Reinforcement Learning for Flexibility and Efficient Production
Year: 2025
Type: article
Abstract: This study presents a review of Reinforcement Learning (RL) as a tool for enhancing the efficiency and flexibility of Manufacturing Production Systems (PS) related to dynamic, Flexible Job Shop Scheduling Problems (DFJSSP) and Flexible Job Shop Scheduling Problems (FJSSP). Specifically, it identifies gaps and emerging trends. Although RL has shown significant advantages over traditional methods, the findings reveal critical gaps, including a lack of comparative studies between RL variants, its application in large-scale systems, and limited validation in the real world. This study emphasizes RL’s role in Industry 4.0, providing a roadmap for implementing adaptive, sustainable, and intelligent PS.
Source: IFAC-PapersOnLine
Institution: University of Concepción
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4414567836