Penggunaan Metode Elbow untuk Pemilihan Jumlah Klaster dalam Identifikasi Bahan Material Shelter Modular

Abstract

Modular shelters have become a popular solution for temporary infrastructure construction, especially in disaster-affected areas. One of the main challenges is selecting the appropriate materials, which can be addressed through cluster analysis to group materials based on similar characteristics. The Elbow Method is used to determine the optimal number of clusters in this analysis, with the "elbow" point on the graph indicating that four clusters are ideal. The K-Means algorithm is then applied to group material data based on the centroid of each cluster. The application of the Elbow Method has proven effective in determining the optimal number of clusters for material identification in modular shelter construction. By analyzing the relationship between the number of clusters and inertia, the Elbow Method successfully indicates that four clusters are the most appropriate. The Elbow graph shows a significant "elbow" after the third and fourth clusters, where the decrease in inertia slows down, indicating that adding more than four clusters does not significantly improve data grouping. Quantitatively, clustering with four clusters provides a balance between data variation and ease of interpretation. Each cluster exhibits distinct characteristics based on the average values of structural and architectural attributes, with variability measured through standard deviation