Reliabilty-based optimization of river bridges using artificial intelligence techniques

Thesis Type: Doctorate

Institution Of The Thesis: Middle East Technical University, Faculty of Engineering, Department of Civil Engineering, Turkey

Approval Date: 2011

Thesis Language: English

Student: Kamil Hakan Turan



Proper bridge design is based on consideration of structural, hydraulic, and geotechnical conformities at an optimum level. The objective of this study is to develop an optimization-based methodology to select appropriate dimensions for components of a river bridge such that the aforementioned design aspects can be satisfied jointly. The structural and geotechnical design parts uses a statisticallybased technique, artificial neural network (ANN) models. Therefore, relevant data of many bridge projects were collected and analyzed from different aspects to put them into a matrix form. ANN architectures are used in the objective function of the optimization problem, which is modeled using Genetic Algorithms with penalty functions as constraint handling method. Bridge scouring reliability comprises one of the constraints, which is performed using Monte-Carlo Simulation technique. All these mechanisms are assembled in a software framework, named as AIROB. Finally, an application built on AIROB is presented to assess the outputs of the software by focusing on the evaluations of hydraulic – structure interactions.