10. Learning-Based Adaptive Large Neighborhood Search for Efficient Pod Repositioning in Robotic Mobile Fulfillment Systems
Invited abstract in session WB-12: AI and Optimization for Warehousing, stream Artificial Intelligence, Machine Learning and Optimization.
Wednesday, 10:45-12:15Room: H10
Authors (first author is the speaker)
| 1. | Lin Xie
|
| Chair of Information Systemes and Business Analytics, Brandenburg University of Technolgy |
Abstract
The pod repositioning problem is a unique challenge in robotic mobile fulfillment systems (RMFS), notably exemplified by Amazon Robotics. In these systems, robots transport racks containing ordered items from the storage area to human pickers. Once the items are retrieved, an optimized decision must be made regarding the repositioning of the pods back into storage to maintain efficiency. Previous studies have addressed this problem using various heuristics.
In this work, we propose an adaptive large neighborhood search (ALNS) approach enhanced by deep reinforcement learning (DRL). Our method dynamically configures selection and acceptance parameters during each iteration, improving solution quality and adaptability. We evaluate our method against existing heuristic-based approaches in the literature, demonstrating its effectiveness in optimizing pod repositioning.
Keywords
- Warehouse Design, Planning, and Control
- Metaheuristics
- Machine Learning
Status: accepted
Back to the list of papers