, computer science, biology and management science(Arne and Nikoleta, 2023) [12]. As a result, Evolutionary game models have been widely applied in studying the innovation behaviors of industry players. By simulating the strategy adjustments and learning processes of participants in dynamic environments, they provide an effective analytical tool for studying the evolution of behaviors in complex systems (Mohammad, 2022; Zhang et al., 2023; Luo et al., 2023; Yuan and Li, 2024) [8,38–40]. For example, Zhang et al. (2023) constructed a tripartite evolutionary game model involving the government, enterprises, and the Energy Regulatory Service Center (ERSC) to explore the impact of government subsidies on the independent innovation of photovoltaic enterprises [38]. Luo et al. (2023) built a tripartite evolutionary game model aimed at agricultural enterprises, universities, and the government. From the perspective of innovation alliances, they examined the issue of low-carbon technology innovation in agriculture [8].

By reviewing relevant literature, it is found that existing studies primarily focus on the current state of innovation in the emergency industry, and the innovation mechanisms in this industry remain at a macro-theoretical level, lacking an in-depth exploration of the collaborative mechanisms between innovation entities from a micro perspective. Meanwhile, although there has been extensive discussion on industrial collaborative innovation mechanisms, existing research has not considered the specific characteristics of the emergency industry, particularly its economic attributes and market return uncertainties. Therefore, the findings of such studies may not effectively guide the practice of collaborative innovation in the emergency industry. Given this, it is both theoretically and practically necessary to conduct research on the micro-mechanisms of collaborative innovation in the emergency industry. Considering that the entities in the emergency industry (enterprises, academic and UR) are not completely rational in decision-making, but are influenced by information asymmetry and the uncertainties of the external environment, evolutionary game theory can better characterize such bounded rational behavior. Additionally, collaborative innovation in the emergency industry is a long-term, dynamic process involving strategic interactions and behavioral adjustments among multiple entities. Using evolutionary game theory can reveal the dynamic evolution of strategies among micro-level entities. Therefore, based on the characteristics of the emergency industry, this study constructs a multi-agent evolutionary game model involving the government, emergency enterprises, and UR to explore the pathways of collaborative innovation in the emergency industry. The innovations and marginal contributions of this study are as follows:

1. (1). This study explores the micro-mechanisms of collaborative innovation in the emergency industry based on evolutionary game theory. By analyzing the strategy choices and interactions between the government, emergency enterprises, and UR in the collaborative innovation process, it fills the gap in existing research at the micro level and expands the application of evolutionary game theory.
2. (2). This study breaks through the traditional analytical framework of industrial collaborative innovation by incorporating the economic attributes and market return uncertainties of the emergency industry into the research scope. It investigates the unique collaborative innovation pathways in the emergency industry. The findings not only reveal the intrinsic patterns of collaborative innovation in the emergency industry but also further enrich the theoretical understanding of industrial collaborative innovation.
3. (3). This study analyzes the collaborative innovation pathways of the emergency industry under the dual influence of government regulation and market mechanisms, clarifying the applicable boundaries of market mechanisms and government regulation in the collaborative innovation process. This study is of great practical significance, providing a reference for the government to develop targeted policies and contributing to the promotion of collaborative innovation in the emergency industry.

3.1. Problem description and model assumptions

In the collaborative innovation development of the emergency industry, three main entities are involved: emergency enterprises (E), university and research institutions (UR), and government departments (G). Emergency enterprises and UR each have their focuses but are interdependent in the process of collaborative innovation in the emergency industry; they promote the rapid development and technological advancement of the emergency industry through close cooperation and joint efforts. Emergency enterprises need to closely monitor market demands, clarify the actual needs for emergency equipment and services through market research and customer feedback, and provide a clear direction for research and development. Furthermore, they are responsible for transforming research and development results into actual products for market promotion and application, thereby gaining economic benefits from innovation investment. UR focus on basic research related to the emergency industry, including the exploration of disaster early warning models, emergency communication technologies, new materials, and other cutting-edge technologies. In addition, they also undertake the task of training professional talents for the emergency industry. Emergency enterprises (E) and UR establish close industry–university–research cooperation through the construction of joint research and development platforms and the joint declaration of projects, actively promoting the sharing of technology, talents, and resources; they carry out technology research and development and achievement transformation through a (cost-)sharing approach, thereby achieving collaborative innovation and development in the emergency industry. Government departments (G) provide financial subsidies to support industry–university–research collaborative research and development projects, or they provide market application space for collaborative innovation results through government procurement methods. In addition, government departments also intervene in the market demand side of emergency products and services, including promoting emergency safety knowledge and culture, establishing emergency safety knowledge popularization and education bases, and the promotion of emergency safety knowledge, to enhance public emergency awareness and accelerate the promotion and application of emergency innovation products.

Based on the above, a theoretical model of collaborative innovation in the emergency industry can be constructed, as shown in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Theoretical model of collaborative innovation in emergency industry.

https://doi.org/10.1371/journal.pone.0322563.g001

Fig 1 shows the relationships among the emergency enterprise (E), UR (U), and government department (G) in the process of collaborative innovation.

Assumption 1: All the entities have bounded rationality, and their behavioral strategies need to be repeatedly learned and adjusted to reach an optimal state. Emergency enterprises and UR have two behavioral strategies, namely Active Innovation and Passive Innovation; the government department’s strategy set is Regulation and Laissez-faire. The probabilities of emergency enterprises and UR choosing active innovation are and , respectively, and the probabilities of choosing passive innovation are and , respectively; the probabilities of the government department choosing regulation and laissez-faire are and , respectively. Here, .

Assumption 2: Regardless of whether emergency enterprises and UR choose active innovation, their basic benefits under normal operations are and , respectively; only when both parties choose active innovation can technological progress and product upgrades be successfully achieved, with the resulting market economic benefits being .Specifically, we incorporates the following characteristic factors of the emergency industry into the assumption of collaborative innovation benefits: First, the size of the collaborative innovation benefits is constrained by the economic attributes of the industry, the higher this economic benefit, the stronger the economic attributes of the industry (Yu and Zheng, 2023) [11]. Some emergency products (such as escape devices for earthquakes, aviation rescue equipment, etc.) have specialized uses, limited market sizes, small demand potential, and restricted growth space. Therefore, collaborative innovation in such emergency industries is unlikely to generate significant economic benefits under market mechanisms, reflecting weak economic attributes. Second, the occurrence of sudden safety events is highly uncertain. Emergency products often exhibit the nature of ‘needed urgently when required, but unnecessary when not required’, which results in considerable uncertainty in the market benefits of emergency products. On the other hand, public awareness of emergency safety triggers preventive behaviors when facing potential risks. This means that even if a disaster does not necessarily occur, individuals will still choose to purchase and stock emergency products in preparation for potential unforeseen dangers (Ehrlich and Becker, 1972) [41]. Clearly, such behavior reduces the uncertainty in market demand and benefits to some extent. To align the model with the specific characteristics of the emergency industry, Referencing the research of Liu et al. (2024) [42], the collaborative innovation benefits is assumed to be , where represents the economic attributes of the emergency industry, represents public emergency safety awareness and is a random variable representing innovation benefits, which follows a uniform distribution over the interval (Yi et al.,2023; Jing et al., 2020) [43,44]. Clearly, the larger the interval length , the broader the range of values for the random variable , meaning the higher the degree of uncertainty. Consequently, the expected benefits from collaborative innovation for industry players will be lower. For the convenience of model description, is defined as the degree of uncertainty in the emergency industry’s benefits. In order to ensure that the additional benefits from collaborative innovation is positive, it is assumed that .

Assumption 3: In the process of collaborative innovation in the emergency industry, emergency enterprises and UR actively promote the sharing of technology, talent, and resources and carry out technology research, development, and the transformation of achievements through a shared benefits and cost-sharing approach. Therefore, it is assumed that the total cost is , with the emergency enterprise bearing a proportion of and UR bearing a proportion of ; the government provides financial support, and the distribution ratios of the innovation benefits are and , respectively, with the emergency enterprise receiving a proportion of and UR receiving a proportion of . Due to the knowledge spillovers present in the collaborative innovation process, under conditions of information asymmetry, one party may engage in opportunistic behavior to serve its own interests by choosing passive innovation [45] (Qi and Li, 2022), thus obtaining spillover benefits. The spillover benefits for the emergency enterprise and UR are and , respectively. At the same time, the party choosing active innovation, in addition to the sunk costs of innovation investment, will also suffer additional economic losses. To simplify the model, it is assumed that the total loss for the party choosing active innovation is . Because the party choosing passive innovation breaches the cooperation agreement, it must pay a penalty when both parties choose passive innovation, additional losses and penalty payments are not considered (Guo et al., 2024) [46].

Assumption 4: When implementing regulatory measures, government departments will provide R&D subsidies to collaborative innovation projects involving emergency enterprises and UR, with the subsidy intensity denoted as . On the other hand, government departments will also create market application opportunities for collaborative innovation outcomes through centralized procurement, with the procurement intensity denoted as . it should be noted that when implementing two types of incentive-based regulatory measures mentioned above, the government will supervise the behavior of emergency enterprises and UR to prevent strategic behaviors such as falsification, which will incur corresponding regulatory costs, denoted as . Punishment will be imposed accordingly, resulting in losses for any party that chooses passive innovation. The punishment intensity is denoted as . In addition, the government enhances public awareness of emergency safety through the promotion of emergency safety knowledge culture, the construction of emergency safety knowledge popularization and education bases and other means. In addition, when the government does not take any measures to enhance public emergency awareness (e.g., promoting emergency safety knowledge and culture, establishing emergency safety knowledge popularization and education bases, etc.), . When such measures are implemented, (). Obviously, the government’s implementation of relevant measures will incur corresponding costs. Moreover, as public awareness of safety increases, it becomes more challenging for the government to further raise this awareness, leading to higher associated costs. Therefore, it is assumed that the cost, denoted as , is a quadratic function of emergency safety awareness, specifically ,where is the cost coefficient. In summary, under government regulation, when the emergency industry achieves high-quality development due to collaborative innovation, it will generate socio-economic benefits, enhance government reputation, and receive rewards from higher-level governments. The total of these benefits is recorded as (Zhang et al., 2023) [38].

3.2. Model construction

Under the assumptions mentioned above, a payoff matrix for the game among the three parties involved in the collaborative innovation process is constructed, as shown in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Payoff matrix.

https://doi.org/10.1371/journal.pone.0322563.t001

Table 1 shows the payoff matrix for the game among the emergency enterprises (E), UR (U), and government departments (G).

Based on the payoff matrix, the following can be derived

1. (1). Replicator Dynamic Equation for the Strategy Choice of the emergency enterprises

where

is the expected payoff for the emergency enterprise when it chooses active innovation;

is the expected payoff for the emergency enterprise when it chooses passive innovation;

is the average payoff for the emergency enterprises.

Thus, the replicator dynamic equation for the strategy choice of the emergency enterprises is

1. (2). Replicator Dynamic Equation for the Strategy Choice of UR

where

is the expected payoff for UR when it chooses active innovation;

is the expected payoff for UR when it chooses passive innovation;

is the average payoff for UR.

Thus, the replicator dynamic equation for the strategy choice of UR is

1. (3). Replicator Dynamic Equation for the Strategy Choice of Government Department

where

is the expected payoff for government department when it chooses regulation;

is the expected payoff for government department when it chooses laissez-faire;

is the average payoff for government departments.

Thus, the replicator dynamic equation for the strategy choice of government department is

3.3. Model solution and analysis

The replicator dynamic equations for the emergency enterprise, UR, and government department are set to 0, i.e., ,,. Solving these equations together can yield eight pure strategy evolutionary equilibrium points, which are ,,,,,,and . By taking the partial derivatives of the replicator dynamic equations ,,and with respect to , and , the Jacobian matrix can be obtained:

Table 2 shows the eigenvalues of the Jacobian matrix corresponding to each equilibrium point.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Eigenvalues of the Jacobian matrix for each equilibrium point.

https://doi.org/10.1371/journal.pone.0322563.t002

An equilibrium point is considered stable (an evolutionarily stable strategy, ESS) if and (Friedman, 1991) [47]. Fluctuations in the stability of equilibrium points due to different parameter values lead to corresponding changes in the evolutionary path of the evolutionary game model. From this, the following proposition can be derived.

Proposition 1 (1) When and , the strategy is evolutionarily stable strategy (ESS).

(2) When and , the strategy is evolutionarily stable strategy (ESS).

Proposition 1 indicates that, whether under market mechanisms or government regulation, emergency enterprises and UR may lose their enthusiasm for innovation investment, leading the collaborative innovation in the emergency industry to fall into a “negative stagnation” of low-end lock-in. The reason lies in the lack of effective supervision and penalty mechanisms during the collaborative innovation process, which fosters passive innovation behavior. In such cases, active investment by either party in the collaborative innovation process not only fails to increase their own returns but may also result in losses due to the opportunistic behavior of the other party. Therefore, passive innovation becomes the dominant strategy for emergency enterprises and UR. For government departments, if they excessively rely on incentive measures such as innovation cost subsidies and emergency product procurement to address the issue of low innovation economic efficiency caused by the weak economic attributes of the emergency industry, they may ultimately abandon regulation due to excessive fiscal cost pressures. Moreover, even if the government adopts a lower regulatory intensity to avoid high regulatory costs, it will neither help suppress opportunistic behavior in the collaborative innovation of the emergency industry nor change the passive innovation behavior of the industry entities, resulting in a situation of “regulatory failure.”

In summary, when the opportunistic behavior of internal entities in the emergency industry lacks constraints, coupled with insufficient or absent government regulation, collaborative innovation in the emergency industry will fall into a “stalemate” of negative stagnation. This situation often occurs during the initial stage of collaborative innovation development in the emergency industry. In this embryonic stage, the cooperative foundation between emergency enterprises and UR is relatively weak, lacking long-term and stable trust relationships. The absence of a trust mechanism increases the costs of cooperation in collaborative innovation, making it easier for opportunistic behavior to arise, thereby reducing the motivation for active participation by all parties. At the same time, the government’s initial interventions to promote collaborative innovation in the emergency industry may prove ineffective due to poorly designed policies or inadequate implementation.

Proposition 2 (1) When and , the strategy is evolutionarily stable strategy (ESS) if ; the strategy is evolutionarily stable strategy (ESS) if.

(2) When and , the strategy is evolutionarily stable strategy (ESS)if ; the strategy is evolutionarily stable strategy (ESS) if .

Proposition 2 indicates that, whether under market mechanisms or government regulation, even if the opportunistic behavior of internal entities in the emergency industry is effectively constrained, there may still be situations where only one party (either emergency enterprises or universities/research institutions) chooses to actively innovate, resulting in “unilateral enthusiasm” in the collaborative innovation of the emergency industry. The reason lies in the fact that, during the collaborative innovation process between emergency enterprises and UR, although the penalty mechanisms for passive innovation behavior are strengthened, the weak economic attributes of the emergency industry make it difficult to establish an effective market-driven revenue mechanism. In other words, active innovation still struggles to generate substantial returns, while entities adopting passive innovation strategies can still obtain certain speculative gains. Additionally, even if the government alleviates the issue of insufficient external returns for emergency industry innovation through measures such as procurement of emergency innovation products, subsidies for innovation costs, and emergency safety awareness campaigns, if internal conflicts such as unbalanced benefit distribution and unreasonable cost-sharing among industry entities remain unresolved, the party receiving lower returns is more likely to adopt short-term passive innovation strategies. Consequently, the government may abandon regulation due to its limited effectiveness and high costs. It is also worth noting that if the government adopts low-intensity regulation (e.g., loose supervision mechanisms or ambiguous policy guidance) to alleviate fiscal pressure, the emergency industry will not escape the “unilateral enthusiasm” dilemma in collaborative innovation. Surprisingly, the government will not abandon regulation in such cases. This implies that even if the “unilateral enthusiasm” dilemma in the collaborative innovation of the emergency industry cannot be effectively resolved, the government will still repeatedly invest resources, which clearly leads to the waste of administrative resources.

From Propositions 1 and 2, it can be concluded that achieving collaborative innovation in the emergency industry faces two main challenges: First, the weak economic attributes of the emergency industry result in limited market-driven returns, which is the root cause of insufficient external incentives for collaborative innovation. Second, there are inherent conflicts among internal industry entities (emergency enterprises and UR), including the emergence of opportunistic behavior, imbalances in benefit distribution mechanisms, and unreasonable cost-sharing rules. These issues further undermine the stability and sustainability of collaborative innovation. Therefore, whether and how the government can implement effective regulation is key to breaking the “low-enthusiasm equilibrium trap” in the collaborative innovation of the emergency industry. The following propositions will further explore this issue.

Proposition 3 (1) When , and , the strategy is evolutionarily stable strategy (ESS).

(2) When , and , the strategy is evolutionarily stable strategy (ESS).

Proposition 3 indicates that the feasible pathways for collaborative innovation in the emergency industry. Specifically, for emergency industries with weak economic attributes, the government regulation-driven pathway for collaborative innovation is significantly viable. This is because such emergency industries often face issues such as insufficient market demand, unstable investment returns, and high risks in technological innovation, making it difficult for market mechanisms to effectively drive their innovation and development. To address this, the government can promote collaborative innovation in the emergency industry through the following two measures: On one hand, the government can enhance public and institutional awareness and demand for emergency products and services, while utilizing government procurement channels to establish a regular mechanism for the storage and renewal of emergency supplies, thereby creating stable market demand. This measure can effectively increase the external returns of collaborative innovation in the emergency industry, reduce the negative impacts of market uncertainty, and provide a foundation for sustainable development. On the other hand, the government can directly reduce the R&D costs of enterprises through policy tools such as innovation subsidies, alleviating the financial pressure on enterprises during the initial stages of technological development. Such support can incentivize enterprises to engage in high-risk, high-investment innovation activities, driving technological advancement and industrial upgrading in the emergency industry.

Additionally, it has been found that if the emergency industry has strong economic attributes, even if the government does not implement regulatory measures, emergency enterprises and UR may still choose to actively innovate during the collaborative innovation process. This indicates that the market mechanism-driven pathway for collaborative innovation is feasible. However, it is important to note that for emergency industries with strong economic attributes, although their inherent market mechanisms have the potential to drive innovation, this potential has not yet been fully realized at the current stage. The main reason lies in the relatively weak emergency awareness among the general public, leading to a lack of sensitivity and proactiveness in the demand for emergency products and services. Taking household emergency products as an example, despite their enormous market potential, the general public’s awareness of emergency safety is generally low, and their understanding of emergency products and services is insufficient. As a result, such products are often perceived as “non-essential” or “low-frequency demand”, leading to weak consumer perception of the upgraded and innovative features of emergency technologies or products. This consumer psychology results in fragmented and short-term market demand, making it difficult to form stable market benefits and thereby constraining the innovation efficiency of the emergency industry. Therefore, for emergency industries with strong economic attributes, the government needs to carefully determine the timing for introducing public resource elements.

The above conclusions provide the following managerial implications: When promoting the development of the emergency industry, the government should formulate differentiated policies based on the economic attributes of different sub-sectors and implement categorized management of emergency products. For emergency sectors with weak economic attributes (e.g., public safety, disaster rescue, etc.), where the social benefits are significant but market-driven incentives are insufficient, the government should remain the main driver, ensuring development through fiscal support, government procurement, and other means, while using market mechanisms as a supplementary force. For emergency industries with strong economic attributes (e.g., household emergency equipment, emergency technical services, etc.), initial development should focus on enhancing public emergency awareness through multi-level interventions (e.g., promoting emergency knowledge through public awareness campaigns, community training, and school education), thereby guiding market demand towards scaling up, strengthening market-driven mechanisms, and improving the external economic benefits of innovation in emergency products, services, and technologies. In the mid-stage of development, both the government and the market should collaborate to further enhance the innovation motivation of the key players in the emergency industry. By the later stages of industrial development, as market mechanisms become more mature, the external economic benefits of innovation in the emergency industry will significantly increase, and industry players will possess intrinsic innovation drivers. At this point, the government should gradually withdraw from market regulation and promote the market-oriented development of emergency industry innovation, achieving self-organization and sustainable development of the industry.

Proposition 4 (1) When the strategy is evolutionarily stable strategy (ESS), the prisoner’s dilemma arises if and .

(2) When the strategy is evolutionarily stable strategy (ESS), the prisoner’s dilemma arises if and

Proposition 4 indicates that the potential ‘Prisoner’s Dilemma’ that may arise during the process of collaborative innovation in the emergency industry. Specifically, when the economic attributes of emergency products are weak, or when the government’s penalties for passive innovation behavior under R&D subsidy policies, the procurement of emergency innovation products, and the intensity of emergency safety awareness campaigns are relatively low, emergency enterprises and academic research institutions may still opt for active innovation if they are able to impose sufficiently high internal penalties on passive innovation behavior. However, in this scenario, although both parties will choose active innovation, their returns will not exceed the level of returns that would have resulted from both parties choosing passive innovation simultaneously, leading to a suboptimal equilibrium resembling the Prisoner’s Dilemma. The implication for reality is that, when the government does not implement regulatory measures or the regulatory efforts are weak, relying solely on the internal penalty mechanisms between industry stakeholders to constrain passive innovation behavior may result in a ‘Prisoner’s Dilemma’ situation during collaborative innovation in the emergency industry. Furthermore, this is more likely to occur in emergency industries with weak economic attributes than in those with strong economic attributes. This also provides further evidence of the necessity of government regulation in promoting collaborative innovation in emergency industries with weak economic attributes. Interestingly, the amounts of R&D subsidies do not significantly affect the above results. What does matter is the government’s punishment for passive innovation behavior under the R&D subsidy policy. This is because R&D subsidies are typically based on ex-ante commitments, meaning that whether or not emergency enterprises and academic institutions actively engage in collaborative innovation does not affect their eligibility for receiving subsidies. The above results suggest the existence of an incentive misalignment problem in policy design: R&D subsidies fail to effectively translate into innovation motivation, while the penalty mechanism becomes a key variable in regulating the behavior of both parties. To resolve this dilemma, it is necessary to reconsider the design logic of policy tools, strengthen dynamic incentive mechanisms and ex-post evaluation systems to substantially enhance the efficiency of collaborative innovation.

5.1. Conclusions

1. (1). The economic attributes of the emergency industry and the uncertainty of market returns are key factors that constrain the collaborative innovation and development of the emergency industry. The economic attributes of the emergency industry determine the applicable boundaries of government regulation and market mechanisms in the collaborative innovation process of the emergency industry. Due to the “marketization paradox” in the emergency industry with weak economic attributes, and the fact that emergency enterprises and UR are more likely to face the “prisoner’s dilemma” in the process of collaborative innovation, collaborative innovation via the government regulation-driven pathway is more feasible. As for emergency industries with strong economic attributes, a market mechanism-driven collaborative innovation path becomes feasible under the condition of high public emergency awareness. This is because the uncertainty of market returns in the emergency sector suppresses the enthusiasm of emergency enterprises and UR to participate in collaborative innovation, while the improvement of public awareness can, to some extent, stabilize market returns and reduce investment risks in collaborative innovation. Therefore, the government should strengthen the market-driven revenue mechanism for emergency industries with strong economic attributes by enhancing public emergency awareness, thereby promoting market mechanism-oriented collaborative innovation development. This approach not only helps alleviate the fiscal pressure of government regulation but also enables the government to better concentrate resources and adopt more targeted regulatory measures for collaborative innovation in emergency industries with weak economic attributes. Existing research on industrial collaborative innovation often separately discusses how government regulatory measures or market mechanisms influence the innovation behavior of industrial entities (Hao et al., 2022; Wang et al., 2021; Zhang et al., 2025;) [7,50,51], without integrating government regulation and market mechanisms into a unified framework to address the applicability boundaries of the two in the industrial collaborative innovation process. Rooted in the characteristics of the emergency industry, this study explores this issue and proposes new perspectives.
2. (2). Regarding the effectiveness of government regulatory measures in promoting collaborative innovation in the emergency industry, we found that government procurement is more effective than subsidy policies. The reason is that while subsidy policies can alleviate some financial pressure, they cannot fully eliminate the risks brought by market uncertainty. Moreover, innovation subsidy policy has a “double-edged sword” effect. Although it can reduce the innovation investment costs for emergency enterprises and UR, thereby playing a positive role in promoting collaborative innovation in the emergency industry under certain conditions, excessively more subsidies can also produce adverse effects, undermining the enthusiasm of emergency enterprises and UR to participate in collaborative innovation. Additionally, increasing the penalties for passive behavior by emergency enterprises and UR in the collaborative innovation process can to some extent avoid the potential effect bias of subsidy policy. This is consistent with the research conclusions of Li ang Gao (2022) [52]. Our study also revealed that government procurement can help address the “prisoner’s dilemma” that may arise during the collaborative innovation process of emergency industries, but innovation subsidy policies have no effect on this. This finding further enriches the understanding of the effects of incentive measures (subsidy policies, government procurement) and punitive measures for behavioral constraints in industrial innovation.
3. (3). Relying solely on penalty mechanisms among internal industry entities to curb passive innovation behavior, emergency enterprises and UR are likely to fall into a “prisoner’s dilemma” during the collaborative innovation process. Moreover, the weaker the economic attributes of the emergency industry, the higher the risk of facing this dilemma. Establishing a reasonable benefit distribution mechanism can not only stimulate the enthusiasm of emergency enterprises and UR to participate in the collaborative innovation process but also enhance the government’s willingness to adopt regulatory measures, thereby creating a win-win situation for emergency enterprises, UR and government departments. This result indicates that collaborative innovation in the emergency industry is a complex systemic project involving multiple stakeholders, including emergency enterprises, UR and government departments. Not only can external policy environments directly influence the behavior of internal industry entities (emergency enterprises and UR), but the internal relationships among these entities can also indirectly affect the implementation of external policies. Therefore, in promoting collaborative innovation in the emergency industry, it is essential to fully consider the interest relationships among internal industry entities as well as external policy environment factors to achieve sustainable development of collaborative innovation.

5.2. Implications

5.3. Limitations

Despite the encouraging results, this study also had the following limitations. First, in order to derive clear insights, we assume that collaborative innovation can always lead to more market returns. However, innovation may face failure. Taking the risks into consideration in the future will enrich the results of this study. Second, this study revealed the key factors and intrinsic mechanisms affecting collaborative innovation pathways from theoretical perspective by constructing an evolutionary game model. Collecting research data for empirical study about the collaborative innovation in the emergency industry is of great importance in guiding significant practices.