.png){kind=icon}
Return
Genetic algorithm-based optimization for multi-resource dispatch planning in liquid freight transportation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jestch.2026.102535.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Efficient dispatch planning is critical in liquid freight transportation, where multiple resources-drivers, vehicles, trailers, and ISO tank containers-must be optimally assigned to transportation requests while considering operational constraints and regulatory requirements. To address this challenge, this study develops an integrated optimization framework that combines a Mixed-Integer Linear Programming (MILP) model with an enhanced Genetic Algorithm (GA), incorporating a novel Food Chain-Based Multi-Parent Selection (FCBS-MP) mechanism to improve solution diversity and convergence speed. The model minimizes total costs, including delayed assignment penalties and fuel consumption, while ensuring balanced utilization of available resources. Extensive experimental results on real-world datasets of varying scales, with 30 independent runs per setting, demonstrate that FCBS-MP significantly improves dispatch planning performance. Relative to manual expert planning, FCBS-MP achieves up to a 63% reduction in driver-distance imbalance and up to an 85% reduction in ISO-tank utilization imbalance, together with a 94-98% reduction in contractual short-haul quota violations in the medium and large instances. Relative to the average performance of conventional crossover operators, it reduces driver- and vehicle-distance imbalance by approximately 49% and 38%, respectively. These results demonstrate the effectiveness and stability of the proposed approach, particularly for large-scale logistics instances, and support GA-based optimization as a scalable and robust alternative for complex multi-resource dispatch planning in liquid freight transportation.
Keywords:
Genetic algorithm
Liquid freight transportation
Logistics
Mixed integer programming model
Intelligent transportation
Resource Allocation Problem
Journal
E
IF:
5.4
Papers:
1.4K
Citations:
6.3K
Organization
Cited Papers
No cited papers available