Interterminal Truck Routing Optimization Using Deep Reinforcement Learning

Area of Science:

• Logistics and Supply Chain Management
• Artificial Intelligence
• Operations Research

Background:

• Global containerized transport growth necessitates improved port productivity through interconnected terminals.
• Multiterminal systems increase complexity and demand for interterminal transport (ITT) via trucks.
• Optimal truck routing is crucial for coordinating ITT flows and enhancing port efficiency.

Purpose of the Study:

• To propose a deep reinforcement learning approach for optimizing truck routing in ITT.
• To consider key factors like origin, destination, time windows, and due dates for routing decisions.
• To evaluate the effectiveness of deep reinforcement learning in this context.

Main Methods:

• Development of a deep reinforcement learning model for truck routing optimization.
• Integration of critical order parameters: origin, destination, time window, and due date.
• Comparative analysis against existing truck routing problem-solving approaches.

Main Results:

• The proposed deep reinforcement learning method demonstrated superior performance.
• Significant improvements in truck routing efficiency were observed.
• The algorithm effectively handled complex ITT scenarios.

Conclusions:

• Deep reinforcement learning offers a promising solution for optimizing truck routing in complex port environments.
• The developed method enhances interterminal transport efficiency and overall port productivity.
• Further research in deep reinforcement learning for logistics optimization is warranted.