This study focuses on the co-evolution of informal organizational structures and individual knowledge transfer behavior within organizations. Our research methodology distinguishes us from other similar studies. We use agent-based modeling and dynamic social network analysis, which allow for a dynamic perspective and a bottom-up approach. We study the emergent network structures and behavioral patterns, as well as their micro-level foundations. We also examine the exogenous factors influencing the emergent process. We ran simulation experiments on our model and found some interesting findings. For example, it is observed that knowledgeable individuals are not well connected in the network, and our model suggests that being fully involved in knowledge transfer might undermine individuals' knowledge advantage over time. Another observation is that when there is high knowledge diversity in the system, informal organizational structure tends to form a network of good reachability; that is, any two individuals are connected via a few intermediates.