Island-based harmony search for optimization problems. Issue 4 (March 2015)

• Record Type: Journal Article
• Authors: Al-Betar, Mohammed Azmi; Awadallah, Mohammed A.; Khader, Ahamad Tajudin; Abdalkareem, Zahraa Adnan
• Is Part Of: Expert systems with applications. Volume 42:Issue 4(2015)
• Page Start: 2026
• Page End: 2035
• Publication Date: 2015-03
• Subjects: Harmony search -- Island model -- Structured population -- Diversity; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2014.10.008
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Island model are embedded within framework of HS algorithm to build i HS. Island number ( I n ), migration frequency ( F m ), migration rate ( R m ) are i HS parameters. Higher values of I n and F m lead to better results but R m value needs to be small. Results outperform others due to manipulating diversity through island concepts. Abstract: Harmony search (HS) algorithm is a recent meta-heuristic algorithm that mimics the musical improvisation concepts. This algorithm has been widely used for solving optimization problems. Moreover, many modifications in this algorithm have been carried out in order to improve the performance of the search. Island model is a structured population mechanism used in evolutionary algorithms to preserve the diversity of the population and thus improve the performance. In this paper, the island model concepts are embedded into the main framework of HS algorithm to improve its convergence properties where the new method is refer to as island HS ( i HS). In the proposed method, the individuals in population are distributed into separate sub-population named (islands). Then the breeding loop is separately involved in each island. After specific generations, a number of individuals run an exchange through a process called migration. This process is performed to keep the diversity of population and to allow islands to interact with each other. The experimental result using a set of benchmark function shows that the island model context is crucial to the performance of i HS. Finally the sensitivity analysis and the comparative study for i HS prove the efficiency of the island model.