Diva Kurnianingtyas, Sza Sza Amulya Larasati, Agus Wahyu Widodo
Capacitated Vehicle Routing Problem (CVRP) is a VRP problem in logistics that focuses on minimizing transportation costs despite limited transport capacity. This study focuses on CVRP for the distribution of paving blocks using three vehicles with specific capacities: 1,250 units for vehicles 1 and 2 and 1,750 units for vehicle 3. The study implements five metaheuristic algorithms to solve CVRP: Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA), and Dragonfly Algorithm (DA). GA, ACO, and PSO are selected for their proven performance, while WOA and DA are included to explore new algorithm potentials. Experiments show that GA achieves the highest average vehicle mileage and consistent results due to its effective use of crossover and mutation. In addition, particularly PSO and DA, significantly improve average mileage and consistency compared to existing conditions. GA and ACO also show meaningful enhancements but with more significant variability. WOA, while capable of highly optimal solutions, requires improved consistency. The practical implication is that metaheuristics are more efficient than existing conditions. Enhancements to WOA and DA, such as parameter adjustments and adaptive mechanisms, can make them more competitive. Further research is needed to optimize these algorithms for real-world applications. This study confirms the significant potential of metaheuristic approaches to enhance logistic efficiency. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
Faculty of Computer Science, Universitas Brawijaya, Malang, Indonesia
