A novel hybrid Harris hawks-simulated annealing algorithm and RBF-based metamodel for design optimization of highway guardrails

In this paper, a novel hybrid optimization algorithm is introduced by hybridizing a Harris hawks optimization algorithm(HHO) and simulated annealing for the purpose of accelerating its global convergence performance and optimizing structural design problems. This paper is the first research study in which the hybrid Harris hawks simulated annealing algorithm (HHOSA) is used for the optimization of design parameters for highway guardrail systems. The HHOSA is evaluated using the well-known benchmark problems such as the three-bar truss problem, cantilever beam problem, and welded beam problem. Finally, a guardrail system that has an H1 containment level as a case study is optimized to investigate the performance of the HHOSA. The guardrail systems are designed with different cross-sections and distances between the posts. TB11 and TB42 crash analyses are performed according to EN 1317 standards. Twenty-five different designs are evaluated considering weight, the guardrail working width, and the acceleration severity index (ASI). As a result of this research, the optimum design of a guardrail is obtained, which has a minimum weight and acceleration severity index value (ASI). The results show that the HHOSA is a highly effective approach for optimizing real-world design problems.