An aggregate model of an order-picking workstation is proposed for an automated warehouse with a goods-to-men order-picking system. A key aspect of the model is that the various stochasticities that contribute to the workstation performance are not modelled in detail. Rather, these are aggregated into a single effective process time distribution, which is calculated from arrival and departure times of products at the workstation. Two model variants have been developed namely for workstations processing products and orders in first-come-first-serve (FCFS) and non-FCFS sequence. Both model variants have been validated using data obtained from an operating order-picking workstation. The use of aggregate models in practice is then elaborated.