Container terminal operation system is a complex discrete-event system,whose characteristics are difficult to be described by mathematical models,while process-based simulation models are in lack of flexibility of equipment dispatching.In order to analyze the effect of the yard truck fleet size on terminal operational performance by precisely depicting the container flows at port,an event-driven multi-agent simulation model for domestic container terminals is proposed.Results indicate that,gross crane rate (GCR) first increases steeply and then slowly as the number of yard truck increases,and both average waiting time (AWT) and average service time (AST) of ships show high negative correlations with GCR.