Spatial clustering for determining rescue shelter of flood disaster in South Bandung using CLARANS Algorithm with Polygon Dissimilarity Function

Abstract

In this research, we provide a solution to the problem of handling the recent flooding in South Bandung, West Java, Indonesia. We offer the solution in determining the locations of the rescue posts. The analysis uses a spatial data clustering algorithm known as CLARANS Algorithm and the spatial similarity is measured using Polygon Dissimilarity Function (PDF). Results showed that clustering of two clusters gives the strongest Silhouette index value of 0.9, and clustering of 4 and 5 clusters have a Silhouette index value of 0.8. Clustering process is done quickly, less than 3 seconds for 2 clusters and less than 4 seconds for 5 clusters.