Data-Driven Route Optimization for Large-Scale Transportation Systems
TL;DR
The findings demonstrate that metaheuristic techniques consistently outperform traditional algorithms in complicated, constraint-rich situations and emphasize the need of cost-effective, data-driven metaheuristic optimization in current logistics planning.
Abstract
The study investigates the use of data-driven optimization strategies to increase routing efficiency in large-scale transportation systems with realistic constraints on operation. The objectives are to combine network cost data with advanced optimization techniques, assess routing performance under capacity, time, and cost constraints. It also compares traditional shortest-path methods to heuristic and metaheuristic approaches. A computational framework was created to compare several algorithms on a large-scale transportation dataset while accounting for real-world constraints such as vehicle capacity and travel costs. Performance was evaluated using total routing cost and route efficiency. The findings demonstrate that metaheuristic techniques consistently outperform traditional algorithms in complicated, constraint-rich situations. Genetic Algorithms, in particular, achieved greater cost reduction and continued to perform well as the problem scale rose. Classical algorithms performed well only in simple circumstances. These findings emphasize the need of cost-effective, data-driven metaheuristic optimization in current logistics planning and encourage future study into hybrid and dynamic routing systems.