Optimal design of steel structures considering welding cost and constructability of beam-column connections

Abstract

Structural design is a process of determining an optimal solution by repeating hypotheses and verification processes with numerous design variables. To acquire a more appropriate solution, a mathematical model for structural design is proposed to find the optimal solution. While many studies have proposed the minimum weight design for steel frames, a mathematical model should be considered in the manufacturing and installation process in order to solve practical issues. In this study, a multi-objective model was proposed to consider not only the cost of materials, but also processing and welding costs. A penalty function was also used to reflect the constructability and on-site applicability of the structure. A genetic algorithm was used to determine the optimal solution, and the results of the optimal design analysis were compared and analyzed based on a three-story numerical example. Analysis of the numerical example resulted in offering a smaller number of types of materials whose size and arrangement allowed for easier construction, compared to that of the optimal design that only considered weight. Furthermore, the welding cost allowed for the optimum design with improved on-site constructability was also considered.