Agent-based Evolutionary Cooperative Approach for Dynamic Multi-Hop Ridematching Problem - Université Polytechnique des Hauts-de-France Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Agent-based Evolutionary Cooperative Approach for Dynamic Multi-Hop Ridematching Problem

Résumé

A Dynamic Multi-hop Ridesharing is a flexible form in which the request can be matched to one or more drivers at different times such that the rider reaches his destination hop by hop. This kind of ridesharing offers more possibilities for the riders. The Multi-Hop Ridematching Problem (M - HRP) in its dynamic context is considered to be NP-Complete. Therefore, we mainly focus on suitable concepts to reduce real time ridematching's high complexity. In this context, we aim to set up a distributed architecture through which an optimizing decentralized process will be performed. In this study, we developed an original distributed approach based on evolutionary cooperative algorithms driven by agents' communication. The cooperation aspect appears into two levels: In the agent architecture level to ensure communication, negotiation and collaboration between agents, and in the algorithmic level to ensure consistency and persistence of the cooperation between these agents. © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

Dates et versions

hal-03659157 , version 1 (04-05-2022)

Identifiants

Citer

Sondes Ben Cheikh, Slim Hammadi, Christian Tahon. Agent-based Evolutionary Cooperative Approach for Dynamic Multi-Hop Ridematching Problem. 15th IFAC Symposium on Information Control Problems in Manufacturing, INCOM 2015, May 2015, Ottawa, Canada. pp.887-892, ⟨10.1016/j.ifacol.2015.06.195⟩. ⟨hal-03659157⟩
12 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More