Container port truck dispatching optimization using Real2Sim based deep reinforcement learning

Jiahuan Jin, Tianxiang Cui, Ruibin Bai, Rong Qu

Abstract: Highlights•A Real2Sim deep reinforcement learning method was proposed for port container truck dispatch problem.•A novel dual-network architecture was used to boost the convergence and generalization.•Our method achieves the state-of-the-art results across diverse scenarios.•The Real2Sim module helps develop valuable operational insights and guidance under complex scenarios.