ABSTRACT
Abstract In this chapter, we derive analytical results for unit-load automated storage/retrieval systems (AS/RS) and compare the throughput performance of two storage/retrieval (S/R) machine dispatching policies over three alternative configurations defined by the locations of the input and output points. These two are simple policies that serve the storage and
retrieval requests on a first-come-first-served (FCFS) basis either across the two request queues combined or individually within each request queue. We assume the expected S/R machine cycle times are given or can be computed from the rack dimensions and S/R machine travel parameters. We also assume that the storage and retrieval requests arrive independently according to a Poisson process and that the arrival rates are given. The numeric example we present assumes randomized storage; however, the results shown in the chapter can be used with other storage policies, provided the appropriate expected S/R machine cycle times can be computed. Although the conclusions may change from one problem instance to another, examining two rack shapes and two workload levels, our numerical results suggest that while one policy performs slightly better than the other one in terms of the expected S/R machine utilization, in most cases there is no significant difference between them for practical purposes. Also, the additional workload imposed on the S/R machine due to unbalanced systems (i.e., having more storage requests per time unit than retrieval requests or vice versa) seems less than what one would have anticipated. We conclude the chapter with a brief presentation of possible research opportunities for AS/RS.
