File(s) under permanent embargo
A GA-PSO layered encoding evolutionary approach to 0/1 knapsack optimization
journal contribution
posted on 2010-08-01, 00:00 authored by S Neoh, N Morad, Chee Peng LimChee Peng Lim, Z AzizThis paper presents a layered encoding cascade evolutionary approach to solve a 0/1 knapsack optimization problem. A layered encoding structure is proposed and developed based on the schema theorem and the concepts of cascade correlation and multi-population evolutionary algorithms. Genetic algorithm (GA) and particle swarm optimization (PSO) are combined with the proposed layered encoding structure to form a generic optimization model denoted as LGAPSO. In order to enhance the finding of both local and global optimum in the evolutionary search, the model adopts hill climbing evaluation criteria, feature of strength Pareto evolutionary approach (SPEA) as well as nondominated spread lengthen criteria. Four different sizes benchmark knapsack problems are studied using the proposed LGAPSO model. The performance of LGAPSO is compared to that of the ordinary multi-objective optimizers such as VEGA, NSGA, NPGA and SPEA. The proposed LGAPSO model is shown to be efficient in improving the search of knapsack’s optimum, capable of gaining better Pareto trade-off front.
History
Journal
International journal of innovative computing, information and controlVolume
6Issue
8Pagination
3489 - 3505Publisher
ICIC InternationalLocation
Kumamoto, JapanISSN
1349-4198Language
engPublication classification
C1.1 Refereed article in a scholarly journalCopyright notice
2010, ICIC InternationalUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC