产学研合作网络及性能分析
本文选题:网络结构 + 产学研合作 ; 参考:《山东师范大学》2017年硕士论文
【摘要】:随着技术进步、知识更新的加快,企业、学校和研究所面对激烈的竞争,三方为了整合各方资源、提高创新的效率,并且降低承担的风险,三方结成战略联盟或者加强合作关系,形成联合开发、优势互补、利益共享、风险共担的产学研合作网络。这种组织介于市场和一体化组织中间,既避免了市场组织的过于松散,又避免受一体化组织的过于规范的约束。产学研合作网络是指高校、科研院所以及企业之间基于某种社会联系,在信任和共享互补资源基础上,就各自需求进行合作形成的,具备社会网络特性和基本结构的组织。文章从社会网络的视角,以山东省产学研联合申请专利数据为例,将企业、高校和研究院所抽象为网络中的节点。两个机构之间有联合申请专利,则认为两个机构有边连接,由这些点和边构建了山东省产学研合作网络。对该合作网络的统计特征,如密度、度分布、平均路径长度等进行了研究。发现产学研合作网络由很多连通子图构成,合作紧密程度低,连通子图内部知识扩散快,效率高,且具有无标度的网络特征;并从网络中的点故障和边故障两个角度对网络的鲁棒性进行了研究,发现网络对节点或边的随机故障及按照边介数删除边有较高的鲁棒性,而对按照节点度优先和点介数优先删除节点极其脆弱。基于网络演化博弈的角度,从网络结构、博弈矩阵及演化规则的三个要素出发,构建了新的产方和学研方在专利合作网络中的演化博弈模型。最优策略的选择是通过网络的平均积极合作概率以及平均收益来测度的,并且从通过改变博弈支付矩阵中知识溢出、惩罚因素和分配系数等参数的大小,来仿真研究其对产学研合作网络合作行为的影响。通过数值模拟发现合作行为受到这些参数变化的影响。发现惩罚成本大于研究开发成本时知识溢出效应对合作行为的影响与惩罚成本小于研究开发成本时的影响有很大差异。结果表明合理的惩罚会有利于积极合作,合适的惩罚能促进度大的节点与其邻居节点合作。通过实验发现合适的产方与学研方的合作收益分配比例为一半对一半。由此为了保证产学研合作网络更好发挥其作用,我们需要设置一定的条件因素来确保网络中的大度节点(即学校和科研院所)积极合作,从而使网络的每个个体均能受益。
[Abstract]:As technology advances and knowledge updates accelerate, businesses, schools, and research institutes face fierce competition. In order to integrate resources, improve innovation efficiency, and reduce risks,The three parties form a strategic alliance or strengthen the cooperative relationship to form an industry-university-research cooperative network with joint development, complementary advantages, shared benefits and shared risks.This kind of organization is between the market and the integration organization, which not only avoids the market organization too loose, but also avoids being restrained by the integration organization too much.The industry-university-research cooperative network is an organization with the characteristics and basic structure of social network which is formed by cooperation on the basis of trust and sharing of complementary resources based on certain social connections among colleges and universities research institutes and enterprises.From the point of view of social network, this paper abstracts enterprises, universities and research institutes as nodes in the network, taking the joint patent application data of industry, university and research institute in Shandong Province as an example.There is a joint patent application between the two institutions, and the two institutions are considered to have edge links, from which the Shandong Province Industrial, academic and Research Cooperation Network has been established.The statistical characteristics of the cooperative network, such as density, degree distribution and average path length, are studied.It is found that the cooperation network is composed of many connected subgraphs with low degree of cooperation, fast knowledge diffusion, high efficiency and scale-free characteristics.The robustness of the network is studied from the point fault and edge fault. It is found that the network is robust to the random faults of nodes or edges and to deleting edge according to the number of edge mediums.But it is very fragile to delete nodes according to node degree priority and point number priority.Based on the theory of network evolution game, a new evolutionary game model is constructed based on the three elements of network structure, game matrix and evolution rules.The selection of the optimal strategy is measured by the average positive cooperation probability and the average income of the network, and the size of knowledge spillover, penalty factor and distribution coefficient in the game payment matrix is changed.To simulate and study its impact on the cooperative behavior of industry, college and research cooperation network.By numerical simulation, it is found that the cooperative behavior is affected by the variation of these parameters.It is found that the impact of knowledge spillover effect on cooperative behavior when the penalty cost is greater than the research and development cost is significantly different from that on the penalty cost being smaller than that on the research and development cost.The results show that reasonable punishment is beneficial to positive cooperation, and appropriate punishment can promote the cooperation between large nodes and their neighbors.It is found that the proportion of cooperative income between the right producer and the partner is half to half.Therefore, in order to ensure that the industry-university-research cooperative network can play its role better, we need to set up certain conditions to ensure the active cooperation of the large nodes in the network (that is, schools and research institutes), so that each individual of the network can benefit.
【学位授予单位】:山东师范大学
【学位级别】:硕士
【学位授予年份】:2017
【分类号】:F273.1
【参考文献】
相关期刊论文 前10条
1 凌守兴;许应楠;仇荣国;;产学研合作演化博弈模型构建及其稳定性分析[J];统计与决策;2015年17期
2 徐翔斌;李恒;;基于供需能力的供应链网络鲁棒性[J];系统工程;2015年08期
3 卢方元;常大华;;基于演化博弈的产学研合作创新稳定性分析[J];科技管理研究;2015年16期
4 孙杨;刘星材;景春光;宋瑞;聂婷婷;;不确定需求下轨道交通网络的鲁棒性优化[J];交通运输系统工程与信息;2015年04期
5 曹武军;韩俊玲;;政产学研协同创新的演化博弈稳定性分析[J];贵州财经大学学报;2015年04期
6 许潇丹;;基于产学研合作的企业网络特征、知识获取及创新绩效的关系研究[J];企业技术开发;2015年14期
7 曹霞;刘国巍;;产学研合作创新网络规模、连接机制与创新绩效的关系研究——基于多主体仿真和动态系统论视角[J];运筹与管理;2015年02期
8 胡鹏;帅斌;狄兆华;;突发状况下危险品运输网络鲁棒性建模和仿真[J];交通运输系统工程与信息;2015年02期
9 刘国巍;;产学研合作创新网络时空演化模型及实证研究——基于广西2000—2013年的专利数据分析[J];科学学与科学技术管理;2015年04期
10 华连连;隋宇心;李弘;;基于复杂网络的供应链鲁棒性分析[J];上海管理科学;2015年01期
相关博士学位论文 前1条
1 温芳芳;专利合作模式的计量研究[D];武汉大学;2012年
,本文编号:1770536
本文链接:https://www.wllwen.com/jingjilunwen/xmjj/1770536.html