Robotic Sorting Systems

Bipan Zou, René de Koster, Yeming Gong (), Xianhao Xu and Guwen Shen
Additional contact information
Bipan Zou: Zhongnan University of Economics and Law [China]
René de Koster: Erasmus University Rotterdam (The Netherlands, Rotterdam) - EUR
Yeming Gong: EM - EMLyon Business School
Xianhao Xu: HUST - Huazhong University of Science and Technology [Wuhan]

Post-Print from HAL

Abstract: Many distribution centers use expensive, conveyor-based sorting systems that require large buildings to house them. In areas with tight space, robotic sorting systems offer a new type of solution to sort parcels by destination. Such systems are highly flexible in throughput capacity and are now gradually being introduced, particularly in express companies. This paper studies robotic sorting system with two layouts. The first layout has two tiers: robots drive on the top tier and sort parcels by destination on spiral conveyors connected to roll containers at the lower tier. The second layout has a single tier with input and output points located at the perimeter, connected by robots. For each layout, we consider both the shortest path topology via dual-lane aisles and the detour path topology via single-lane aisles. We build closed queueing networks for performance estimation, design an iterative procedure to investigate robot congestion in the two-tier layout, and use a traffic flow function to estimate robot congestion in the single-tier layout. Random, closest, dedicated, and shortest-queue robot-to-loading-station assignment rules are examined. We validate analytical models by both simulation and a real case of Deppon Express and analyze the optimal system size and operating policies for throughput capacity and operating cost. The results show that the system throughput capacity is significantly affected by robot congestion in the single-tier layout with the detour path topology, but it is only slightly affected in the other systems. A square layout fits the shortest path and a rectangular layout fits the detour path. Both the random assignment rule and the shortest-queue assignment rule are superior for a large number of robots, whereas the dedicated assignment rule is superior for a small number of robots. We apply these insights at Deppon Express for different allocations in peak and off-peak hours. Our analysis shows that a robotic sorting system typically has lower overall annual cost than a traditional cross-belt sorting system when the required throughput capacity is not too large.

Keywords: queueing network; robotic sorting; automated warehouse; AI (search for similar items in EconPapers)
Date: 2021-11-01
References: Add references at CitEc
Citations:

Published in Transportation Science, 2021, 55 (6), pp.1430 - 1455. ⟨10.1287/trsc.2021.1053⟩

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-05635784

DOI: 10.1287/trsc.2021.1053

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().