DocumentCode :
3496013
Title :
Solving Traveling Salesman Problem by a hybrid combination of PSO and Extremal Optimization
Author :
Khakmardan, S. ; Poostchi, H. ; Akbarzadeh-T, M.-R.
Author_Institution :
Dept. of Artificial Intell., Islamic Azad Univ., Mashhad, Iran
fYear :
2011
fDate :
July 31 2011-Aug. 5 2011
Firstpage :
1501
Lastpage :
1507
Abstract :
Particle Swarm Optimization (PSO) has received great attention in recent years as a successful global search algorithm, due to its simple implementation and inexpensive computation overhead. However, PSO still suffers from the problem of early convergence to locally optimal solutions. Extremal Optimization (EO) is a local search algorithm that has been able to solve NP hard optimization problems. The combination of PSO with EO benefits from the exploration ability of PSO and the exploitation ability of EO, and reduces the probability of early trapping in the local optima. In other words, due to the EO´s strong local search capability, the PSO focuses on its global search by a new mutation operator that prevents loss of variety among the particles. This is done when the particle´s parameters exceed the problem conditions. The resulting hybrid algorithm Mutated PSO-EO (MPSO-EO) is then applied to the Traveling Salesman Problem (TSP) as a NP hard multimodal optimization problem. The performance of the proposed approach is compared with several other metaheuristic methods on 3 well known TSP databases and 10 unimodal and multimodal benchmark functions.
Keywords :
computational complexity; convergence; mathematical operators; particle swarm optimisation; search problems; travelling salesman problems; NP hard multimodal optimization problem; convergence; extremal optimization; global search algorithm; local search algorithm; mutated PSO-EO; mutation operator; particle swarm optimization; traveling salesman problem; Benchmark testing; Convergence; Databases; Flowcharts; Optimization; Particle swarm optimization; Traveling salesman problems;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks (IJCNN), The 2011 International Joint Conference on
Conference_Location :
San Jose, CA
ISSN :
2161-4393
Print_ISBN :
978-1-4244-9635-8
Type :
conf
DOI :
10.1109/IJCNN.2011.6033402
Filename :
6033402
Link To Document :
بازگشت