Premium
Site Characterization by Neuronets: An Application to the Landfill Siting Problem
Author(s) -
Basheer Imad A.,
Reddi Lakshmi N.,
Najjar Yacoub M.
Publication year - 1996
Publication title -
groundwater
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.84
H-Index - 94
eISSN - 1745-6584
pISSN - 0017-467X
DOI - 10.1111/j.1745-6584.1996.tb02048.x
Subject(s) - subsoil , permeability (electromagnetism) , probabilistic logic , artificial neural network , sensitivity (control systems) , computer science , spatial variability , environmental science , data mining , soil science , artificial intelligence , engineering , statistics , mathematics , membrane , electronic engineering , biology , soil water , genetics
For a large number of geotechical/ground‐water applications, determining the spatial distribution of the subsoil properties constitutes a major challenge in both design and construction phases. Models that are frequently used for characterization of site properties emphasize the use of probabilistic models that incorporate a number of model parameters to be evaluated. In this paper, neural networks (or neuronets) are used to map the variation of permeability for purposes of identifying boundaries of landfill to be constructed on a real site. The neural network, as a simple technique, was found to be able to logically predict the variation. The sensitivity of the produced permeability maps to both the quality and number of observations was also studied to investigate the accuracy of the proposed mapping methodology. The use of neuronets as a mapping tool can help identify the regions within a site where additional subsoil exploration is warranted.