Structural Feedback Approach to Modeling Behavioral Decision Making in Queuing Systems: Model

Existing queuing models either leave human judgment and decision making outside the scope of the system, ignoring their role as determinants of the system performance, or capture the feedback between system state and human agents’ behavior using approximations at an aggregate level. However, empirical evidence has shown that individual agent’s behavior and interactions can substantially alter the system’s output. In this paper, we consider full behavioral feedback in queuing systems, developing an approach that combines stochastic events and responses to system-wide performance indicators while tracking individual agents. We formulate human responses by explicitly capturing the agent’s objectives and the information available about the system state, while accounting for delays and possible information distortions. Applying this approach to explore a long-standing problem of oscillation in queuing systems with delayed announcement, we demonstrate that customers’ individual behavioral responses are the primary drivers of oscillatory behavior, continuously agitating the system and causing fluctuations between periods of low and high utilization. By formalizing behavioral responses as feedback loops, we analyze the observed system behavior and identify system-wide managerial levers that mitigate oscillations by affecting individual agents’ behavior. The proposed modeling and analysis framework can guide system design and improve performance in scenarios where dynamics are driven by both feedback structure and stochasticity, providing generalizable structural explanations of the impact of human behavior on system output.