Resumen

ALVAREZ-CRUZ, Cesar Dario; MUNARI, Pedro Augusto  y  MORABITO, Reinaldo. Bounds for the vehicle allocation problem. Dyna rev.fac.nac.minas [online]. 2019, vol.86, n.208, pp.329-335. ISSN 0012-7353.  https://doi.org/10.15446/dyna.v86n208.68504.

The Vehicle Allocation Problem (VAP) consists in allocating a fleet of vehicles to attend the expected demand for freight transportation between terminals along a finite multiperiod planning horizon. The objective is to maximize the profit generated for the completed services. Given the geographical dispersion of freight transportation demand services, it is common that some vehicles accumulate where they are not needed or they lack where they are indeed needed, thus it is important to balance the supply of vehicles and demand for services along the planning horizon. The size of practical problems encountered by logistic transportation companies are significantly large to solve in reasonable computational times, especially in road freight transportation. Consequently, heuristic methods are used to obtain feasible solutions, however, without a quality certificate of the solution. For this reason, this work applies two decomposition methods, which provides quality certificate of the solution, to obtain bounds in good computational times. The methods used are lagrangean relaxation with subgradient optimization and Dantzig-Wolfe decomposition with column generation. Computational experiments with realistic instances show great potential of the bounds in terms of computational efficiency for the second method.

Palabras clave : optimization; vehicle allocation; Dantzig-Wolfe; column generation; lagrangean relaxation.

        · resumen en Español     · texto en Inglés     · Inglés ( pdf )