Skip to content

Author

Shaokai An

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Aug 2026

Dynamic Scheduling Implementation of Reverse Logistics Network Simulation Based on Deep Reinforcement Learning

Driven by the dual momentum of the circular economy and digital transformation, reverse logistics serves as a critical nexus for resource regeneration while facing volatile recycling volumes, fragmented collection networks, and the intricate sorting constraints inherent to fiber recovery. With the growing demand for intelligent information interaction and data transmission in modern industrial systems, including electromagnetic-enabled sensing and communication infrastructures, traditional static scheduling modes are increasingly unable to adapt to dynamic operational environments. Deep reinforcement learning (DRL) provides an effective technological pathway for dynamic scheduling of reverse logistics networks through its strong capability for adaptive decision-making and environment-aware optimization. Centering on the integrated framework of simulation modeling, algorithm optimization, dynamic scheduling, and empirical verification, this study employs deep reinforcement learning to optimize the dynamic scheduling of reverse logistics networks, specifically addressing the nonlinear characteristics of fiber feedstock recovery and multi-grade material flows. The proposed approach enhances scheduling adaptability and resource coordination efficiency while offering a practical reference for intelligent logistics management and information-driven optimization in advanced industrial communication environments.

Jing Zhang, Shaokai An · 0 citations