A VND-Based Heuristic for the Heterogeneous Load Allocation Problem in Road Transportation

This work proposes a heuristic for the heterogeneous load allocation problem in road transportation, extending a previously developed exact formulation. The approach combines a partially greedy initial construction with an improvement phase based on VND (variable neighborhood descent). The heuristic is designed to preserve the logic of the objective function and the constraints of the mathematical model, allowing direct comparison between both approaches. Computational experiments consider literature-based and author-generated instances, comparing the heuristic with the exact model in terms of solution quality and computational time. The results show that the heuristic substantially reduces computational effort and, for several instances, obtains solutions close to those of the exact model. However, larger average differences and greater variability were observed in some of the more demanding instances.