Title: Modified Adaptive Cuckoo Search (MACS) algorithm and formal description for global optimisation
Authors: Yongwei Zhang; Lei Wang; Qidi Wu
Addresses: College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China. ' College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China. ' College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China
Abstract: Bio-inspired algorithms, through imitating the regular pattern of life forms, often produce unexpected results. A novel global optimisation algorithm, Cuckoo Search (CS), is an example that simulates the brood behaviour of some species of cuckoos. By using Lévy distribution, the flying pattern of cuckoos is also imitated. However, the potential of cuckoo's search pattern is not fully discovered in CS algorithm. In this article, we introduce the CS algorithm and associated Lévy flights. A Modified Adaptive Cuckoo Search (MACS) is then proposed by introducing grouping, parallel, incentive, adaptive and information-sharing characteristics. Also, the formal descriptions of improving strategies are given. The proposed algorithm improves the basic CS algorithm without losing the characteristic of high-efficiency search of Lévy flights. Experiment results show that MACS outperforms basic CS algorithm on most test problems and possesses application potential for real-world problems.
Keywords: cuckoo search; global optimisation; Levy flights; swarm intelligence; foraging strategy; population diversity; bio-inspired computation; formal description.
International Journal of Computer Applications in Technology, 2012 Vol.44 No.2, pp.73 - 79
Available online: 23 Aug 2012 *Full-text access for editors Access for subscribers Purchase this article Comment on this article