Solving the Vehicle Routing Problem with Stochastic Demands with a Multiple Scenario Approach

Victor Pillac (1, 2) , Christelle Guéret (1) , Andrés Medaglia (2)

1 IRCCyN - Institut de Recherche en Communications et en Cybernétique de Nantes
2 COPA - Centro para la Optimización y Probabilidad Aplicada, Departamento de Ingeniería Industrial

Traditional approaches for the VRPSD aim at designing a-priori robust plans that avoid potential route failures. However, the widespread and inexpensive real-time communication and geolocalization technologies have opened promising perspectives in this field. We illustrate on the VRPSD the flexibility of jMSA, a generic framework for Multiple Scenario Approach. Preliminary results show that a continuous re-optimization leads to reductions in route failures and improvements in cost efficiency.

Mots clés: vehicle routing; dynamic vehicle routing; multiple scenario approach; vrpsd

Domaines: Recherche opérationnelle [math.OC]

Victor Pillac, Christelle Guéret, Andrés Medaglia. Solving the Vehicle Routing Problem with Stochastic Demands with a Multiple Scenario Approach. ALIO-INFORMS Joint International Meeting 2010, Jun 2010, Buenos Aires, Argentina. ⟨hal-00623472⟩