Efficient Truss Design: A Hybrid Geometric Mean Optimizer for Better Performance

Abstract

The emphasis on sustainable development is increasingly recognized as a global imperative, leading to significant transformations in methods and technologies within the construction industry. Specifically, the optimization of the mass of truss structures is aimed at enhancing sustainable construction efforts. Such optimization is crucial for reducing dependence on natural resources and contributing to a decrease in CO2 emissions from the production and transport of construction materials. This study presents an innovative approach to the truss design challenge using a hybrid geometric mean optimizer (hGMO). The geometric mean optimizer (GMO) faces challenges in effectively exploring the solution space and avoiding local optima. To address these issues, GMO has been integrated with the variable neighborhood search (VNS) technique, thereby enhancing its exploration capabilities. The effectiveness of the hGMO model has been evaluated through four distinct truss design scenarios: 10-bar, 52-bar, 72-bar, and 200-bar structures. The results demonstrate that hGMO outperforms traditional methods, achieving optimal weights of 5060.915 lb, 1902.605 kg, 389.334 lb, and 25453.62 lb, respectively. These findings confirm that hGMO is a crucial tool in advancing sustainable construction practices by focusing on the strategic optimization of truss structure mass.