A retrieval sequencing heuristic for miniload end-of-aisle automated storage/retrieval systems

In this paper, a retrieval sequencing scheme aimed at improving the throughput of miniload automated storage/retrieval systems in an order picking environment is developed. We assume that an order composed of retrieval requests is always available such that dual command cycles are always performed. A nearest-neighbour retrieval sequencing heuristic is presented, an analytical model is developed to predict its performance, and this model is validated using simulation. The heuristic is shown to improve throughput by 5–15% over traditional first-come-first-served retrieval sequencing. The heuristic achieves this improvement by appropriately sequencing retrieval requests within an order and also optimizing retrieval requests among successive orders. An upper bound for throughput under any sequencing rule is established, and under the conditions tested, the heuristic is found to perform within 3–6% of this bound over different order sizes.