A practical tool for economical design of I-shaped steel beams: development and application for parametric study
Abstract
Steel beams are efficient structural elements widely used in industrial buildings due to their remarkable strength and ductility characteristics. Despite numerous studies exploring the optimization of steel beam for cost-effective designs, the utilization of optimization within the construction industry remains rare. This scarcity can be attributed to the complexities associated with applying optimization algorithms and the limited understanding of the structural behavior of optimized designs. In light of these challenges, this study aims to leverage the immense computational power of artificial intelligence (AI), specifically the Evolutionary Algorithm (EA), to develop an innovative AI-driven spreadsheet-based tool for cost optimization of I-shaped steel beams. Additionally, a parametric study investigates the influence of various design variables on the optimized cost. The EA within the Solver tool of MS Excel is used to perform the optimization. Design variables are subject to strength and serviceability-related constraints in accordance with AISC 360–22. The effectiveness of the developed optimization approach is demonstrated by optimizing four steel beam design examples from the literature. It is found that up to 51% of the beam cost can be optimized by keeping the beam depth and steel grade as variables. The parametric study reveals that the optimal range for the beam depth varies depending on the steel grade used. Further, the trends obtained for beam cost with respect to other variables such as flange and web slenderness ratio, and beam depth provide valuable insights for structural engineers undertaking steel beam optimization in the future.