Please use this identifier to cite or link to this item:
Type: Conference paper
Title: Investigation into the robustness of artificial neural networks for a case study in civil engineering
Author: Shahin, M.
Maier, H.
Jaksa, M.
Citation: MODSIM 2005 International Congress on Modelling and Simulation: Modelling and Simulation Society of Australia and New Zealand, December 2005 / Andre Zerger and Robert M. Argent (eds.): pp.79-83
Publisher: mssanz
Publisher Place:
Issue Date: 2005
ISBN: 0975840029
Conference Name: International Congress on Modelling and Simulation (16th : 2005 : Melbourne, Victoria)
Statement of
M. A. Shahin, H. R. Maier and M. B. Jaksa
Abstract: Artificial neural networks (ANNs) have been used as a prediction tool in many areas of engineering. In order to test the robustness and generalisation ability of ANN models, the approach that is gener-ally adopted is to test the performance of trained ANNs on an independent validation set. If such performance is adequate, the model is deemed to be robust and able to generalise. However, this is not necessarily the case. In this paper, the robust-ness of ANN models is investigated for a case study of predicting the settlement of shallow foun-dations on granular soils. A procedure that tests the robustness of the predictive ability of ANN models is introduced. The results indicate that good performance of ANN models on the data used for model calibration and validation does not guarantee that the models will perform in a robust fashion over a range of data similar to that used in the model calibration phase. The results also indi-cate that validating ANN models using the proce-dure provided in this study is essential in order to investigate their robustness.
Description: © 2005 Modelling & Simulation Society of Australia & New Zealand
RMID: 0020052449
Description (link):
Published version:
Appears in Collections:Civil and Environmental Engineering publications
Environment Institute publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.