Accelerative Factor Based Spider Monkey Optimization
B. Bhagwanti, HARISH KUMAR SHARMA, Nirmala Sharma · 2018
Swarm intelligence (SI) based algorithms are very efficient and popular techniques since the last decade. Spider monkey optimisation (SMO) algorithm is a recent addition to the arena of SI. SMO is triggered from the food searching behaviour of spider monkeys that follow fission-fusion social system (FFSS). Although SMO has been performing very well sometimes there are some issues like stagnation, slow convergence etc. To overcome the above-mentioned issues, an efficient variant of SMO has been introduced. The proposed variant is entitled as Accelerative Factor based Spider Monkey Optimization (AFSMO) algorithm. In the proposed AFSMO to improve global convergence and get rid of stagnation, two phases of basic SMO, global leader phase, and local leader decision phase are modified by introducing an accelerative factor. This accelerative factor is decreasing the step size in an intelligent manner. Further, the efficiency and reliability of proposed AFSMO are validated over 15 different benchmark functions and Comparative study of the results of AFSMO is being done with several algorithms. The outcome of the proposed algorithm is clearly distinguishable and outperforming.