A comparative analysis of preying behavior-based metaheuristic algorithms for optimization of laser beam drilling processes

Increasing demand for miniaturized components as micro-holes with diameters ranging from a few microns to hundreds of microns used in automobile, aviation, electronics, and medical industries makes laser beam drilling (LBD) a viable option in today's manufacturing environment. The LBD processes must be optimized for the best settings of their different input parameters to achieve enhanced performance and the most desired micro-hole characteristics. The earlier researchers have already attempted numerous metaheuristic algorithms. In this paper, the applicability of five robust preying behavior-based metaheuristic algorithms, i.e., sailfish optimizer, Harris hawks optimizer, Aquila optimizer, bat optimizer, and grey wolf optimizer, is validated while optimizing two LBD operations. The first example considers CO2 laser-based micro-drilling on polycarbonate using laser power, exposure time, and focal plane position as the input variables, and hole depth, hole diameter, and heat-affected zone as the outputs. The second example deals with micro-drilling on stainless steel with gas pressure, current, cutting speed, and pulse frequency as the process parameters, and circularity at entry and exit, and hole taper angle as the responses. Although all the considered preying behavior-based metaheuristics are very robust and have excellent search ability, there has been a lack of comparative analysis of their relative performance while solving complex optimization problems with minimum computational effort. Their optimization performance is compared with respect to quality of the solutions, convergence speed and computational effort, identifying grey wolf optimizer as the most efficient one. The results of Friedman's mean rank test also prove the superiority of grey wolf optimizer over its competitors.