Velocity‐based stowage policy for a semiautomated fulfillment system

Abstract Online retail fulfillment is increasingly performed by semiautomated fulfillment systems in which inventory is stored in mobile pods that are moved by robotic drives. In this paper, we develop a model that explores the benefits of velocity‐based stowage policies for semiautomated fulfillment systems, also known as robotic mobile fulfillment systems. The stowage policies decide which pods to replenish with the received inventory. Specifically, we model policies that account for the velocity of the units being stowed. By stowing higher (lower) velocity units on higher (lower) velocity pods, we expect to increase the heterogeneity of the pod velocities. Greater heterogeneity in pod velocities can yield a greater reduction in pod travel distance from velocity‐based storage policies for the pod. Reducing pod travel distance decreases the number of robotic drives that are needed for the system to maintain a certain throughput rate. We analyze an M ‐class velocity‐based stowage policy. We stow units from each velocity class onto pods dedicated to that velocity class; each class of pods then has its own storage zone, where the zones are ordered based on the distance to the stationary pick and stow stations. We characterize the pod travel distance as a function of the skewness of the demand distribution and the number and size of the classes. We find that with two or three classes we can achieve most of the benefits possible from a velocity‐based stowage and storage policy. For representative demand distributions, we find that a two‐class policy achieves 75% of the maximum possible travel‐time reduction and that a three‐class policy improves this to 90%.