Fuzzy optimization model for electric vehicle routing problem with time windows and recharging stations. (1st May 2020)

• Record Type: Journal Article
• Authors: Zhang, Shuai; Chen, Mingzhou; Zhang, Wenyu; Zhuang, Xiaoyu
• Is Part Of: Expert systems with applications. Volume 145(2020)
• Publication Date: 2020-05-01
• Subjects: Vehicle routing problem -- Credibility theory -- Electric vehicle -- Fuzzy simulation -- Adaptive large neighborhood search; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2019.113123
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Investigated a novel electric VRP considering uncertain environment. Proposed a fuzzy optimization model with fuzzy simulation to the presented problem. Integrated ALNS algorithm and VND algorithm to solve the presented model. Performed experiments to verify the effectiveness of the proposed algorithm. Abstract: As fuel prices increase and emission regulations become increasingly strict, electric vehicles have been used in various logistics distribution activities. Most studies have focused on the electric vehicle routing problem under a deterministic environment, neglecting the effects of uncertain factors in practical logistics distribution. Thus, a novel fuzzy electric vehicle routing problem with time windows and recharging stations (FEVRPTW) is investigated in this study, and a fuzzy optimization model is established based on credibility theory for this problem. In the presented model, fuzzy numbers are used to denote the uncertainties of service time, battery energy consumption, and travel time. Moreover, the partial recharge is allowed under the uncertain environment. To solve the model, an adaptive large neighborhood search (ALNS) algorithm enhanced with the fuzzy simulation method is proposed. In the proposed ALNS algorithm, four new removal algorithms are designed and integrated for addressing the FEVRPTW. To further improve the algorithmic performance, the variable neighborhood descent algorithm is embedded into the proposed ALNS algorithm and five local search operators are applied. The experiments were conducted to verify the effectiveness of the proposed ALNS algorithm for solving the presented model.