Three-learning strategy particle swarm optimization for air-ground collaborative logistics transportation scheduling...

Area of Science:

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

Background:

• Traditional logistics models face challenges in optimizing complex delivery and pickup scenarios.
• Integrating Unmanned Aerial Vehicles (UAVs) with ground vehicles offers potential for enhanced efficiency.
• Customer priority and cost factors are critical in logistics scheduling.

Purpose of the Study:

• To propose a novel air-ground collaborative logistics distribution model for the deliver-before-pickup scenario.
• To develop an intelligent optimization algorithm for solving the Air-ground Collaborative Logistics Transportation Scheduling Problem with Pickup and Delivery Considering Customer Priority (ALTSPPDCP).
• To minimize total travel costs by considering customer priority, vehicle costs, and UAV costs.

Main Methods:

• Formulation of the Air-ground Collaborative Logistics Transportation Scheduling Problem with Pickup and Delivery Considering Customer Priority (ALTSPPDCP).
• Design of a multi-layer, multi-stage encoding and decoding strategy.
• Implementation of the Three-Learning Strategy Particle Swarm Optimization (TSLPSO) algorithm, incorporating ascending order sorting and dynamic segmentation methods.

Main Results:

• The vehicle-UAV collaborative schema achieved a 14.86% lower total cost compared to a vehicle-only schema in a 50-node scenario.
• The TSLPSO algorithm demonstrated superior performance, reducing costs by 39.99% compared to Particle Swarm Optimization (PSO) and 27.94% compared to Randomized Particle Swarm Optimization (RPSO).

Conclusions:

• Air-ground collaborative logistics significantly reduces operational costs in complex delivery scenarios.
• The proposed TSLPSO algorithm is an effective and efficient method for solving the ALTSPPDCP.
• Intelligent optimization algorithms provide superior solutions for logistics scheduling problems compared to traditional methods.