Prediction of Slope Failure Susceptibility of an Iron Ore Mine Using PCA and K-Means Clustering

Slope failure in mines is one of the most frequent accident events that cause damage to both property and life every year. This study uses remote sensing and GIS technique to predict and generate a slope susceptibility map of an iron ore mine area in India. First, the required remote sensing data and satellite images are collected. The final results of this study will help to find the vulnerable zones for slope failure, and eventually, this will help in effective land use planning to save the environment and lives. The basic data sources required to generate the final susceptibility map are various thematic maps such as land use, elevation, curvature. They are generated with the help of Digital Elevation Model (DEM) data collected from the satellite images of the study area. First, the weights for each thematic map are calculated using Analytical Hierarchy Process (AHP) method. Then, the PCA algorithm is implemented using PYTHON code to associate all the attributes to each pixel of the final slope susceptible map. Finally, each pixel of the final map is again classified into high, moderate, low, and very low slope susceptibility indexes using the K-means Clustering algorithm.