Neural networks for assessing the failure load of a construction

In this contribution, neural networks are applied to the field of structural engineering. The prediction of the failure load of a construction is a tedious task if one considers the geometrical imperfections of the construction. Experiments on models in a laboratory lead to test results. The guaranteed strength of a construction can be found by searching for a lower bound of the test results. In this contribution, this lower bound is obtained with neural networks. The method is based on the knowledge of the tolerances on the geometrical features of the construction.