A SLOPE FAILURE TYPE CLASSIFICATION ALGORITHM FOR MULTIPLE SIMULTANEOUS SLOPE FAILURE HAZARD ASSESSMENT

Abstract

 Pointing out the necessity of evaluating hazardous areas affected by different types of multiple simultaneous slope failures, this paper proposes a slope failure type classification algorithm based on multigroup discriminant analysis of the Hayashi’s quantification method type II. Three kinds of training data sets of “surface slope failure, deep-seated slope failure and landslide in Nagaoka area” induced by the Niigata Chuetsu earthquake are prepared. Quantitative model constructs the relationship between those training data sets of endogenous variables and causal factors of exogeneous variables (i.e. Soil, Surface geology, Topography, Vegetation index, Slope, Aspect, Drainage, Elevation, and Relief). The scatter diagram based on multigroup quantification method type II indicates that the training samples can be grouped into three classes corresponding to above slope failure types. As a final outcome, a slope failure type classification map (termed FTC map) with overall accuracy of 84% and kappa coefficient of 0.76 can be produced through cluster analysis of nearest neighbor method. The FTC map is superior to previous susceptibility map for specified type of slope failure in evaluating hazardous areas affected by different types of multiple simultaneous slope failures.