| Since the Fifth Plenary Session of the 18 th Central Committee of the Communist Party of China in 2015,innovation has been positioned as the national strategy to promote the China's modernization through accelerating the development of the science and technology.To be an important way to the construction of an innovative country,industrial clusters provide an efficient way to comprehensively improve the quality of China's core products.Although as a large manufacturing country,China has thousands of industrial clusters,there is a big gap for the core and key products between China and the developed countries in the world,making it a huge challenge for these clusters to transform or upgrade from traditional low-cost clusters to innovative ones.Even for some innovative clusters with good development foundation,it is still faced with some serious problems including insufficient collaboration,misalignment among different segments,low innovation efficiency,and slow innovation diffusion.Therefore,how to improve the innovation efficiency and promote firms' collaboration within an industrial cluster has become a critical problem which needs to be explored from both academic as well as practical domains.To addresses this problem,this study explores the ways of the collaborative governance from cluster innovation networks perspective.While previous researches on cluster innovation networks mainly discuss the enterprise innovation behaviors or network structure characteristics separately,rare studies combine them even though both factors are important for collaborative innovation in an industrial cluster.To fill this gap,in this dissertation,the multi-agent models are established based on the interaction between enterprise innovation behavior and network structure,and the collaborative governance strategies of cluster innovation networks are discussed systemically.Specifically,this study tries to address the following questions:(1)what are the key factors that determine the performance of cluster collaborative innovation?(2)at different stages of industrial cluster life-cycle,what strategies are necessary to promote collaborative innovation of cluster innovation networks?(3)from the perspective of information diffusion efficiency,how do the cluster innovation networks achieve collaborative innovation?In order to study the proposed questions,this dissertation first reviews the related literature and theories to find the research gap,which includes cluster innovation networks,network governance theory,collaborative innovation theory and multi-agent models.Then taking growing and matured clusters as samples,the key factors affecting cluster collaborative innovation are explored by utilizing in-depth interview and grounded theory,which lays a theoretical foundation for the follow-up model construction and simulation.After that,according to the key factors and the characteristics of different stages of the cluster,the corresponding multi-agent models are constructed respectively,so as to carry out in-depth research on the governance of cluster collaborative innovation.In order to describe the innovation behavior of cluster enterprises,a dynamic model of enterprise innovation behavior depending on the accumulation of innovation knowledge is established.Based on this model,from the perspective of network structure stability,two different stages of clusters including mature and growing clusters are distinguished since mature clusters have stable network structures and growing clusters have unstable ones.Furthermore,at each stage,according to the symmetry of innovation information,the different situations of instant information transmission and information transmission delay are considered and the corresponding multi-agent models are constructed respectively.For all these multi-agent models,the relevant theoretical analyses are conducted and numerical simulations are carried out.Finally,different governance strategies based on event trigger are proposed to better achieve the collaborative innovation within a cluster.Through the research on the collaborative governance of cluster innovation networks by using multi-agent model,the main contributions of this dissertation can be obtained as follows:1.By using the grounded theory,the main factors affecting the cluster collaborative innovation are investigated,which include information transmission efficiency,network topologic structure,innovation potential differences among enterprises,enterprise's investment on innovation and government supporting policy.2.Based on the innovation characteristics of cluster enterprises,the dynamic model of enterprise innovation behavior relying on the accumulation of innovation knowledge is established and the leader-follower multi-agent model of the mature-stage cluster innovation networks is constructed.Based on the proposed model,an event-triggered collaborative governance strategy of cluster innovation networks is proposed.Theoretical analyses and simulation results show:(1)there is an inverted U-shaped relationship between the accumulation of innovative knowledge and the degree of collaborative innovation;(2)incentive intensity of leading enterprises is positively related to the efficiency of achieving collaborative innovation;(3)network governance intensity is positively associated with the degree of collaborative innovation;(4)compare to the random network,scale-free network is more conducive to the cluster collaborative innovation at the matured stage.3.Considering asymmetric information situation in matured clusters,this dissertation designs the collaborative governance strategies for cluster innovation networks which contains the determinants found in Chapter 3,that is,network structure,enterprise innovation potential differences,and information transmission time delay.The main findings include:(1)fixed governance period is negatively related to the frequency of collaborative governance;(2)information transmission time delay is negatively correlated with the degree of collaborative innovation;(3)the network governance intensity and the degree of collaborative innovation have an inverted U-shaped relationship;(4)scale-free network tends to be the most efficient structure to realize the collaborative innovation in the mature-stage cluster.4.The T-S fuzzy method is introduced to deal with the instability of the growth-stage cluster innovation network.T-S fuzzy multi-agent model is established to explore collaborative governance strategy for cluster innovation networks.The main results indicate:(1)accumulation of innovation knowledge is negatively correlated with the frequency of collaborative governance;(2)there is a negative relationship between the frequency of collaborative governance and the fixed governance period;(3)network governance intensity is positively correlated with the degree of collaborative innovation;(4)at the growing stage of an industrial cluster,network structures from random network to small-world network are more conducive to collaborative innovation;(5)the increase of network size will not affect the designed governance strategy,showing the robustness of the proposed strategies.5.Considering the influence of asymmetric information on the collaborative innovation in the growing cluster,a switching multi-agent model of cluster innovation networks is established based on the information transmission time delay and the collaborative governance strategy with aperiodic sampling for cluster innovation networks is designed.The main findings include:(1)information transmission time delay is negatively correlated with the degree of collaborative innovation;(2)network governance intensity is positively related to the degree of collaborative innovation;(3)the changes of network size and network structure will not affect the effectiveness of the proposed strategies for cluster innovation networks;(4)during the growing phase of an industrial cluster,small-world network is the most efficient to the collaborative innovation and by contrast,random network is not conducive to the collaborative innovation. |
PDF Full Text Request