Solving a bi-objective mathematical model for location-routing problem with time windows in multi-echelon reverse logistics using metaheuristic procedure

Subject Areas : Mathematical Optimization

V. R.  Ghezavati ¹ (School of Industrial Engineering, Islamic Azad University, South Tehran Branch, Tehran, Iran)
M.  Beigi ² (School of Industrial Engineering, Islamic Azad University, South Tehran Branch, Tehran, Iran)

Received: 2020-10-04 Accepted : 2020-10-04 Published : 2016-12-01

Keywords: II, Reverse logistics network . Location, routing problem . Time window . Bi, objective model . e, Constraint method  NSGA,

Abstract :

During the last decade, the stringent pressures from environmental and social requirements have spurred an interest in designing a reverse logistics (RL) network. The success of a logistics system may depend on the decisions of the facilities locations and vehicle routings. The location-routing problem (LRP) simultaneously locates the facilities and designs the travel routes for vehicles among established facilities and existing demand points. In this paper, the location-routing problem with time window (LRPTW) and homogeneous fleet type and designing a multi-echelon, and capacitated reverse logistics network, are considered which may arise in many real-life situations in logistics management. Our proposed RL network consists of hybrid collection/inspection centers, recovery centers and disposal centers. Here, we present a new bi-objective mathematical programming (BOMP) for LRPTW in reverse logistic. Since this type of problem is NP-hard, the non-dominated sorting genetic algorithm II (NSGA-II) is proposed to obtain the Pareto frontier for the given problem. Several numerical examples are presented to illustrate the effectiveness of the proposed model and algorithm. Also, the present work is an effort to effectively implement the ε-constraint method in GAMS software for producing the Pareto-optimal solutions in a BOMP. The results of the proposed algorithm have been compared with the ε-constraint method. The computational results show that the ε-constraint method is able to solve small-size instances to optimality within reasonable computing times, and for medium-to-large-sized problems, the proposed NSGA-II works better than the ε-constraint.