First EAI International Conference on Computer Science and Engineering

Research Article

Hybridizing Bat Algorithm with Modified Pitch-Adjustment Operator for Numerical Optimization Problems

Download679 downloads
  • @INPROCEEDINGS{10.4108/eai.27-2-2017.152269,
        author={Waheed Ali H. M. Ghanem and Aman Jantan},
        title={Hybridizing Bat Algorithm with Modified Pitch-Adjustment Operator for Numerical Optimization Problems},
        proceedings={First EAI International Conference on Computer Science and Engineering},
        publisher={EAI},
        proceedings_a={COMPSE},
        year={2017},
        month={2},
        keywords={Bat algorithm; Harmony search algorithm; Global Optimization problem; Pitch adjustment operator},
        doi={10.4108/eai.27-2-2017.152269}
    }
    
  • Waheed Ali H. M. Ghanem
    Aman Jantan
    Year: 2017
    Hybridizing Bat Algorithm with Modified Pitch-Adjustment Operator for Numerical Optimization Problems
    COMPSE
    EAI
    DOI: 10.4108/eai.27-2-2017.152269
Waheed Ali H. M. Ghanem1,*, Aman Jantan
  • 1: School of Computer Science, Universiti Sains Malaysia, Pulau Pinang, Malaysia
*Contact email: waheed.ghanem@gmail.com

Abstract

This article introduces a new metaheuristic approach that is a hybrid of two known algorithms, for solving global optimization problems. The proposed algorithm is based on the Bat Algorithm (BA), which is inspired by the micro-bat echolocation phenomenon, and addresses the problems of local-optima trapping and low precision using an adjusted mutation operator from the Harmony Search (HS) algorithm. The proposed Hybrid Bat Harmony (HBH) algorithm attempts to balance the good exploitation process of BA with a fast exploration feature inspired by HS. The design of HBH is introduced and its performance is evaluated against fourteen of the standard benchmark functions, and compared to that of the standard BA and HS algorithms and to another recent hybrid algorithm (HS/BA). The obtained results show that the new HBH method is indeed a promising addition to the arsenal of metaheuristic algorithms and can outperform the original BA and HS algorithms.