| This research investigates the utility of Artificial Neural Network (ANN) technology in producing Engineering Geological Maps. Specifically, a preliminary map describing the stability for sanitary landfill use was created. To accomplish this, three technologies were combined: ANNs, Geographic Information Systems (GISs), and engineering geological mapping techniques.;ANNs are computer programs designed to model the human brain and its ability to learn tasks. GIS are computer systems that store, analyze, and display mapped features and related information. Engineering geological mapping classifies the surficial materials and bedrock geology according to their suitability for engineered construction or modification. Engineering geology relies on basic geologic, hydrologic, and topographic data. This data is created, stored and displayed in the GIS, which is also used to format the data into the single-column vector needed by the ANN. Within the ANN, a sequence of training and testing was undertaken to create a tool that can be used to create preliminary maps of engineering geology themes. This is additional information for the engineering geologist to use in the investigation of a specific area. It is a tool that can reduce the level of effort, and thus the ultimate cost, of the engineering geological map.;This research shows that ANN technology is appropriate for creating preliminary engineering geological maps. The trained ANN correctly estimated the suitability for sanitary landfill use in 93% of the area being investigated. |
