An Overview of the Potential Applications of Neural Networks in Civil Engineering

by Jamshid Ghabousssi, Univ of Illinois at Urbana-Champaign, Urbana, United States,

Document Type: Proceeding Paper

Part of: Structural Engineering in Natural Hazards Mitigation


Neural networks are computational models inspired by the structure and operation of the brain. They are massively parallel systems, made up of a large number of highly interconnected, simple processing units. Signals travel along the interconnections between the processing units. Neural networks have capabilities ideally suited to application to civil engineering problems. The most important is their capability for learning from examples. Neural networks can learn complex associations in an unstructured environment. The knowledge acquired through learning is encoded in the interconnections of the neural network. Neural networks are adaptive models; they can easily acquire new information and modify the stored knowledge. As computational models neural networks are robust and noise and fault tolerant. These properties are important in Civil Engineering, where the problems are complex and the data is often scarce, incomplete or noisy. Neural networks have potential applications in a wide range of civil engineering problems, from construction management problems involving scheduling and allocation to problems of structural control, damage diagnostics and identification. Neural networks also have important applications in computational mechanics and material models.

Subject Headings: Neural networks | Construction engineering | Computer models | Mathematical models | Structural models | Structural control | Structural analysis | Structural systems

Services: Buy this book/Buy this article


Return to search