DOI: 10.4018/IJAEC.2018010101
International Journal of Applied Evolutionary Computation
Volume 9 • Issue 1 • January-March 2018
Copyright © 2018, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
1
Efcient Golden-Ball Algorithm
Based Clustering to solve the Multi-
Depot VRP With Time Windows
Lahcene Guezouli, University of Batna 2, Batna, Algeria
Mohamed Bensakhria, University of Batna 2, Batna, Algeria
Samir Abdelhamid, University of Batna 2, Batna, Algeria
ABSTRACT
In this article, the authors propose a decision support system which aims to optimize the classical
Capacitated Vehicle Routing Problem by considering the existence of multiple available depots and
a time window which must not be violated, that they call the Multi-Depot Vehicle Routing Problem
with Time Window (MDVRPTW), and with respecting a set of criteria including: schedules requests
from clients, the capacity of vehicles. The authors solve this problem by proposing a recently published
technique based on soccer concepts, called Golden Ball (GB), with different solution representation
from the original one, this technique was designed to solve combinatorial optimization problems, and
by embedding a clustering algorithm. Computational results have shown that the approach produces
acceptable quality solutions compared to the best previous results in similar problem in terms of
generated solutions and processing time. Experimental results prove that the proposed Golden Ball
algorithm is efficient and effective to solve the MDVRPTW problem.
KEyWoRDS
Clustering, Golden Ball Algorithm, Multi-Depot Vehicle Routing Problem, Routing, Scheduling
1. INTRoDUCTIoN
Within the wide scope of logistics management, transportation plays a central role and is a crucial
activity in the delivery of goods and services. The transport problem is one of the mainly essential
combinatorial optimization problems that have taken the interest of several researchers. Huge research
efforts have been devoted to the study of logistic problems and thousands of papers have been written
on many variants of this problem such as Traveling Salesman Problem (TSP) (Toth, 2001), Vehicle
Routing Problem (VRP) and supply chain management (SCM) (Pisinger, 2007).
The Vehicle Routing Problem (VRP) has been one of the central topics in optimization since
Dantzig proposed the problem in 1959(Dantzig, 1959). A simple general model of VRP can be
described as follows: a set of service vehicles need to visit all customers in a geographical region with
the minimum cost. The VRP is also named Single-depot VRP (SDVRP). In cases with more than one
depot, VRPs are known as multi-depot VRPs (MDVRP) (Filipec, 2000), (Mirabi, 2014). Single-depot
VRPs are not suitable for practical situations though they have attracted researchers in a wide sense.
The Multi-Depot Vehicle Routing Problem (MDVRP) is a generalization of SDVRP in which
the vehicles start from multiple depots and return to their original depots at the end of their assigned
tours. As there are a large number of depots, it is a difficult task for decision makers to determine