Algoritmos genéticos en la asignación de recursos y optimización de rutas en transporte: revisión sistemática

Abstract

Operational scheduling in transportation companies involves the efficient assignment of vehicles, drivers, routes, and schedules, a highly complex combinatorial optimization problem when capacity constraints, time windows, costs, emissions, and demand variability are considered. In this context, genetic algorithms (GA) and their hybrid variants have been used as metaheuristics to obtain good-quality solutions within reasonable computational times. This systematic review, conducted according to Kitchenham's methodology and reported through a flow diagram adapted from PRISMA 2020, aimed to identify the types of resource allocation and route optimization problems addressed with GA, the predominant variants, the quantitative results reported, and the main limitations. Articles published between 2022 and 2026 were reviewed from ScienceDirect, Springer Nature Link, IEEE Xplore, MDPI, and other academic publishing platforms with verifiable DOI; the final selection comprised 31 studies. The results show a predominance of the Vehicle Routing Problem and its variants (80.65%), together with a strong presence of hybrid GA, multi-objective strategies, and adaptive operators. It is concluded that GA are relevant for operational transportation scheduling, although challenges remain regarding parameter tuning, scalability, premature convergence, and validation in real-world scenarios.