A. Agustín Martín, Á. A. Juan Pérez, J. Faulín Fajardo, M. Á. Llorente García

In this work, an alternative procedure for solving vehicle routing problems (VRP) with different types of vehicle capacities, say Heterogeneous Vehicle Routing Problem (HVRP), has been proposed to deal with the HVRP. The main issue for VRP is to serve node demands with a vehicle fleet located in a depot and minimize the routes costs. Our approach is based on the Successive Approximations Method (SAM) which solves, in each iteration, the Capacitated Vehicle Routing Problem (CVRP) for the non- served nodes and the maximum capacity of unused vehicles. Any efficient algorithm can be used to solve the CVRP and the so-called Simulation Routing via Generalized Clark and Wright (SR-GCW) is applied in this work. Then, successively, the largest routes that hold from the CVRP solution are saved as a partial solution for the HVRP. Finally, regarding problem benchmarks, competitive results are obtained in reasonable computing times.

Palabras clave: heterogeneous vehicle routing problem, combinatorial optimization, heuristics, local search

Programado

XA5 Algoritmos metaheurísticos
18 de abril de 2012  09:00
Sala Viena

Heurísticas para el problema de planificación de la extracción de bloques en minas a cielo abierto

E. Jelvez, N. Morales, J. Peypouquet, P. Reyes Valenzuela

Algoritmo genético para el problema de secuenciación en máquinas paralelas con tiempos de cambio y minimización del retraso/adelanto ponderado

E. Vallada Regalado, R. Ruiz García