On Evolutionary Dynamics Of Knowledge-transfer Networks

In the era of knowledge economics, knowledge has become the core competitiveness of the organizations. For modern organization, how to effectively acquire and share knowledge and to make knowledge transfer successfully in the organization becomes the chief problem. Therefore, knowledge transfer has become an important direction of the knowledge management field. As the transfer of knowledge usually takes places on a network of social relations, which itself may in turn evolve due to the activities of knowledge transfer, in this thesis the evolution of knowledge transfer network is studied from the aspect dynamical modeling.Two factors are critical for studying the evolution of knowledge transfer network. The first is the behavior of individuals in the knowledge transfer and network evolution process; and the second is the managerial regulations on the knowledge and relationships of groups. According to this view, two agent-based simulation models are respectively developed, based on the complexity analysis of the network elements of knowledge transfer and its transfer characteristics. The interacting effects between knowledge transfer dynamics and network structure evolution are correspondingly studied in the present thesis.(1) Evolution of Knowledge Transfer Network Model Based on Individual Behavior. In the proposed model, three main factors are taking into account, namely the stickiness of social relations, the cohesion of local cliques, and individual’s tendency for homophily in knowledge exchange. The local rules for knowledge exchange and neighborhood adjustment are accordingly defined; and the macroscopic phenomena of network evolution and knowledge transfer are simulated and analyzed. The simulation results show that the local rewiring of neighborhood of the knowledge-seeking agents may have prominent influences on both the evolution of the network structure and the performance of knowledge transfer. In particular, the spontaneous emergence of the "small-world" structure from the initial ER random network is analyzed in the present model.(2) Evolution of Knowledge Transfer Network Model Based on Group Regulation. From the aspect of administrative coordination, this thesis further studies the effects of the adjustment of network structure on the performance of knowledge transfer with another agent-based model. In the model, we assume some exogenous force to purposefully adjust the "knowledge-connection" structure of the knowledge-transfer network and test how such adjustment influences the performance of knowledge transfer. By comparing the performance of knowledge transfer in a series of networks which differ from one another in their "knowledge-connection" structures, the simulations illustrate that the maintenance of knowledge heterogeneity in agents’neighborhood, by e.g. employee turnover, is a positive factor for boosting knowledge transfer.According to the simulation results, this thesis may establish a theoretical foundation for the further study on the knowledge management methods that promote the organization knowledge transfer.