Abstract
This paper proposes a self-organizing hierarchical monkey algorithm (SHMA) with a time-varying parameter to improve the performance of the original monkey algorithm (MA). In the proposed SHMA, we adopt a hierarchical structure to organize the climb, watch, and somersault operations and apply a self-organizing mechanism to coordinate these operations. Moreover, a time-varying parameter is employed to adjust the exploration ability and exploitation ability during the optimization process. The SHMA also applies the fitness information of solutions to guide the optimization process and introduces a selection operator, a fitness-based replacement operator, and a repulsion operator into the climb, watch and somersault operations, respectively. To investigate the performance of the SHMA, we compare it with eight different metaheuristic algorithms on 30 benchmark problems and four real-world optimization problems. The simulation results show that the SHMA exhibits better overall performance than the eight compared algorithms
Original language | English |
---|---|
Pages (from-to) | 3245–3263 |
Number of pages | 18 |
Journal | Neural Computing and Applications |
Volume | 31 |
DOIs | |
Publication status | Published - 10 Nov 2017 |
Externally published | Yes |
Keywords
- self-organizing hierarchical monkey algorithm
- monkey algorithm
- metaheuristic algorithm
- global optimization
- time-varying parameter