集群自组织的复杂网络仿真研究

Self-organization of industrial cluster:A computer simulation from complex network perspective

集群演进过程是一个很难进行实证研究的问题,影响因素的复杂性以及动态数据的匮乏严重制约了对集群演进的定量研究.基于传统管理理论和方法的局限性,尝试从复杂理论角度研究集群演进过程.首先认为集群演进实质上是自组织过程,然后再通过复杂网络聚类系数、度分布、平均路径长度、网络的时效和质量等指标,采用仿真模拟方法对集群自组织过程进行了量化研究.主要结论有:(1)在集群自组织过程中,企业会首先利用地理位置的优势广泛地与群内其他企业接触,通过不断学习过程,从而建立相对稳定的交流圈层;(2)给定企业不同的能力,每个企业所享受到的集群优势是有差异性的,而同一集群内部企业在自组织过程中会逐步产生趋同性;(3)集群自组织过程是企业总体从无序逐步到有序的自发过程,企业与环境之间,以及企业与企业之间都会呈现出相互适应而逐步稳定的趋势.

Industrial cluster evolution has long been a challenge for scholars, especially for quantitative research, in part due to both the complexity of simulating the evolutionary process and the demand for longitudinal data. To partially fill this gap, we attempt to introduce insights from complexity theory perspective and to examine the mechanism by applying computer simulation. Taking the cluster evolution as a self-organizing process, we first develop an analytical framework, and then simulate the evolutionary process by employing five statistic indexes drawn from complex network domain. The preliminary results indicate that （ 1 ） in the self-organizing process of industrial cluster, an enterprise will at the beginning make full use of geographic proximity to connect with the incumbents, and through the learning process, the enterprise will establish its stable communication circle; （2）different enterprises vary in their benefits from cluster, yet there appears to be a tendency of convergence with the deepening of self-organizing process, and （3） as part of an autonomous self-organizing process in which individual firm＇ s behavior appears to be random, yet collectively certain kind of order emerges in a cluster, a industrial cluster will exhibit a tendency toward temporary equilibrium the adaptations between firms as well as between firms and their environment.