Simulation Research Of Knowledge Network And Diffusion Model Based On Enterprise Cluster

As human society entered the era of knowledge economy, knowledge is considered as one of the most competitive resources within enterprises. At the same time, enterprise cluster has become one of the major features in current globalization economy. Therefore, the research of knowledge network and diffusion model based on enterprise cluster is very important and necessary.In this paper, two knowledge network models are proposed according to above problems.(1) Enterprise's knowledge diffusion model based on learning. Enterprise's knowledge resource can be classified into subsets as nodes and relationships between them being viewed as edges. In this way knowledge acquisition of employee can be simulated as an agent wandering cross the hierarchical networks while the visited nodes are viewed as the acquired knowledge. A multi-agent random walk scheme is proposed to improve the efficiency of knowledge learning. Four visiting strategies (i.e. random choice, preferential choices of edges, nodes, and degree distribution) are put forward towards two network models, namely, Barabasi-Albert and Erdos-Renyi, by which the agents walk on the network in different ways. The simulating experiments show that different models and strategies may bring different degree of knowledge acquisition. Besides, a kind of node-weighted network is introduced.(2) Cluster's knowledge diffusion model based on client's requirement. A knowledge diffusion network is established considering the various requirements from clients in the market. The whole diffusion process of enterprise clusters based on this network is simulated by the computer, in which the average knowledge level (AKL) and the knowledge deviation (KD) of the cluster are recorded as diffusion efficiency index according to the process. The effect of market size and knowledge dimension are also taken into account and studied in the paper.For each case, simulation result is discussed and explained. Several suggestions are given to improve the knowledge management, and the whole research work is summarized in the end.