Estimating cycle type distributions in multi-shuttle automated storage and retrieval systems

A model is presented for configuring storage racks in automated storage and retrieval systems where multi-shuttle storage and retrieval machines can be used. The models extend traditional rules of thumb for sizing storage racks based on target levels of system utilization. By using a stochastic, analytical model of system interleaving, the extended models predict the relative proportion of different types of order picking cycles used in a system. This enables estimation of the average operating time of storage and retrieval machines per transaction served and machine utilization for a given rack configuration. For fixed levels of transactions demand, the model is applied in a search process for finding efficient storage rack configurations defined by the number of storage aisles, the rack height and aisle depth. The model can be used to search for cost-effective machine design and storage rack combinations prior to simulation validation. Significance: A key limitation of traditional rule-of-thumb methods for multi-shuttle ASRS sizing is overcome through development of a stochastic model for prior estimation of the distribution of order picking cycle types based on transactions demand and cycle times.