Jump to Main Content
Assessing the factor of safety using an artificial neural network: case studies on landslides in Giresun, Turkey
- Gelisli, Kenan, Kaya, Türkan, Babacan, Ali Erden
- Environmental earth sciences 2015 v.73 no.12 pp. 8639-8646
- case studies, cohesion, data collection, friction, landslides, neural networks, rapid methods, safety factor, Turkey (country)
- In this study, the calculability of slope stability using the artificial neural networks (ANN) method was examined. Initially, 100 synthetic slope models were created to be used in calculations and the factors of safety of these slopes were calculated by a conventional stability calculation method using slope parameters. Then, factors of safety were calculated by through ANN method. 80 of the datasets from the generated data were used for training while 20 were used for testing in these calculations with the ANN method. In both conventional calculation of stability and the ANN method, input parameters included slope height, height of the water level, slope angle, unit weight, cohesion and angle of internal friction, while the output parameter was factor of safety (SF). A good level of consistency was obtained between the SFs calculated through the conventional method and the ANN method. Furthermore, SFs were calculated separately via the ANN method by assigning range values to unit weight, cohesion and angle of internal friction from amongst the parameters that affect SF for Giresun landslides (Eastern Turkey). The data obtained in this scope revealed that cohesion was the parameter with the highest level of effect on SF. Consequently, it was established that the factors of safety of slopes could be calculated by means of the ANN method in a rapid and convenient manner, the effects of slope parameters on the factors of safety in landslide were examined and the factors of safety for Giresun landslides were calculated through the ANN method.