An exact penalty method with nonmonotone line search and rapid infeasibility detection for constrained multiobjective optimization: Application in supervised machine learning

M. Mohd Rashidi

doi:10.1016/j.cor.2025.107351

Deep reinforcement learning for real-time scheduling in large-scale mixed-model production systems with dynamic order insertion

2025-12-03

0

PRE

AI

D

Donghai Wang

邹璟 (Jing Zou) *

X

Xinan Zhou

J

Jin Sun

DOI：10.1016/j.cor.2025.107345

Original

Original request for help

Save

Abstract

En 中文

• A Deep Reinforcement Learning framework for real-time scheduling in large-scale mixed-model production systems with dynamic order insertion. • A modified disjunctive graph model that consolidates similar job processes, significantly reducing computational complexity in large-scale scheduling. • Integration of enhanced graph neural networks into the DRL framework to improve adaptive scheduling decisions for dynamic production events. • Extensive experiments on benchmark and industrial-scale datasets demonstrating improved scheduling efficiency, scalability, and adaptability over conventional methods.

Deep reinforcement learning for real-time scheduling in large-scale mixed-model production systems with dynamic order insertion

Abstract

Journal

Organization

Cited Papers

Citing Papers