Proposition of a geotechnical mapping based on artificial neural networks for the Town of Caucaia, Ceará, Brazil for paving purposes

Abstract

This research focuses on the development of a method, based on Artificial Neural Networks (ANN), aimed to infer the geotechnical characteristics of the subgrade of Caucaia Town, located in the metropolitan region of Fortaleza – Ceará, Brazil based on biophysical variables (pedology, geology, geomorphology and phytophysiognomy). The main objective of this research was to systematize method to assist with the recognition of geotechnical characteristics of the subgrade from data obtained in projects and pre-existing studies (secondary data), so as to help with the decision-making process regarding land use for paving purposes. For that, GIS techniques were used to cross spatial information on biophysical variables from several layers, with the geotechnical classification of the subgrade obtained by the Transportation Research Board (TRB) method. We developed an ANN a model to provide estimates of the geotechnical classification based on the biophysical variables studied, creating a neural-geotechnical map presenting the prediction of TRB classification of the subgrades of the Town of Caucaia. In order to check the efficacy of the proposed mapping, 20 samples were collected and classified according to the TRB categories. The results show the potential of the technique of artificial neural networks in predicting the properties of materials occurring in the town of Caucaia.