Artificial Neural Networks for Civil Engineers: Advanced Features and Applications

This is the second in a two-monograph series on the application of artificial neural networks (ANN) to civil engineering. The first monograph, "Artificial Neural Networks for Civil Engineers: Fundamentals and Applications", published by ASCE in 1997, was aimed at providing potential users with an understanding of the scope of application of neural networks. This second monograph complements the first by illustrating advanced methods and novel developments in the application of ANN to civil engineering. The objective is to illustrate the wide range of alternative ways neural networks can be used in problem solving. With this objective in mind, the chapters explore such areas as: evaluating new construction technologies; using multi-layered ANN architecture to overcome problems with conventional traffic signal control systems; increasing the computational efficiency of an optimization model; predicting carbonation depth in concrete structures; detecting defects in concrete piles; analyzing pavement systems; using neural network hybrids to select the most appropriate bidders for a construction project; and predicting the Energy Performance Index of residential buildings. Many of the ideas and techniques discussed in this book can commute across the disciplinary boundaries and, therefore, should be of interest to all civil engineers.