【作者】 阮爱清；

【导师】 刘思峰；

【作者基本信息】 南京航空航天大学 ， 管理科学与工程， 2008， 博士

【摘要】 产业集群是当今世界经济地图上重要的经济现象,而且产业集群被视作发展区域经济的一种有效组织模式。把握产业集群的演化机理,探索产业集群的成长效益对于区域经济管理部门借助产业集群提升区域经济水平具有重要的意义。本论文应用系统科学相关的理论和方法,从不同的视角研究了产业集群的演化机理与成长效益问题。本文首先应用进化博弈理论,从整个生命周期的角度研究产业集群的演化。作者根据现实生活中间存在的收益与进化阶段强相关的现象,改进并提出了一类收益矩阵与进化阶段强相关的进化博弈链模型,用于研究产业集群的演化问题;应用灰色系统理论,将收益与进化阶段强相关的进化博弈模型扩展到不确定(灰数)条件的情形。研究了对称状态和线性收益函数条件下的求解问题,通过仿真模拟,对产业集群的发生与形成、演化与发展同区域经济条件的关系进行了深入研究,较好地解释了产业集聚演化的多种现象和表现形式,明确区域政府在推动产业集群演化中的重要作用。研究结论显示,区域经济环境在一定条件下对产业集群的演化和发展起着重要的作用。不利的条件会明显地延缓和阻碍产业集群的形成;区域经济环境的改善则会大大地促进产业的集聚。在一定的条件下,产业集群是可以被创造出来的。在产业集群整个演化轨迹研究的基础上,本文提出企业集群剩余与产业集群租金的概念,揭示出产业集群演化的机理。通过产业集群租金的系统动力学模型的仿真,并依据产业集群租金的递增率特征点的识别,提出基于集群租金的产业集群演化周期的划分:集群“种子”→“核”生长期→“网”发展期→“群”稳定期→“区”衰退(转型)期;同时揭示出产业集群“核”生长期中存在的“萌发期”,指出其对于产业集群演化的重要意义:“核”生长期中的“萌发期”是政策的重要作用点和着力点。在演化阶段划分的基础上,运用随机网络理论,构建区域产业集群形成与发展的GERT网络模型(Graphical Evaluation Review Technique,简称GERT),研究了区域产业集群形成与发展的“夭折”概率与“成熟”概率等概念,并研究政府在产业集群的不同阶段给予支持时,各种方案对产业集群形成与发展的作用效果,说明政府在集群发展的初期给予支持是最优的选择。此外,本文提出灰色GERT网络技术,给出它的定义及其分类。为衡量灰色GERT网络计算结果的灰度,在灰数覆盖特性的基础上,提出基于顾客需求的灰数估计精度的概念,从而给出在论域无法确定的情况下计算结果的判断方法。对于产业集群成长效益的研究,本文从微观和宏观两个视角进行。微观地看,产业集群是众多企业群体决策的结果,而且参与集群就是参与了一个特殊的企业联盟,企业参与集群的目的就是获得个体联盟租金即体现为企业租金。本文基于Shapley值法的公理和性质,将运用Shapley值法分配得到的联盟个体收益分解成两个部分:基于博弈方投资或实力的收益以及基于博弈方对联盟重要性的收益;并证明了结论的一般性,从而给出了产业集群租金的定量的刻划。通过企业联盟的分析,得出结论:处于关键环节或拥有竞争性资源的企业往往获得最大份额的联盟租金(或集群租金)。本文还将群决策的理论引入Shapley值法模型中建立Shapley值群决策模型来研究企业联盟的建立问题,同时也克服了经典的Shapley值模型无法确定特征函数的困难。本文还讨论以区间数为特征函数估计参数的Shapley值模型,并研究由此产生的高估风险和低估风险及其度量方法。针对群决策中判断矩阵存在的区间不确定性和离散不确定性两种类型,提出了灰数判断矩阵信息集结的理论和方法。同时对于求解过程中发现二元的信息集结的新问题,提出二元组群决策问题及其求解思路。从宏观的视角研究产业集群的成长效应,本文着重定量地研究产业集群与区域经济竞争力提升的互动效应。在界定产业集群内涵的基础上,作者选取了行业增加值以及行业就业人数两个互补的指标来建立产业集群度评价模型,并建立了行业综合集聚度的评价模型和区域综合集聚度的测度模型。依据经济竞争力的不同层次,将区域经济发展动能的概念拓展和应用到行业,建立行业经济竞争力的测算模型,同时提出并建立了区域动态经济竞争力的测度模型。研究两个层次的问题,一是行业的综合集聚度与行业经济竞争力提升的互动作用,二是区域综合集聚度对区域经济竞争力提升的互动效应。本文创造性地借用柯布-道格拉斯生产函数的思想和形式,将产业集群视为不同于与传统的资本和劳动力投资的新的投入要素,寻求区域综合集聚度与区域经济竞争力之间的联系。运用灰色关联理论,通过对大量数据测算、分析,考察区域经济竞争力与区域综合产业集群的相关关系,发现区域经济竞争力与区域综合产业集群度之间存在高度相关关系。结论说明了产业集群与区域经济竞争力之间存在着互相作用、互相促进的关系。产业集群水平的提高对区域经济竞争力的提升有很大的促进作用;同样区域经济竞争力水平的提升,也能为产业集群创造更好的集聚优势和集聚动力,二者之间可以互相促进,形成良性循环。但是,当两者(或其中一个)出现停滞或衰退,就会产生负面效应,形成恶性循环。在理论研究的基础上,本文重点研究了中关村高科技产业集群以及温州鞋革产业集群的演化过程,并用本文的理论方法对它们进行演化阶段的划分和分析,通过实际的案例,证明了产业集群演化过程中“萌发期”的存在,以及政府在产业集群发展初期,尤其是核生长的“萌发期”给予引导和支持的必要性和重要性,从而更好的阐明了本文研究的现实意义。本论文对产业集群演化机理和成长效应方面的研究取得了一些初步的成果,但有些结论还需要更多的产业集群的实践检验,有些问题还需要更深入的理论研究,这些都是下一步的工作的目标。更多还原

【Abstract】 Industry Cluster is one of the most economic phenomena of the global economy map, and it has been seemed as one of critical ways to promote the regional economic level. It is needed to know the evolvement mechanism of industry cluster and study its growing effect for those who will develop regional economy by way of industry cluster. This research on the evolvement mechanism and growing effect of industry cluster is done from some different angles of the question by theory and method of system engineering in this thesis.Firstly, by evolutionary game theory, the thesis discusses the evolvement of industry cluster from the angle of whole life cycle. According to some realistic economic conditions that the income of some one is in strong correlation with the stage at that time he makes decision, the author advances one category of evolutionary game under strong correlation between individual pay-off and evolution stage, which is used to study evolution of industry cluster. Also by grey system theory, this model is extended to one category chain model of grey evolutionary game in condition that decision person is with limited knowledge and limited information. The thesis discusses its evolutionary stable strategy (ESS) with linear functions and symmetric pay-off matrix condition, and some important conclusions are found, that in some situations ESS could be gotten directly, but in other conditions ESS could not be worked out and it will be sensitive to initial condition. On the basis of these, this model is used to explain the success and failure of industry cluster, and an irradiative conclusion is gotten that in some situation industry cluster could be created by some ways.Secondly, the evolutionary process and stages of industry cluster is studied. Industry cluster could be seemed as an organism with life force and lifecycle, and the development of industry cluster goes through three states, from seed state, through core state, and then to cluster state. On the base of that, the conception of the rent of industry cluster is put forward, and the dynamic evolutionary model of it is established. Moreover according to change rate of the rent of industry cluster from simulation result of system dynamic model, the whole lifecycle of industry cluster should from seed state experience these phases, core growth phase, expand phase, mature phase, and decline or upgrade phase. It is discovered that during core growth phase, there exists a“bourgeon phase”or“difficult phase”, which is critical to development of industry cluster. Based on the grow stages, using Grey Graphical Evaluation Review Technique(GERT), a GERT network model is built up to describe the process of formation and development of regional industry cluster, and the concepts of probability of being premature and probability of mature are proposed. Taking example of high-tech industry cluster, the probability of mature are discussed when support from government are provided at different stages of industry cluster. The conclusions account for the necessary and affection of government support. Moreover Grey Graphical Evaluation Review Technique (GERT in brief) Network is put forward and its model is built up, for that the method of GERT Network requires accurate parameters of action, and however, people usually can not find out exact value of parameters because of limited information of real life. By Grey GERT Network, many actions are easier to be described and discussed. This thesis defines several types of Grey GERT Network and searches for the approach to solution. And the estimating precision is advanced to measure the uncertainty of estimation on the condition that the degree of greyness can not be calculated.The problem of growing effect of industry cluster is research from two different visual angles. From micro angle, the enterprise alliance is the result of group decision-making of all members. So to enter an industry cluster is to attain an enterprise alliance, and the core of establishment of enterprise alliance is the profit allocation of the member. Based on model deducing from the axiom and character of Shapley value of cooperative game, author divides the result of player by Shapley value into two parts, one part is the income decided by the investment and capability of player, the other part is the benefit coming from the importance of player in alliance. And these two parts present the advantage of Shapley value that it includes the long-time and short-time factors,and universality of conclusion is also proved. And the definitions of rent of alliance and efficiency of alliance are advanced and the calculation methods are given. By discussing the example of enterprise alliance, it is found and proved that the key enterprise that holds competitive resource always gains the most quantity of rent of alliance. Then the thesis advances a method combined with Shapley value and group decision making model to solute the problem of the function of enterprise alliance in Shapley value model. Also the thesis sets up the model of Shapley value in the condition of uncertainty based on the incomplete information, finite knowledge and bounded rationality, and discusses the way to solution, based on which the overrating risk and the underrating risk of Shapley value model with parameter of interval number are put forward and defined, and measure methods are given. On the basis of these, a new kind of binary group decision-making problem is put forward.From the macro angle of view, the thesis stresses on the mutual affection of industry cluster and regional economic capability by quantitative analysis method. In order to measure the agglomerate degree of industry cluster and integrated agglomerate degree of region, two kinds of statistic index, including added value of industrial output and quantity of employment of industry are selected to build up measure model. Then the thesis puts forward the measure method of industry competition capability, also the definition and measure model of regional dynamic competition capability. By the improved C-D production function model, the contribution degree of industry cluster to industry competition capability is measured. And the grey relational analysis of integrated agglomerate degree of region and regional competition capability is analyzed. It is found that the integrated agglomerate degree of region and regional competition capability are under strong correlation.Finally, on the foundation of theory analysis on industry cluster, the thesis does case studies by two typical industry clusters: high-tech industry cluster of Beijing Zhongguancun Science Park (Z-Park) and Wenzhou industry cluster of shoes and leather. The development stages of these two industry clusters are discussed and existent of“difficult phase”are analyzed, and practical cases illuminate that it is necessary and significant and helpful for regional administrator to provide support to industry cluster especially at the“difficult phase”of core growth phase.However it must be claimed that the present study is only at its beginning stage, and the obtained results at present are expected to be verified in a larger scope of samples and more deeply theoretical studies. To research deeply the underlying evolution law of real-life industry cluster is the author’s next goals.更多还原