The down-peak period is an important aspect of hoist operations in construction projects. To make the vertical transportation service more efficient, the operation parameters of down-peak should be modeled precisely in the process of planning. A simulation model is developed to analyze the operations of elevator in down-peak. The model incorporates hybrid use of discrete-event simulation and agent-based modeling to provide a robust methodology to analyze the two most important parameters for elevator operation planning: time spent and average waiting time. The developed model is validated under several conditions, and its usage is expanded to random situations.