TY - GEN
T1 - A sensitivity analysis for harmony search with multi-parent crossover algorithm
AU - Abu Doush, Iyad
AU - Santos, Eugene
N1 - Funding Information:
This research project is funded by the Dartmouth College and American University of Kuwait (Dartmouth-AUK) fellowship program.
Publisher Copyright:
© Springer Nature Switzerland AG 2020.
PY - 2020
Y1 - 2020
N2 - Harmony search algorithm with multi-parent crossover (HSA-MPC) is a hybrid algorithm that relies on benefiting from the crossover operation to combine more than one harmony to generate a new harmony. The picked harmonies are taken from an archive pool with best harmonies. In a previous study, the algorithm proves its efficiency when compared to other harmony search algorithms. In this paper, we will study the effect of harmony memory size (HMS), harmony memory consideration rate (HMCR), multi-parent crossover rate (MPCR), and the archive pool size on the quality of the generated solution. Eleven different scenarios are evaluated using a set of eight real-world numerical optimization problems introduced for CEC 2011 evolutionary algorithm competition. The analysis provides fixed values for all operators except the one under investigation. The obtained results prove the sensitivity of the algorithm to these operators and suggest a set of recommendations to improve the algorithm performance.
AB - Harmony search algorithm with multi-parent crossover (HSA-MPC) is a hybrid algorithm that relies on benefiting from the crossover operation to combine more than one harmony to generate a new harmony. The picked harmonies are taken from an archive pool with best harmonies. In a previous study, the algorithm proves its efficiency when compared to other harmony search algorithms. In this paper, we will study the effect of harmony memory size (HMS), harmony memory consideration rate (HMCR), multi-parent crossover rate (MPCR), and the archive pool size on the quality of the generated solution. Eleven different scenarios are evaluated using a set of eight real-world numerical optimization problems introduced for CEC 2011 evolutionary algorithm competition. The analysis provides fixed values for all operators except the one under investigation. The obtained results prove the sensitivity of the algorithm to these operators and suggest a set of recommendations to improve the algorithm performance.
KW - Evolutionary algorithms
KW - Harmony search algorithm
KW - Hybrid harmony search algorithm
KW - Numerical optimization
UR - http://www.scopus.com/inward/record.url?scp=85072835926&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-29516-5_21
DO - 10.1007/978-3-030-29516-5_21
M3 - Conference contribution
SN - 9783030295158
T3 - Advances in Intelligent Systems and Computing
SP - 276
EP - 284
BT - Intelligent Systems and Applications - Proceedings of the 2019 Intelligent Systems Conference IntelliSys Volume 1
A2 - Bi, Yaxin
A2 - Bhatia, Rahul
A2 - Kapoor, Supriya
PB - Springer Verlag
T2 - Intelligent Systems Conference, IntelliSys 2019
Y2 - 5 September 2019 through 6 September 2019
ER -