[Paper Review] Biogeography-Based Optimization of RC structures including static soil-structure interaction

Iván Negrin, Dirk Roose+2 more|arXiv (Cornell University)|Mar 8, 2021

Innovations in Concrete and Construction Materials20 references4 citations

TL;DR

This paper presents a biogeography-based optimization (BBO) framework for minimizing the cost of reinforced concrete (RC) structures while accounting for static soil-structure interaction (SSSI) using SAP2000 as the structural analysis engine. The method reduces direct construction costs by 21% compared to traditional design, with validated cost savings and sensitivity analysis on SSSI effects.

ABSTRACT

A method to minimize the cost of the structural design of reinforced concrete structures using Biogeography-Based Optimization, an evolutionary algorithm, is presented. SAP2000 is used as computational engine, taking into account modelling aspects such as static soil-structure 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. Strategies to reduce the computational cost of the optimization procedure are proposed and an extensive parameter tuning was performed. The resulting tuned optimization algorithm allows to reduce the direct cost of the construction of a particular structure project with 21% compared to a design based on traditional criteria. We also evaluate the effect on the cost of the superstructure when SSSI is takeninto account.

Motivation & Objective

• To develop a cost-minimization framework for RC structures that incorporates realistic design constraints and SSI effects.
• To reduce computational cost in evolutionary optimization by tuning algorithm parameters and implementing efficient search strategies.
• To evaluate the impact of static soil-structure interaction (SSSI) on superstructure cost and design efficiency.
• To ensure design feasibility by restricting reinforcement bar sizes to commercially available sections.
• To validate the optimization framework against traditional design practices using a real-world structural project.

Proposed method

• Employed biogeography-based optimization (BBO), an evolutionary algorithm, to explore design parameter spaces for RC sections and reinforcement.
• Integrated SAP2000 as the structural analysis engine to perform nonlinear static pushover analysis with SSI effects.
• Formulated the optimization problem to minimize direct construction cost while respecting code-compliant constraints on bar sizes and section dimensions.
• Applied parameter tuning and computational efficiency strategies, including adaptive migration rates and population control, to reduce runtime.
• Incorporated static soil-structure interaction (SSSI) by modeling foundation flexibility and subsoil stiffness in SAP2000.
• Used a hybrid approach combining BBO with local search heuristics to accelerate convergence and improve solution quality.

Experimental results

Research questions

• RQ1 To what extent can biogeography-based optimization reduce the direct cost of RC structures when SSI is included?
• RQ2 How does the inclusion of static soil-structure interaction (SSSI) affect the optimal design and cost of RC superstructures?
• RQ3 What is the impact of computational efficiency strategies on the convergence and runtime of the BBO algorithm in structural optimization?
• RQ4 How do commercially available reinforcement bar sizes influence the optimal design outcomes and cost savings?
• RQ5 Can the proposed BBO framework consistently outperform traditional design methods in terms of cost and structural feasibility?

Key findings

• The proposed BBO-based optimization reduced the direct construction cost of the target RC structure by 21% compared to a traditional design approach.
• Incorporating static soil-structure interaction (SSSI) significantly influenced the optimal design, leading to changes in section sizes and reinforcement layout.
• The tuned BBO algorithm achieved convergence within a feasible computational time, demonstrating the effectiveness of parameter tuning and efficiency strategies.
• The optimization results were feasible in practice, with all reinforcement bar sizes selected from standard, commercially available sections.
• Sensitivity analysis confirmed that SSSI has a measurable impact on cost and structural behavior, justifying its inclusion in the optimization model.
• The integration of SAP2000 with BBO enabled accurate modeling of SSI effects and reliable structural performance evaluation during optimization.