Mitigating Dimensionality in 2D Rectangle Packing Problem under Reinforcement Learning Schema
Year: 2024
Type: preprint
Abstract: This paper explores the application of Reinforcement Learning (RL) to the two-dimensional rectangular packing problem. We propose a reduced representation of the state and action spaces that allow us for high granularity. Leveraging UNet architecture and Proximal Policy Optimization (PPO), we achieved a model that is comparable to the MaxRect heuristic. However, our approach has great potential to be generalized to nonrectangular packing problems and complex constraints.
Source: arXiv (Cornell University)
Authors: Kołodziejczyk, Waldemar, Kaleta, Mariusz
Institution: Warsaw University of Technology
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4403780422