Copyright (c) 2026 Documentos de Trabajo ECACEN
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Smart Manufacturing Operations Scheduling: State of theArt inIndustry 4.0.
Production scheduling has evolved into an adaptive and connected function within cyber-physical systems and cloud platforms. This transition, analyzed over a period spanning from 1980 to 2025 using the PRISMA methodology, reveals a convergence toward hybrid approaches that integrate discrete event simulation, mathematical optimization, and reinforcement learning to manage the complexity of Industry 4.0. In this ecosystem, Digital Twins emerge as the critical infrastructure for synchronizing the physical plant with decision-making, while Deep Reinforcement Learning (DRL) and graph neural networks stand out in dynamic environments due to their real-time rescheduling capabilities. Although challenges in scalability and interpretability persist, the current value of operations scheduling transcends classical optimization by incorporating resilience, traceability, and a high capacity for reacting to industrial disturbances. This research provides a comparative taxonomy of methods from various authors through a historical timeline.
