Computational intelligence tools for the prediction of slope performance

The current paper illustrates the application of computational intelligence tools in slope performance prediction both in static and dynamic conditions. We present the results obtained by using the back-propagation algorithm, the theory of Bayesian neural networks and the Kohonen self-organizing maps, one of the most realistic models of the biological brain functions. We estimate slope stability controlling variables by combining computational intelligence tools with generic interaction matrix theory. Our emphasis is given to the prediction and estimation of the following: slope stability, coefficient of critical acceleration, earthquake induced displacements, unsaturated soil classification, classification according to the status of stability and failure mechanism for dry and wet slopes. Finally, we present an integrated methodology for assessing landslide hazard coupling computational intelligence tools and geographical information systems.