SUPR
Reinforcement Learning for Production Logistics
Dnr:

NAISS 2024/22-563

Type:

NAISS Small Compute

Principal Investigator:

Ayca Ă–zcelikkale

Affiliation:

Uppsala universitet

Start Date:

2024-04-15

End Date:

2025-05-01

Primary Classification:

20202: Control Engineering

Webpage:

Allocation

Abstract

The goal of this project is to enhance sustainability and competitiveness of Sweden's manufacturing sector through the application of artificial intelligence (AI). Our research focuses on developing AI-powered tools for decision support using virtual production models. Specifically, we are working to improve the sequencing and allocation of resources for production logistics tasks. A key area of our study involves testing the effectiveness of deep reinforcement learning for addressing real-world job scheduling challenges in a production setting. This research addresses a dynamic resource allocation problem, where resources are allocated to various jobs dynamically to optimize performance metrics. This requires intricate decision-making, considering numerous operational variables on the factory floor and diverse product details.