STR-914: RELIABILITY-BASED DESIGN OPTIMIZATION OF PRESTRESSED GIRDER BRIDGES

Prestressed concrete girder bridges are popular structures due to their simplicity of fabrication, speed of construction, ease of inspection, maintenance and replacement. Many design factors play an essential role in deciding the members’ dimensions, geometry, weights, and cost. This reflects the importance of developing optimization tools that provide cost-effective design by determining certain design variables to achieve the best measurable objective function under given constraints. However, in order to maintain appropriate safety levels while performing the optimization process, it becomes necessary to adopt a probabilistic approach that considers the uncertainties associated with the basic design variables. In this paper, a reliability-based optimization model, for the design of prestressed girder bridges, is proposed. The proposed model adopts a simulation-based optimization technique. The simulation engine utilizes Monte Carlo analysis, while the optimization engine performs metaheuristic scatter search with neural network accelerator. A case study is provided to investigate the applicability of the proposed model. The implementation of the proposed method will serve bridge engineers who intend to achieve cost-efficient design solutions with guaranteed safety levels reserved in the optimized bridge configuration