Bombardier Beetle Optimizer (BBO) for Optimizing Engineering Design Problems

Abstract

This paper provides a comprehensive comparative study of various metaheuristic optimization algorithms with the intention of emphasizing the most adequate method for optimizing well-established design problems of engineering. The algorithms chosen for this study include “Bombardier Beetle Optimizer (BBO)”, “Particle Swarm Optimization (PSO)”, and “Grey Wolf Optimizer (GWO)”. The experimental results of this paper are extracted in two phases. In the first phase, these algorithms are utilized to optimize the “Congress on Evolutionary Computation (CEC) 2017” test bed suite. While in the second phase, they are applied to solve the well-known problems of engineering design, such as Speed Reducer Design, Pressure Vessel Design, Cantilever Beam Design, and Robot Gripper optimization problems. Moreover, two techniques for evaluating the proposed algorithms are selected in this paper, which are quantitative and qualitative analyses. For the former one, a set of key performance indicators are used to measure solution quality, which are mean, best fitness, and standard deviation. While for the later one, the convergence curves are drawn to estimate the efficiency of algorithms over iterations. This helps for a robust estimation of each algorithm's ability to poise exploration and exploitation in the domain of any problem. These case studies provide more evidence of the algorithms' versatility and applicability in managing challenging, multifaceted, and limited optimization jobs. The outcomes show that BBO has the merit in optimizing these problems.