Research On The Evolution Of Alliance Network Structure And Performance Based On Evolutionary Game Perspective

Alliance network refers to the collection of all enterprises that have alliance with the core enterprise acting as the center.Building alliance network is helpful for enterprises to obtain complementary resources,improve organizational learning ability and cope with environmental uncertainty,so it becomes an important strategic choice in enterprise growth.However,with the development of information globalization,the interaction and cooperation between core enterprises have formed a complex network topology.Core enterprises are facing more and more uncertainties in choosing or giving up their own alliance partners in their network groups.The rapid changes of market,position and ecological environment make it difficult for core enterprises to apply the decision results generated by traditional game model to complex networked social group decision making.Because the network is a collection of alliance relationship among core enterprises,the dynamic evolution results of core enterprises choosing to establish or abandon alliance relations through different game strategies will directly lead to the dynamic evolution of the whole alliance network topology,and the topology of the alliance network to which the enterprise belongs will affect the performance of the core enterprise.According to this logic thinking,based on the traditional classic game model,we proposes two game models with social universality to be applied to the selection of core enterprise alliance partners in the alliance network,and puts forward the following main research questions: First,in the alliance network,how will the continuous game decisions of core enterprises under the minority game model lead to the evolution of the entire alliance network structure? Second,in alliance network,how does the continuous game decisions of core enterprises under the ranking game model lead to the evolution of the entire alliance network structure? Third,what influence does the topology of alliance network have on the performance of core enterprises?Based on the above research questions,combined with complex network theory,evolutionary game theory and social network theory,this dissertation comprehensively applied theoretical research,evolutionary game theory,empirical research and other research methods to conduct qualitative and quantitative research.The main research work and conclusions are as follows:(1)The alliance network,which is the basic structure of the whole social system,involves business,management,society,economy and other fields.The sustainability and dynamics of social system depend on the evolution of topological structure.The minority game model has been successfully applied in the fields of social science,economics,management and engineering to clarify the evolution and cooperation dynamics of competitive core enterprises in binary decision situations.By extending a few game models in alliance networks,this dissertation analyzes the cooperative evolution of network alliances with different initial network topologies.In this study,numerical simulation method was used to study three kinds of initial structural alliance networks(random network,scale-free network and small-world network).The results show that network reconstruction strategy can improve the overall performance of alliance network cooperation,and alliance network can obtain higher returns and higher decision-making stability.The minority game strategy that allows network reconnection is beneficial to bring the alliance network from the bad initial stage to a higher level of cooperation.However,intransigence in decision-making can worsen the situation if cooperation is not encouraged;Partial profit strategy may lead to global imbalance,while global strategy may achieve relatively stable results.(2)In the general social ecology and alliance network environment,in order to obtain better and more returns,the core enterprise does not need to have an absolute advantage over competitors,but only needs to have a relatively high level or industry ranking than competitors.In this way,the core enterprises in the alliance network will launch game competition in order to improve the social level.To study this social nature,a ranking game model based on social network is proposed in this dissertation.This dissertation introduces three decision-making strategies based on the promotion of ranking position.We conducted systematic simulation in the random network,small world network and scale-free network.The results show that the ranking of the whole alliance group has different changes in the network topology of these different alliance networks.The centrality strategy leads to the improvement of the overall centrality of the alliance network,while the following strategy tends to make the network have significantly greater edge density,that is,to increase the level of cooperation of the alliance network.In addition,the central strategy resulted in the lowest ranking similarity,the lowest survival rate and the highest mobility,showing significant social structural changes and better social mobility.On the contrary,for random and follow strategies,social networks become more rigid.(3)Network evolution is a process and result of participant evolutionary game.Based on the two kinds of network evolutionary game above,this study further discusses the relationship between alliance network performance and the structure of the alliance network.In many industries,enterprises build a project-based alliance network through the selection of partners,which has a limited lifetime and a clear boundary.The transparency of the film industry provides a unique opportunity to study the impact of alliance network topology on project alliance performance from the perspective of social network and organization theory.In this study,box office and score were taken as the proxy variables of financial performance and word-of-mouth performance of project alliance network,and the influence of structural characteristics of production alliance network and distribution alliance network on box office and score was discussed.In this study,the degree(degree centrality),betweenness centrality,eigenvector centrality and structural hole characteristics of the alliance network are introduced into the regression model as the variables of the structural characteristics of the alliance network.The results show that the financial performance of the project alliance network is significantly affected by the attributes of the alliance network.In particular,the total degree of the alliance network of production companies contributes positively to the box office.The total feature vector center of the production company alliance network has no positive contribution to the box office.In addition,the overall intermediary centrality of the affiliate network has a significant negative effect on the score.The total structure hole of the distribution company alliance network has a significant and positive contribution to the box office,while the total structure hole of the production company alliance network has a significant negative impact on the score.The above research conclusion deepens the research of evolutionary game theory in the topological structure evolution and performance micro aspects of alliance network.The research conclusion has guiding significance for the development and planning practice of alliance network.The research of this dissertation is mainly innovative in the following aspects:Firstly,the alliance network is regarded as an ecological network system which can evolve dynamically.Based on complex network theory and evolutionary game theory,the evolutionary game model on network is constructed,and the dynamic evolution of alliance network is discussed.Secondly,by introducing the method of computer simulation,the simulation model of the influence of game behavior on the evolution of alliance network structure in different situations is constructed.The computer simulation software is used to simulate the high frequency operation.Based on the simulation results,the relationship between the dynamic evolution of alliance network structure and the game behavior of core enterprises is further revealed.Thirdly,in this dissertation,the minority game model is applied to the dynamic evolution of alliance networks for the first time.For the first time,three indexes are proposed to measure the degree of solidification of competitive ranking in alliance networks.Fourthly,this dissertation,for the first time,takes the film project alliance network as the research object,respectively discusses the influence of the network attributes of the production company alliance network and the distribution company alliance network on the performance of the project alliance network,and analyzes the effect rule of the project-based alliance network structure on its performance.