A multi-objective hybrid genetic algorithm to minimize the total cost and delivery tardiness in a reverse logistics

• Published in: Multimedia tools and applications Vol. 74; no. 20; pp. 9067 - 9085
• Main Authors: Lee, Jeong-Eun, Chung, Kyung-Yong, Lee, Kang-Dae, Gen, Mitsuo
• Format: Journal Article
• Language: English
• Published: New York Springer US 01-10-2015; Springer Nature B.V
• Subjects: Alliances; Computer Communication Networks; Computer Science; Corporate image; Cost engineering; Costs; Customer services; Data Structures and Information Theory; Distilleries; Fuzzy logic; Genetic algorithms; Integer programming; Inventory control; Literature reviews; Logistics; Marketing; Mathematical analysis; Mathematical models; Molybdenum; Multimedia Information Systems; Optimization; Propensity to consume; Reusable; Reverse logistics; Special Purpose and Application-Based Systems; Statistical analysis; Supply chain management; Supply chains; Busan South Korea; South Korea; Fuzzy logic controller (FLC); Pri-awGA (priority-based GA with adaptive weight approach); Multi-objective reverse logistics network (mo-RLN); Multi-objective hybrid genetic algorithm (mo-hGA)
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-013-1594-6

In the recent environmental protection the reverse logistics of the used product is one of the most important research topics. The reverse logistics is the process flow of used-products that are collected to be reproduced so that they can be sold again to customers after some processing. We propose a multi-objective hybrid genetic algorithm (mo-hGA) combined with Fuzzy Logic Controller (FLC) for efficiently dealing with multi-objective reverse logistics network (mo-RLN) problem. The aim of this paper is firstly to formulate mo-RLN model, and secondly to optimize it by mo-hGA method proposed with reusable system configuration. In particular two objective functions to be minimized total costs of mo-RLN, (i.e. fixed opening cost, transportation cost and inventory cost) and also minimized delivery tardiness in all periods are considered in the model. We will clear each objective function (i.e. total costs and total delivery tardiness), computational time and number of Pareto solutions with LINGO, pri-awGA (priority-based GA with adaptive weight approach) and mo-hGA proposed with numerical examples. For demonstrating the effectiveness of the proposed model, we evaluate with the numerical examples and simulate it with a bottles distilling/sale company as a case study in Busan, Korea.