Abstract
This paper simulates the behavior mechanism of each subject in the equipment manufacturing industry-university-research (IUR) collaborative innovation system based on NetLogo, and constructs the input-output model of each subject. Secondly, the reasonable value of each parameter is determined through expert consultation, and finally 6 scenarios are selected to simulate the evolution of the IUR collaborative innovation system. The results show that different decision-making behaviors of each subject will lead to different operating states of the innovation collaborative system. When the conditions of each subject are exactly the same, the innovation system can achieve stable and continuous operation. However, if the initial conditions are different, the IUR collaborative innovation system needs an authoritative intermediary to coordinate the behavioral strategies of each subject, strengthen the willingness to cooperate, and ensure its stable operation. Reasonable reward and punishment measures are also an effective and important measure, which can mobilize the enthusiasm of all subjects for collaborative innovation. This paper is of great significance for the IUR collaborative innovation in equipment manufacturing industry.
1. Introduction
In recent years, China has been in a critical period of economic transformation and high-quality development. Its technological innovation system has evolved from the initial “IUR cooperation innovation model” to the current “IUR in-depth integration innovation model,” which shows that IUR collaborative innovation plays a vital role in promoting national construction and efficient economic and social development. The equipment manufacturing industry is an important pillar industry in China, and the improvement of IUR collaborative innovation capabilities will promote the growth of the high-end equipment manufacturing industry.
Over the past few decades, China has made outstanding achievements in independent innovation, along with its economy and technology. Its R&D investment, scientific research achievements, and the amount of patent grants reached the international leading level. However, the integration rate of scientific research achievements and the economy is still relatively low. The conversion rate of patented technology is only approximately 10%, which is far lower than the 40% converted in developed countries, while that of scientific research achievements in universities is particularly low: only 5% form strong technical support for economic development [1]. Above all, the role of technology as the primary productive force has not been fully utilized.
The collaborative innovation of IUR serves as the in-depth cooperation and resource integration link among knowledge and technological innovation subjects—with universities, enterprises, and research institutions as the core elements, and with government, financial institutions, intermediary organizations, innovation platforms, and nonprofit organizations as auxiliary elements. This is an important mechanism design and development mode that promotes technological innovation and achievement transformation [2]. Therefore, the IUR collaborative innovation is a complex system with multiple agents, cross-organizations, and multiple elements. The decision-making behaviour of each subject is the key to determining whether it can exert the advantages of the subject and form synergistic effects. In conclusion, based on the social experiment method, this paper explores the impact of the decision-making behaviour of each subject on IUR collaborative innovation to provide valuable research conclusions and countermeasures for IUR collaborative innovation in the equipment manufacturing industry.
Based on system theory, this paper employs the NetLogo platform to construct an artificial society that basically conforms to the actual situation. It uses computer technology to study the evolution mechanism of equipment manufacturing IUR collaborative innovation, the interaction between the subject and the environment, and the dynamic behaviour of each subject. Through the microlevel analysis of subjects, behaviours and their interactions, this paper studies the evolutionary process of the overall state change of the IUR collaborative innovation system and further reveals the evolutionary law of the equipment manufacturing IUR collaborative innovation system.
2. Literature Review
The equipment manufacturing industry is a knowledge-intensive and technology-intensive industry with the characteristics of strong technological innovation, high industrial added value, and a high-end value chain. Currently, most scholars mainly focus on the network evolution and temporal and spatial distribution characteristics of equipment manufacturing IUR cooperation. Lu Guoqing has discovered that the IUR innovation system architecture of the equipment manufacturing industry in the Yangtze River Delta is at an early stage, and its evolution has obvious periodic characteristics. The members with higher centrality were mostly universities. The spatial differentiation characteristics of the IUR in each city in the region were obvious. Geographical proximity, similar innovation policies, and the same size of knowledge resources are important factors that influence innovation entities to establish innovative cooperation links. Zhao Shuang points out that the scale of China’s equipment manufacturing IUR cooperation network has increased, but member connections are not high, and the network centrality and network central potential have not truly formed [3]. In addition, some scholars are committed to exploring the organizational model of high-end equipment manufacturing technology innovation alliances and trying to build an evaluation index system for IUR innovation capabilities [4]. Furthermore, some studies discussed how to improve and perfect the IUR system of China’s equipment manufacturing industry from the aspects of fiscal and taxation policies, system innovation, strategic management, and foreign successful experience. That is, scholars at home and abroad are still exploring the research methods, theories and perspectives of equipment manufacturing IUR collaboration, while these studies provide important theoretical support, methodological references, and perspective innovation for this paper.
Collaborative innovation is a process of system optimization and cooperative innovation of each innovation subject; thus, collaborative innovation can be analysed from the two dimensions of integration and interaction [5]. The collaborative innovation system needs to integrate the advantages of all subjects in IUR innovation and establish a platform and mechanism that is conducive to R&D. Simultaneously, it is a process that requires scientists and entrepreneurs to interact and universities and enterprises to participate in the development of new technologies [6]. Essentially, IUR collaborative innovation is a process of learning organization innovation and tacit knowledge sharing, accompanied by collaborative innovation [7]. That is, the process of knowledge acquisition, transfer, application and feedback needs to go through three processes: knowledge sharing, knowledge creation and the formation of knowledge advantages [8]. At the same time, it is necessary to grasp the fluctuation law of the innovation process, clarify the nonlinear effect of the system, strengthen the openness of the system and improve the absorptive capacity of the enterprise [9]. Consequently, the IUR collaborative innovation system has the characteristics of high complexity, low stability, and unstable scale of collaborative innovation, and the key to the choice of each subject’s behaviour is the choice of organizational model.
According to the organization level and closeness of IUR collaborative innovation, Wang Zhangbao et al. [10] divide IUR innovation into five organizational models: project type, construction type, entity type, alliance type and virtual type. However, the IUR collaborative innovation system established only through contracts is subject to changes in the external environment. He Yubing and Zhang Yingchun [12] embed the networking model in the IUR collaborative innovation system, pointing out that members are not only restricted by contractual relationships but also affected by various “embedded” relationships, and divide the IUR collaborative innovation model into weak relationship-sparse networks, strong relationship-sparse network, strong relationship-dense networks and weak relationship-dense networks. Due to the diversity of organizational models, the classification criteria are not unique, and the stability of the IUR collaborative innovation system was a dynamic equilibrium. Wu et al. [11] believed that the stable state was determined by the evolutionary path of the collaborative innovation behaviour of various innovative subjects. When studying its evolution path, Cheng and Shi [12] believe that the IUR collaborative innovation system should be open and nonequilibrium, and the evolutionary path is phase transition and bifurcation. Phase transition refers to the reorganization at the microlevel to form the change of the macro level state; bifurcation means that when the coevolution process reaches a critical state, the IUR collaborative innovation system will have a stable branch and unstable branches.
Zoogah and Peng [13] consider that the microskills and abilities of strategic alliance managers are the key to maintaining alliance cooperation, and their decision-making can strongly determine the stability of the alliance. Li and Yuan [14] studied the specific strategies of government-industry-university-research collaborative innovation to effectively act on microsubjects. However, when constructing and evolving the IUR collaborative innovation system, we have to consider the barriers and trust issues between the subjects. Bruneel et al. [15] pointed out that there are two barriers between universities and enterprises. One is the different positioning of universities and enterprises, and the other is transaction barriers, such as the ownership of intellectual property rights. Furthermore, Ran et al. [16] proposed an improvement plan for selecting the best partner university in the process of enterprise technological innovation. Xue Kelei and Pan Yu [19] use the evolutionary game method to analyse the trust relationship between IUR collaborative innovation subjects and show that the trust relationship of IUR collaborative innovation will affect the efficiency of collaborative innovation. Similarly, Li and Liu [20] also use the evolutionary game method to study the behavioural decisions of subjects in the IUR collaborative innovation and point out that the evolution direction of the IUR collaborative innovation depends on the parameter values and initial state of the game matrix.
In summary, it can be seen that the stable operation of the IUR collaborative innovation system largely depends on the behavioural strategies of each subject. Although many scholars have begun to study this topic from the macro and microaspects, the research on the collaborative innovation subjects mostly uses the evolutionary game method; in the process of the game, scholars have not regarded the collaborative innovation of IUR as an organic whole.
3. Model
3.1. System Components
The IUR collaborative innovation of the equipment manufacturing industry is complex and can be divided into three parts: an innovation subject subsystem, innovation resource subsystem and an innovation environment subsystem [21]. As shown in Figure 1, the innovation subject subsystem refers to the realization of mutual coordination and functional connection between subjects with universities, enterprises and scientific research institutions as the main participants. The innovation resource subsystem refers to resource elements, including input elements such as talent, funds, knowledge, materials, equipment, information, and venues, as well as output elements such as innovative patents, new products, new knowledge, scientific research papers, scientific research projects, new talent, and social reputation. The innovation environmental subsystem mainly involves the external environment, for example, political environment, economic environment, talent environment, and market environment. Figure 1 is right here.
3.2. Theoretical Hypothesis
Hypothesis 1. To more accurately calculate the input of each subject to the equipment manufacturing IUR collaborative innovation system, we introduce an efficiency amplification factor to compare the input efficiency of each subject, and work on the following assumptions: ① the input of universities and research institutions is mainly talent and knowledge. As talent invested by universities and scientific research institutions has more professional knowledge, the talent efficiency amplification factor should be greater than 1, while that invested by enterprises should be less than 1. The input of universities is mostly basic knowledge, which is not as efficient as enterprises and scientific research institutions. That is, the knowledge efficiency amplification factor of universities should be less than 1, and that of enterprises and scientific research institutions should be greater than 1 that. ② For various reasons, the funds invested or obtained by each subject are not fully used for R&D, and its capital efficiency amplification factor is less than 1.
Hypothesis 2. Since knowledge innovation is relatively complex, its diffusion, dissemination and transformation require a certain amount of time [22, 23]; therefore, it can be considered that the knowledge invested in the short term is unchanged.
Hypothesis 3. Universities and scientific research institutions seldom consider economic benefits in equipment manufacturing IUR collaborative innovation and pay more attention to the output of scientific research papers, followed by patent output and the least attention to new products.
Hypothesis 4. Enterprises consider economic benefits more in collaborative innovation and pay attention to new products, followed by patents, and scientific research papers.
Hypothesis 5. In independent innovation, universities and scientific research institutions pay attention only to the output of scientific research papers and patents rather than the output of new products.
Hypothesis 6. In independent innovation, companies only focus on the output of new products and patents, not the output of scientific research papers.
3.3. Model Composition Analysis
3.3.1. Input Analysis
The input elements of the equipment manufacturing IUR collaborative innovation mainly involve talents, funds, knowledge, materials, equipment, information, venues, and elements such as materials, equipment information and venues can all be obtained through money. Therefore, this paper assumes three categories as talent, capital, and knowledge, where represents talent, capital, and knowledge. Talent consists of talent invested by universities , that of enterprises , and that of scientific research institutions . Capital refers to capital invested by universities , enterprises and scientific research institutions . Knowledge involves knowledge invested by universities , enterprises , and scientific research institutions .
The talent input of the equipment manufacturing IUR collaborative innovation system is written as follows:
The capital input is as follows:
The knowledge input of the equipment manufacturing IUR collaborative innovation system is written as follows:
The talent efficiency amplification factor of universities is , that of enterprises , and that of scientific research institutions .
The capital efficiency amplification factor of universities is , that of capital enterprises , and that of scientific research institutions .
The knowledge efficiency amplification factor of universities is , that of enterprises , and that of scientific research institutions . From Hypothesis 1, we know ; ; ; ; ; ; ; ; .
3.3.2. Output Analysis
Similar to input analysis, the output is also diverse, including patents, new products, new knowledge, scientific research papers, scientific research projects, new talents, social reputation, etc., in which patents, new products and scientific research papers are the key. Therefore, this study assumes that the output of collaborative innovation is divided into three categories: patents, new products, and scientific research papers, where patents represent , new products , and scientific research papers .
To construct the output model of the equipment manufacturing IUR collaborative innovation, the study adopts the Cobb–Douglas production function, and the general form is written as follows:where represents technical level, labour input, capital input, elasticity of labour input, elasticity coefficient of capital input, and random interference.
According to Hypothesis 2, the patent output of the equipment manufacturing IUR collaborative innovation system is written as follows:where , , and are undetermined parameters. On the basis of the Cobb–Douglas production function, we can obtain:
The new products output is as follows:where , , and are undetermined parameters. Similarly, we can get:
The scientific paper output is written as follows:where , and are undetermined parameters. Similarly, we can obtain:
3.3.3. Analysis of Expected Output of Collaborative Innovation of Each Subject
The general expectation of universities in collaborative innovation is written as follows:
The degree of attention of universities to patents is , that to new products and that to scientific research papers is . In line with 3, we can determine that . The benefit sharing of universities for patents is , that for new products , and that for scientific research papers .
The general expectation of enterprises in collaborative innovation is written as follows:
The degree of attention of the company to patents is , that to new products , and that to scientific research papers . According to Hypothesis 4, we can obtain . The degree of attention of the company for patents is , that for new products , and that for scientific research papers .
The general expectations of scientific research institutions in collaborative innovation are represented by equation (13):
The degree of attention of scientific research institutions for patents is , that for new products , and that for scientific research papers . In light of Hypothesis 3, it is clear that . Meanwhile, the degree of attention of scientific research institutions for patents is , that for new products , and that for scientific research papers .
3.3.4. The Expected Output of Each Subject in Independent Innovation
The equation of patent output of universities in independent innovation is written as follows:
, and are undetermined parameters.
Scientific research papers Output in universities is denoted as follows:
, and are undetermined parameters.
If the degree of attention of universities to patents is and that to scientific research papers is . . The output equation of universities for patents and scientific research papers when the university is independently innovating is denoted as follows:
In line with Hypothesis 5, universities pay attention only to the output of scientific research papers and patents. Therefore, the overall expectation of universities in independent innovation is presented as follows:
Similarly, the patent output of enterprises in independent innovation is written as follows:
, and are undetermined parameters.
The new product output of enterprises in independent innovation is written as follows:
, and are undetermined parameters.
If company’s degree of attention to patents is and that to scientific research papers is , then it is clear that . The expectation of enterprises for patents and new product output in independent innovation is as follows:
According to Hypothesis 6, the general expectation of the enterprise in independent innovation is denoted as follows:
The patent output of scientific research institutions in independent innovation is as follows:
, and are undetermined parameters.
The output of scientific research institutions for scientific research papers in independent innovation is written as follows:
, , are undetermined parameters.
If the attention of scientific research institutions for patents is , and that for scientific research papers , we can obtain . According to Hypothesis 6, the expected output of scientific research institutions is written as follows:
3.4. Judgement Rule
The mathematical expectation of universities in collaborative innovation can be expressed as and that in independent innovation . The difference between and can be used to judge the attitude of universities towards collaborative innovation and independent innovation. To analyse this impact more accurately, a new variable is introduced: the current strength of universities’ willingness to cooperate . can be calculated as the relative value of the difference between and .
Furthermore, to explore how changes at different moments, it supposes is the changing situation of universities’ willingness to cooperate at two different moments.
Obviously, can directly determine the changing situation in universities’ addition or reduction of talent, capital, and knowledge input at the next moment. The specific logic flow is shown in Figure 2.
The judgement of enterprises and scientific research institutions is similar to that of universities, so there is no need to repeat them. Figure 2 is right here.
4. Simulation Analysis
4.1. Determination of Model Parameters
To determine the reasonable value of each parameter, we adopt the Delphi method. In the process, a total of 42 experts, scholars, R&D personnel and business management personnel participated in the consultation. Thirty-two valid expert consultation questionnaires were received, including 8 professors (professor-level senior engineers), 14 associate professors (senior engineers), 6 lecturers (engineers), and 2 doctoral students and 2 master’s students. According to expert advice, statistical parameters and the average values of weights are shown in Table 1, which basically meets expectations. Table 1 is right here.
4.2. Evolution Analysis
We adopt NetLogo 6.0.4 to construct a simulation environment for the collaborative innovation of IURs in the equipment manufacturing industry, and analyse the system evolution of each subject under different conditions. In practice, the types, quantities, benefit sharing types, and conditions of input elements of each subject are extremely different. For the convenience of research and analysis, this research will assume the following scenarios for simulation.
4.2.1. Scenario I
First, in scenario I, the inputs and benefits of each subject in collaborative innovation are the same. We perform a system simulation on scenario 1, and the simulation results are shown in Figure 3. In the figure, yu0, ye0, and yr0 represent the strength of the initial cooperation willingness of universities, enterprises and scientific research institutions, respectively. Under the same conditions in all aspects, the strength of the initial cooperation willingness of the industry, university and research leaders is basically the same. Moreover, the system evolution diagram Y represents the change in the intensity of cooperation willingness of each subject, and the strength of cooperation willingness of each subject remains basically stable. System evolution diagram C represents the output of the collaborative system, where C1, C2, and C3 represent the output of patents, new products, and scientific research papers respectively. In addition Figure 3 shows that under the same initial conditions and equal benefits sharing, the output of collaborative innovation remained basically stable. As shown in Figure 3, qu, qe, and qr represent the expected innovation benefits of universities, enterprises, and research institutes in collaborative innovation respectively, and quu, qee, and qrr represent the expected innovation benefits of universities, enterprises, and scientific research institutions in independent innovation conditions respectively.
In scenario I, the expected benefits of collaborative innovation of each subject are always higher than those of independent innovation. The equipment manufacturing IUR collaborative innovation system can maintain continuous and stable operation, and all entities will not exit the system. Figure 3 is right here.
4.2.2. Scenario II
In practice, the asymmetry of external information may lead to the uneven distribution of benefits between the entities of IUR collaborative innovation under the same initial investment conditions. Scenario II is that the initial conditions of universities, enterprises, and scientific research institutions are the same, but the benefits of scientific research institutions are greater than those of universities and enterprises. The evolution result is shown in Figure 4. In this case, the intensity of the initial cooperation willingness of scientific research institutions is much higher than that of universities and enterprises. Moreover, during the cooperation period, the intensity of cooperation willingness of scientific research institutions is basically greater than that of universities and enterprises, while that of universities and enterprises is basically the same. The output of innovation continued to rise at the beginning, and as cooperation deepened, the output of innovation began to gradually decline. The analysis shows that at the time of initial cooperation, that of collaborative innovation between universities and enterprises was greater than the benefits of independent innovation. However, after several cooperation, all subjects gradually determined the operation of collaborative innovation. The benefits of collaborative innovation between universities and enterprises continue to decrease until they are the same as those of independent innovation, while the benefits of scientific research institutions in collaborative innovation are always higher than those of independent innovation.
Therefore, universities and enterprises choose to withdraw from the collaborative innovation relationship, leading to the failure of the IUR collaborative innovation system. Although the asymmetry of external information has contributed to the formation of the IUR collaborative innovation system, all subjects choose to withdraw when their benefits are not guaranteed. Figure 4 is right here.
It can be seen from the above that it is difficult for the various subjects of the equipment manufacturing industry to achieve the same level of initial investment and technical level. Therefore, we have to consider different initial investment or different technical levels. To have a deeper understanding of the influence of the subjects’ willingness to cooperate, we need to discuss the analysis of the equal and unequal distribution of benefits under the conditions of different initial investments or different technical levels.
4.2.3. Scenario III
Scenario III is that the initial investment of the enterprise is greater than that of universities and scientific research institutions, and the subjects have the same technical level and the benefits. The evolutionary result is shown in Figure 5. In this case, the initial cooperation of enterprises is extremely low, while that of universities and research institutions is much higher. As the system continues to operate, the intensity of the company’s willingness to cooperate has fluctuated greatly. Although the initial investment of enterprises is relatively high, the investment of the enterprise will be gradually reduced to ensure that their benefits are not lost. Similarly, universities and scientific research institutions will gradually reduce their investment to prevent losses, which leads to a trend of first rising and then falling in the output of system collaborative innovation in the initial stage of cooperation. Eventually, the benefits of collaborative innovation between the IUR subjects of the equipment manufacturing industry are the same as those of independent innovation. Each subject chooses to withdraw and the system collapses. Figure 5 is right here.
4.2.4. Scenario IV
Based on the above simulation, it can be seen that in scenario III, a certain subject in the system can easily “free ride”. Therefore, we change the initial conditions, and conduct simulations under the condition that the technical level of each subject is the same, the initial investment of the enterprise is higher than that of scientific research institutions and universities, and the benefits are divided according to the proportion of the initial investment. The evolutionary result is shown in Figure 6. In this case, the differences in the initial cooperation intentions of each subject are very small, but the cooperation intentions of scientific research institutions are stronger. When the system began to operate, the input of each subject remained stable, and the output of collaborative innovation also remained stable. However, as the investment of enterprises changes, the willingness to cooperate of various entities begins to fluctuate. As the system continues to operate, the benefits of collaborative innovation between enterprises and scientific research institutions continue to approach the benefits of independent innovation, causing a certain subject to choose to withdraw from the system and ultimately leading to the termination of the system and the collapse of the collaborative innovation system. Figure 6 is right here.
4.2.5. Scenario V
Scenario V begins with all subjects having the same initial investment and equal benefits, while universities have the highest technical level, followed by scientific research institutions, and the lowest enterprise. The evolutionary result is shown in Figure 7. Enterprises have a strong initial willingness to cooperate, followed by scientific research institutions and universities. Because universities have the highest technical level, the income of universities in the process of collaborative innovation is lower than that of enterprises and scientific research institutions, which leads to the continuous reduction of their later cooperation investment. Until the benefits of collaborative innovation in universities are consistent with those of independent innovation, the universities then choose to withdraw from collaborative innovation. Figure 7 is right here.
4.2.6. Scenario VI
Scenario VI is a situation where each subject of collaborative innovation has the same initial investment, different technical levels, and uneven benefits. The evolutionary result is shown in Figure 8. The results show that the depth of cooperation between the subjects of the system is significantly higher than when the benefits are equal, and although the initial strength of cooperation willingness is different, there is not much difference, and the overall collaborative innovation benefit only appears to be significantly reduced in the later period. Unexpectedly, the benefits of collaborative innovation of various entities are basically stable, but the benefits of independent innovation of universities and scientific research institutions have undergone great changes. As cooperation progresses, the technology of universities is continuously absorbed by enterprises and scientific research institutions, which has improved the technological development of enterprises and scientific research institutions, resulting in an increase in their own innovation activities and continuous improvement of the benefits of collaborative innovation. Regardless of whether the technical level of either universities or scientific research institutions is raised to a higher level, they will withdraw from the collaborative innovation system to obtain more benefits and carry out independent innovation. Figure 8 is right here.
5. Conclusion and Suggestion
This paper adopts software NetLogo 6.0.4 to simulate and analyse different scenarios of IUR collaborative innovation in the equipment manufacturing industry. Our research indicates:
(1) The equipment manufacturing IUR collaborative innovation system can only maintain stable operation under the condition that all subjects have the same initial investment, technical level, and benefits. In other cases, the system will be unstable, leading to the final failure. Therefore, when selecting partners in the equipment manufacturing IUR collaborative innovation, the subject should choose as much as possible partner with a strong desire to cooperate, and having the characteristics of similar economic strength and technical level. In reality, such conditions are too harsh to make the existing equipment manufacturing IUR collaborative innovation mostly in a difficult situation. In particular, IUR collaborative innovation will also be affected by factors such as geographic proximity and organizational proximity.
In others words, an effective measure to solve this dilemma is to introduce an authority or intermediary platform to intervene and regulate. As the main force of macroeconomic regulation and control, the government has the functions of coordinating the overall situation and grasping the advantages of all subjects. Appropriately strengthening government intervention can not only promote the effective cooperation of each subject of equipment manufacturing IUR collaboration innovation but also adjust internal conflicts to ensure the smooth progress of IUR collaboration innovation.
(2) Universities, enterprises, and scientific research institutions are more inclined to cooperate when they are dominant, which leads to difficulties in the coordination of the benefits of the equipment manufacturing IUR collaborative innovation system. On the basis of the above evolutionary results, it is clear that the unequal division of benefits can promote the development of the IUR collaborative innovation system when the initial investment of each subject of IUR is different. According to the difference of input and contribution, each subject receives the benefits of collaborative innovation accordingly, which is beneficial to the stability of the collaborative system.
(3) In the IUR collaborative innovation of the equipment manufacturing industry, the introduction of intervention measures will benefit the collaboration of all subjects, and the formulation of rewards and punishments is an effective measure. The establishment of rewards and punishment measures can also play a role in screening equipment manufacturing IUR collaborative innovation subjects. When looking for a partner, each subject will conduct a comprehensive assessment of the partner’s situation as much as possible to effectively avoid opportunistic risk and trust risk in equipment manufacturing IUR collaborative innovation.
Data Availability
The figure and table data used to support the findings of this study are included within the article.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The authors thank for the funding from National Social Science Fund of China (Major Program, number: 18VSJ087).