Joint design of LGV and pallet shuttle fleets for warehouse automation

This paper deals with emerging automation technologies for pallet handling in multi pallet-deep rack storage systems. In particular, this study investigates opportunities resulting from the adoption of pallet shuttles (PSs) for material handling within the rack structure, i.e. for put-away and retrieval operations from the rack face to the specific pallet location and viceversa. PSs may be used in combination with both manually-driven forklift trucks and Laser Guided Vehicles (LGVs) providing material transport from the end-of-line area to the rack storage and from the rack storage to the loading area. The focus of this study is on the combined use of PSs and LGVs so as to discuss the design of a fully automated warehouse. PSs help to overcome the disadvantages related to one of the most commonly used and space efficient type of racks storage, i.e. drive-in racks or similar solutions. Pallets stored in drive-in racks are not independently accessible so that all the levels of a certain lane must be devoted to the same item type and the amount of time to access the pallet locations increases as the lane depth and the number of levels increase. This may lead to efficiency losses. On the contrary, PSs allow different item types to be stored in independent levels of the same lane, so that an higher storage efficiency can be obtained even in long racks. It should be pointed out that methods to support the design of a full automated logistics system are still in an early stage of development. In particular, to the authors' knowledge, studies about the joint design of LGV and PS fleets are not proposed in literature. Thus, this study aims to develop an analytical model for the integrated dimensioning of the fleet of PSs and the fleet of LGVs in a storage system. Given a fleet of PSs acting as the server of the system, the proposed model provides the pallet shuttle utilization rate as a function of the number of LGVs, by taking into consideration the storage area layout and the production and delivery flows. The model identifies sets of feasible solutions, expressed in terms of number of LGVs and PSs, with the same PS utilization rate, i.e. that can be represented by iso-utilization contour lines. Finally, through the application of queuing theory results, the PS utilization rate of a certain set of solutions can be related to the expected cycle time (order release to shipment). The validity of the model is proved by discussing a real case study.