This paper develops an order throughput time evaluation model for a symmetrical warehousing system incorporating the COI-based storage policy, with traversal and return routing strategies. The first and second moments of the order picker's service time are derived and applied into an E_k/G/1 queueing network for estimating the expected order throughput time accurately. The results generated by the model are compared and validated via the eM-Plant simulation package. The results show that the proposed model can provide sufficiently accurate evaluations of the warehousing system performance for practical purposes. Furthermore, the provided model can be used to determine the optimal number of orders in a batch and analyze the effects of popularity skewness and warehouse layout on the order throughput time.