Optimizing throughput in multi-aisle automated storage and retrieval systems: a comprehensive analysis of performance under generally distributed fully sequenced orders through merging queuing systems

Abstract

This paper proposes a method to accurately determine the throughput of a multiple-aisle automated storage and retrieval system. The main focus is on the order sequence of the retrieval process from different storage aisles. The process is modeled as a merging process with several inputs, each with limited capacity. A queuing system with finite capacities is used to describe this complex process. Specifically, an M/G/1/K-queuing model is employed for every input stream, and these models are combined to evaluate the performance of different numbers of aisles. A discrete event simulation is used to validate the accuracy of the proposed method. The paper concludes with an example that demonstrates the impact of variables such as the storage aisle throughput, the merging point performance, and its coefficient of variation on the overall performance of the storage aisle. The paper offers a decision tool that can accurately determine the throughput of a multiple-aisle storage system.