| With the coming of knowledge-based economy,instead of the traditional production factors such as earth,labor and capital,knowledge has exceeded to be the uppermost economic resource and the dominant source of competitive advantage. Therefore,knowledge management has been blooming in the fields of theory and practise.A comprehensive literature study was conducted to find that knowledge network has become a hot issue.Much research on knowledge network is converted from quality and point,simplicity to integration and complexity.Knowledge network is a dominate effect existed in industry clusters,and knowledge spillovers and organization learning provide a foundation for knowledge network.The concept of knowledge network of this paper is different from staff knowledge network.It is a kind of complex network existed in special industry clusters,which has knowledge exchange relation among nodes.Because of knowledge diffusion,cluster knowledge network is also a part of local and global knowledge network.This paper,from the view of complex system,gives a thorough study on network structure and its evolution,by means of an empirical study of industry clusters.The paper research obtains main conclusions as follows:(1)Knowledge network in industry clusters is a complexity adaptive system.The author first use complexity adaptive system theory to interpret the system thought and multi-level facet structure.Then,propose a conceptual model and some evolutionary strategies with high practical ability.Research shows knowledge network has explicit characters of complex system.The first,knowledge network has different kinds of coupling among actors.The second,knowledge network is provided with multi-feedback mechanism,positive or negative,through which system can make information and knowledge flow.The third,the relationship among actors possesses characters of repeated game during a long time.(2)Based on the deep-rooted coupling mechanism,we elaborate the complexity including three levels:elements,ties and evolution.Realizing the attracting complexity emerging from knowledge network seems to frustrate scholars' interest of modeling,but it also brings us huge challenges.With multi-hierarchy and nesting each other,the structure of knowledge network is complex and different.There are much influence and coupling among the subsystems.Any small perturbation would cause network system catastrophe.The whole behavior pattern emerging from the local and mutual relationship leads network structure to durative change and alternation.(3)The measurement of knowledge network structure can be depicted by four dimensions.An increasing number of attentions to knowledge network in industry clusters are paid by practitioners and scholars,and it is becoming important how to measure knowledge network structure.Through theory analysis,the paper builds a research framework and constructs SEM,and then goes on an empirical research on 117 firms which come from Zhejiang province.The results show that the measurement model has good construct validity and the instrument includes four dimensions:centrality of an actor,strength of a tie,quality of relationship and density.(4)The evolutionary process is provided with some characters of dissipative system,such as an open system far from the balanced state,the non-linear and fluctuation.Limited to narrow understanding of knowledge management,in the past, people focus on special subset or some part in knowledge network system.With the relationship between knowledge management and its environment complex,the shortcomings of narrow understanding are emerging.Therefore,we need an organic method consider knowledge network evolution.The innovation of this part is the author analyses dynamic characters of evolutionary process,from the perspective of dissipative structure theory,and focuses on four aspects,that is,an open system far from the balanced state,the non-linear and fluctuation.(5)We can simulate the evolutionary behavior of knowledge network with the help of artificial neural networks(ANN).With the opinion evolutionary modeling of knowledge network is a kind of system simulation,we have established ANN-based method of evolutionary model and concluded its steps.A neural network has a massively parallel and distributed structure,which is composed of many artificial neurons.And the neurons can be used in applications for information processing.An exact simulation method improves the present research and the simulation shows it is right to carry through dynamics analysis.An intelligent measurement was achieved by neural networks with self-adaptation and self-study,and the results from the simulation are satisfied.Compared with previous research findings,the following aspects are unique points of this research:(1)The paper constructed a new perspective to interpret complex knowledge network system.Based on many relative studies,we use complexity adaptive system to interpret the internal attribute,mechanism,environment and synergy evolution. This paper constructed a new analytical paradigm to interpret knowledge network system,and the technology path is "embedded system thought-system layers-concept model- complexity emerging from network system".The result shows the analytical paradigm can better describe the essence of the appearance,innovation,learn and adaptation behavior of complex knowledge network system in industry clusters.The perspective and the conclusions have referenced value for future research.(2)This paper outlined a general framework for studying knowledge network, based on two dimensions including structure and evolution.The framework can give a systemic interpret for knowledge network in industry clusters,such as knowledge spillover,knowledge transfer and knowledge integration.Meanwhile,the study broadens research path and perspective,and the explorative findings will provide reference and experience for future research.Being an active actor in cluster network and environment,firms are embedded in knowledge network.It is useful to study the dynamic relationships and reactions among firms,universities,R&D institutions, government and agencies,and to understand the internal mechanism.Through the former study,the validity of the framework is tested.(3)Using the purified technique,an instrument of knowledge network structure is developed through an empirical study based on Zhejiang industry clusters.The measurement of knowledge network structure is a radical issue of knowledge network theory and unwonted in relative research in China.The results will contribute great value to forming system info of knowledge network,especially for the future empirical study.The measure model of knowledge network structure is developed based on three levels,including actor level,inter-actor level and network level.Pretest, pilot test,and survey are implemented to purify the items.And,reliability ad validity of the instrument is tested empirically.(4)The evolutionary framework is purified from abstract evolution,which provides some strategies for firms' management practice of knowledge network.In the research fields on knowledge network,many terms and concepts aren't unified, and the foundation,framework and theory model belongs to the seedtime.The analytical framework of this part is "evolutionary mechanism/process-dynamic model and simulation-optimizing network structure",and this framework,in fact,is based on two hypothesis accepted by us.One is that knowledge network system is a dissipative structure with non-equilibrium and order;and can maintain only in the opening system with matter flow,information flow and energy flow.The other is that the ultimate evolutionary cause is the non-linear reactions of subsystems(coupling elements).From this significance,the concept model constructed by the paper will contribute great value to any firms.Based on the explorative study,the results enrich the theory of knowledge network,and it is instructive to direct management practice of enterprise.Restricted to time,ability and data collection,this paper has some shortcomings waiting for improvement in future.Possible development is to use the instrument to test the relationship of knowledge network structure,knowledge sharing and innovation performance.In addition,with the improvement of data collection,based on more effective test methods,the measurement of the evolving results is worthy to be further explored.
|
PDF Full Text Request