AI-Driven Digital Twin Framework for Warehouse Logistics Automation Optimization
Summary
- Researchers proposed an AI-driven digital twin framework to optimize warehouse logistics and dynamic product placement.
- The system integrates a reinforcement learning agent within a FlexSim-based digital twin environment using the Asset Administration Shell framework.
- Experiments demonstrated a 29.66 percent reduction in average picking time and a 35.42 percent increase in system throughput.
- The approach also achieved an 18.54 percent decrease in travel distance in a modeled four-zone warehouse layout.
Why it matters
Combining reinforcement learning with digital twins provides a practical approach to significantly improve warehouse operational efficiency and material handling throughput.