Performance Evaluation of Capuchin Search Algorithm Through Non-linear Problems, and Optimization of Gear Train Design Problem

The purpose of this paper is to demonstrate the superiority of the Capuchin Search Algorithm (CapSA), a metaheuristic, in competitive environments and its advantages in optimizing engineering design problems. To achieve this, the CEC 2019 function set was used. Due to the challenging characteristics of the CEC 2019 function set in reaching a global solution, it effectively showcases the algorithm's quality. For this comparison, sea-horse optimizer (SHO), grey wolf optimizer (GWO), sine-cosine algorithm (SCA), and smell agent optimization (SAO) were chosen as current and effective alternatives to the CapSA algorithm. Furthermore, the gear train design problem (GTD) was selected as an engineering design problem. In addition to the CapSA algorithm, a hybrid of SCA and GWO algorithm (SC-GWO) and genetic algorithm (GA) were chosen as alternatives for optimizing this problem. The performance superiority and optimization power of the CapSA algorithm were assessed using statistical metrics and convergence curves, then compared with alternative algorithms. Experimental results conclusively demonstrate the significant effectiveness and advantages of the CapSA algorithm.