Expected Retrieval Time Comparison Among Robotic Storage and Retrieval Systems With Various Structures

Warehouses recently are increasingly using autonomous robots and compact structures to handle fluctuating demands in a cost-efficient way. While organizing items in a compact cube can save floor space cost, it sacrifices the operational efficiency due to reshuffling and robot detouring to get target items. This study investigates this trade-off in three typical robotic storage and retrieval systems (RS/RSs), including a compact structure that organizes stacks in a grid (RCS/RS), a disperse structure that establishes vertical aisles for robot traveling (VRS/RS), and a structure with an aisle between any two adjacent racks for robot traveling (AVS/RS). We first build an expected travel time model of retrieval transactions, and validate the accuracy of the model by simulation. Then, with the objective of minimizing the expected retrieval travel time of robots, we derive the optimal configuration for three systems. Finally, we construct a cost minimization model of the system with a required throughput capacity obtained by a closed queuing network. We derive the minimum system cost by determining optimal system configuration and the numbers of robots and workstations. The comparison results show that the AVS/RS is the most efficient in robot travel time. The VRS/RS needs the least investment to reach a required throughput capacity, and the AVS/RS is the most expensive. Moreover, when the budget for system configuration, robots and workstations is limited, the VRS/RS can provide the largest throughput capacity.