Using neural networks for initial design decisions in buildings

Abstract

With the development of mathematics and ability to process symbols as opposed to numbers in computer technology, the artificial intelligence applications developed rapidly. Artificial Neural Networks (ANN) and Case Based Reasoning (CBR) are some of examples for the developed tools. Here we are trying to use ANN application in Structural Design as a prediction tool for arriving at column spacing and sizing in multistory buildings, based on historical examples. The trained back propagation networks were used to establish the significance of the required resuh. KEYWORDS Artificial Neural Networks, Preliminary design, Column Spacing, Column Sizing, Significance Testing, Sensitivity analysis.