Biogeography-Based Optimization of RC structures including static soil-structure interaction

A method to minimize the economic cost of the structural design of spatial reinforced concrete (RC) frame structures is presented. SAP2000 is used as computational engine, taking into account modelling aspects such as static soilstructure interaction (SSSI). The optimization problem is formulated to properly reflect an actual design problem, limiting e.g., the size of reinforcement bars to commercially available sections. The resulting discrete optimization problem is solved by using Biogeography-Based Optimization (BBO), an evolutionary algorithm selected for its convergence properties. Strategies to reduce the computational cost of the optimization procedure are proposed and an extensive tuning of the parameters of the BBO algorithm is performed, using a novel utility metric, evaluated for models of six simple RC frame structures. The parameters to deal with more complex structures are selected based on the use of utility landscapes. The resulting tuned optimization algorithm allows to reduce the direct cost of the construction of a particular structure project with 21% (15% when SSSI is not taken into account), compared to a design based on traditional criteria. The effect of considering SSSI on the cost of the superstructure is also evaluated, showing that this is an aspect that should not be neglected during modeling.