The fleet size and mix dial-a-ride problem with reconfigurable vehicle capacity

Oscar Augusto Tellez Sanchez (1) , Samuel Vercraene (1) , Fabien Lehuédé (2, 3) , Olivier Péton (2, 3) , Thibaud Monteiro (1)

1 DISP - Décision et Information pour les Systèmes de Production
2 IMT Atlantique - DAPI - Département Automatique, Productique et Informatique
3 LS2N - équipe SLP - Systèmes Logistiques et de Production

This paper introduces a fleet size and mix dial-a-ride problem with multiple passenger types and a heterogeneous fleet of reconfigurable vehicles. In this new variant of the dial-a-ride problem, en-route modifications of the vehicle's inner configuration are allowed. The main consequence is that the vehicle capacity is defined by a set of configurations and the choice of vehicle configuration is associated with binary decision variables. The problem is modeled as a mixed-integer program derived from the model of the heterogeneous dial-a-ride problem. Vehicle reconfiguration is a lever to efficiently reduce transportation costs, but the number of passengers and vehicle fleet setting make this problem intractable for exact solution methods. A large neighborhood search metaheuristic combined with a set covering component with a reactive mechanism to automatically adjust its parameters is therefore proposed. The resulting framework is evaluated against benchmarks from the literature, used for similar routing problems. It is also applied to a real case, in the context of the transportation of disabled children from their home to medical centers in the city of Lyon, France.

Mots clés: large neighborhood search; reconfigurable vehicles; Fleet Size and Mix Problem; set covering; check; feasibility; Dial-A-Ride Problem

Domaines: Recherche opérationnelle [math.OC]

Oscar Augusto Tellez Sanchez, Samuel Vercraene, Fabien Lehuédé, Olivier Péton, Thibaud Monteiro. The fleet size and mix dial-a-ride problem with reconfigurable vehicle capacity. [Research Report] Laboratoire DISP. 2017. ⟨hal-01619103v1⟩