Emergency departments are currently facing major pressures due to rising demand caused by population growth, aging and high expectations of service quality.  
With changes continuing to challenge healthcare systems, developing solutions and formulating policies require a good understanding of the complex and dynamic nature of the relevant systems. 
However, as a typically complex system, it is hard to grasp the non-linear association between macro-level features and micro-level behavior for a systematic understanding.

This article presents an approach to discover knowledge of emergency department through simulating individual behavior of its components. 
Agent-based modeling technique was used to simulate the behavior of system components. 
The behavior simulation model can generate interaction information under various configuration scenarios. 
Analyzing this interaction information thoroughly enables knowledge discovery towards a better understanding of the complex systemic behavior. 
This makes it possible to explore association between micro-level behaviors of individuals and macro-level patterns that emerge from their interactions, thus assisting users to better understand a system's behavior under various conditions. 
Additionally, a layer-based architecture was used to achieve flexibility and configurability.

Finally, case studies are used to demonstrate the potential use of the proposed approach. Results show that the proposed framework can significantly reflect the non-linear association between micro-level behavior and macro-level features.