Authors: Henrique Ewbank; Peter Wanke; Henrique L. Correa
Addresses: COPPEAD Graduate School of Business, Federal University of Rio de Janeiro, Rua Pascoal Lemme, 355 – Cidade Universitária, Rio de Janeiro, RJ, Brazil ' COPPEAD Graduate School of Business, Federal University of Rio de Janeiro, Rua Pascoal Lemme, 355 – Cidade Universitária, Rio de Janeiro, RJ, Brazil ' Crummer Graduate School of Business, Rollins College, 1000 Holt Ave., Winter Park, FL 32789, USA
Abstract: This work solves a real-world multi-depot vehicle routing problem (MDVRP) with a homogeneous fleet and capacitated depots. A pipeline company wants to establish a vehicle policy in order to own part of its fleet and serve its customers for a period of one year. The company also wants to know the schedule of the visits for collecting ethanol from 261 producers and taking it to their three terminals located in Brazil. This problem presents uncertain demand, since weather conditions impact the final crop and uncertain depot capacity. Due to the vagueness of managers' speech, this problem also presents uncertain travel time. In this paper, fuzzy logic is used to model uncertainty and vagueness and to split the initial instance into smaller ones. Besides solving a real-world problem with fuzzy demand, fuzzy depot capacity and fuzzy travel time, this paper contributes with a decision making tool that reports different solutions for different uncertainty levels.
Keywords: multi-depot vehicle routing problem; MDVRP; fuzzy logic; job scheduling; real-world problem; expert system.
International Journal of Industrial and Systems Engineering, 2020 Vol.34 No.1, pp.65 - 83
Received: 30 Sep 2017
Accepted: 05 May 2018
Published online: 20 Dec 2019 *