Multi-stage production cost optimization of semi-rigid steel frames using genetic algorithms

The response of a steel structure is closely related to the behavior of its joints. This means that it is necessary to take explicit account of joint properties in order to ensure a consistent approach to design optimization of steel frames. Semi-rigid design has been introduced into steel construction standards such as Eurocode 3 and AISC. However, in the absence of appropriate guidelines, engineers encounter difficulties when bringing in semi-rigid design to everyday engineering practice. Moreover, connection design significantly affects the production cost of steel frame structures. Thus, a realistic optimization of frame design should take into account the effective costs of different stages of production including manufacturing and erection activities. This paper presents a Genetic Algorithm based method for multi-stage cost optimization of steel structures. In the objective function, the total cost of different production stages is minimized. A new cost model is presented that itemizes costs of all stages of production (material supply, manufacturing, erection and foundation). Design examples are used to validate the proposed methodology. Numerical validation shows that the multi-stage design optimization results in substantial cost benefits between 10% and 25% compared to traditional design of steel frames. Furthermore, the developed methodology is shown to be capable of measuring the possible impact of design choices in the early design stage thus assisting designers to make better design decisions.