Article Preview
Top2. Routing Algorithms And Reverse Logistics Network
The idea of reverse logistics has been developed by businesses for planning their materials returns. There are different reasons that motivated the development of research in this area such as environmental conservation, customer demand and economically driven opportunities to reuse products (Dekker, 2004).
Reverse logistics is defined as the process of planning, developing, and efficiently controlling the flow of materials, products, and information from the place of origin to the place of consumption in such a way that while satisfying the consumer’s needs, the available remaining material is managed to be reintroduced into the supply chain, obtaining an added value and/or if not possible procuring a suitable disposal of this remaining material Rubio (2003). One of the main factors in reverse logistics networks is to have enough quantity of returns to ensure a stable flow of material in the recycling and returns network.
The prototypical formulation for vehicle routing models is known as Capacitated Vehicle Routing Problem (CVRP). It is well known that this problem is NP-hard with optimal solutions to only a small to medium problem instances. For these problems, heuristics are frequently used for larger instances to reach solutions close to optimal.
For the case we are dealing in this paper, the problem can be classified as a Cumulative Capacitated Vehicle Routing Problem with Multiple Trips with Time Windows (CCVRPMTW) which is a variant of the Vehicle Routing Problem that also include the following particularities:
- 1.
Vehicles can be deployed in more than one route
- 2.
Time windows constraints
- 3.
Finite vehicles capacity
- 4.
Single visit to all clients based on profit
- 5.
Asymmetric geographic distribution of trips
Top3. Literature Review
The vehicle routing problem described in this paper has been studied partially in the last years.
A hybrid metaheuristic algorithm to tackle a districting problem with pickups and deliveries over a determined service region has been developed by González-Ramírez et al. (2011). The problem studied in this paper also deals with a collecting region divided into districts in which each district is served by vehicles departing from only one central depot. Geetha, Poonthalir and Vanathi (2010) developed a population-based search algorithm using Particle Swarm Optimization (PSO) with crossover and mutation operators, for the VRP problem. The main difference is that Geetha, Poonthalir and Vanathi. (2010) focus on a symmetric problem from the cost perspective, while this study contemplates an asymmetric cost scenario. Additionally, Geetha, Poonthalir and Vanathi (2010) consider only traveling costs, while the PVA also considers operations costs, time costs and revenues.