Efficient Multi-Objective Constrained Bayesian Optimization of Bridge Girder

2025-09-24

AI Summary

The construction sector accounts for approximately 37% of global greenhouse gas emissions, with cement production alone contributing ~7%. This paper addresses the multi-objective optimization of cost and embodied carbon emissions in bridge girder design—a challenge complicated by stringent structural safety constraints. We propose a constraint-aware Bayesian optimization framework explicitly tailored to decision-makers’ trade-off preferences—not Pareto-frontier exploration. Our method integrates Proper Orthogonal Decomposition (POD) with Kriging-based Partial Least Squares (KPLS) to construct an efficient, high-fidelity reduced-order surrogate model, and employs the constrained Expected Improvement (cEI) acquisition function to handle complex engineering constraints. Applied to a three-span prestressed concrete bridge, the approach achieves a 10–15% reduction in construction cost and ~20% reduction in embodied carbon emissions—while strictly satisfying all structural safety requirements—outperforming conventional designs. The key innovation lies in the first synergistic integration of trade-off-driven optimization and POD-KPLS surrogate modeling for low-carbon, simulation-intensive structural design.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Optimization
• Search and Optimization: Mixed Discrete/Continuous Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

The buildings and construction sector is a significant source of greenhouse gas emissions, with cement production alone contributing 7~% of global emissions and the industry as a whole accounting for approximately 37~%. Reducing emissions by optimizing structural design can achieve significant global benefits. This article introduces an efficient multi-objective constrained Bayesian optimization approach to address this challenge. Rather than attempting to determine the full set of non-dominated solutions with arbitrary trade-offs, the approach searches for a solution matching a specified trade-off. Structural design is typically conducted using computationally expensive finite element simulations, whereas Bayesian optimization offers an efficient approach for optimizing problems that involve such high-cost simulations. The proposed method integrates proper orthogonal decomposition for dimensionality reduction of simulation results with Kriging partial least squares to enhance efficiency. Constrained expected improvement is used as an acquisition function for Bayesian optimization. The approach is demonstrated through a case study of a two-lane, three-span post-tensioned concrete bridge girder, incorporating fifteen design variables and nine constraints. A comparison with conventional design methods demonstrates the potential of this optimization approach to achieve substantial cost reductions, with savings of approximately 10% to 15% in financial costs and about 20% in environmental costs for the case study, while ensuring structural integrity.

Problem

• Optimizing bridge girder design to reduce greenhouse gas emissions
• Developing efficient Bayesian optimization for expensive structural simulations
• Balancing cost reduction with structural integrity constraints

Innovation

• Multi-objective Bayesian optimization with specified trade-offs
• Integrates proper orthogonal decomposition for dimensionality reduction
• Uses constrained expected improvement as acquisition function