Deep Reinforcement Learning for the Job Shop Scheduling Problem: Reference Axes for Modelling, Implementation and Validation

The purpose of this paper is to indicate how the association of deep learning with reinforcement learning, known as deep reinforcement learning (DRL), influences recent research on one of the most representative problems in the production planning and control (PPC) area, the job shop scheduling problem (JSSP), by identifying both the main approaches and tools applied in its modelling, implementation and validation, and the direction in which future research moves.