Performance Evaluation of the K-Means Clustering Method in Grouping Indonesian Provinces Based on Potential Disaster Impact

Sitti Masyitah Meliyana; Anugra S. S. Dunggio; Ansari Saleh Ahmar

doi:10.35877/454RI.jinav4696

Authors

Sitti Masyitah Meliyana Department of Statistics, Universitas Negeri Makassar, Makassar, 90223, Indonesia
Anugra S. S. Dunggio Department of Statistics, Universitas Negeri Makassar, Makassar, 90223, Indonesia
Ansari Saleh Ahmar Department of Statistics, Universitas Negeri Makassar, Makassar, 90223, Indonesia

DOI:

https://doi.org/10.35877/454RI.jinav4696

Keywords:

Davies-Bouldin Index (DBI), Disaster Impact, Elbow, K-Means Clustering, Silhouette Coefficient

Abstract

This study aims to cluster the provinces in Indonesia based on their level of potential disaster impact, which consists of hazard area, exposed population, physical losses, economic losses, and environmental damage, using the K-Means clustering algorithm and to evaluate the performance of the resulting model. The optimal number of clusters was determined using the Silhouette Coefficient and the Elbow Method with the Within-Cluster Sum of Squares (WSS) approach. The performance evaluation of the K-Means clustering was conducted using the Davies–Bouldin Index (DBI). Based on the selection of the optimal number of clusters, the Silhouette Coefficient produced the highest value at K=3, with a score of 0.699. Similarly, the Elbow Method showed a significant decrease in the mean WSS at K=3, indicating that three clusters were optimal. The performance evaluation using DBI for K=3 resulted in a score of 0.30. According to the principle of DBI evaluation, the closer the DBI value is to zero (without being negative), the better the clustering quality. Therefore, it can be concluded that the K-Means clustering algorithm successfully produced a very good clustering structure in grouping Indonesian provinces based on their potential disaster impact.

References

Akram, A., Risal, N., Maryani, D., Fadillah, N., Alviadi, A., & Risal, N. (2024). Implementasi K- Means Clustering Untuk Rekomendasi Kelas Unggulan di SMK 1 Teknologi dan Rekayasa Mimika. JESSI: Journal of Embedded System Security and Intelligent System, 5(3), 255–261. https://doi.org/10.59562/jessi.v5i3.5446

Baldah, A., Duarisah, A. V., & Maulana, R. A. (2023). Clustering Daerah Rawan Bencana Alam Di Indonesia Berdasarkan Provinsi Dengan Metode K-Means. Jurnal Ilmiah Informatika Global, 14(2), 31–36. https://doi.org/10.36982/jiig.v14i2.3186

BNPB. (2025). Rangkuman Bencana Indonesia tahun 2025. Data Informasi Bencana Indonesia. https://dibi.bnpb.go.id/superset/dashboard/1/?standalone=0&expand_filters=0

Dhewayani, F. N., Amelia, D. N., Sari, B. N., & Jajuli, M. (2022). Implementasi K-Means Clustering untuk Pengelompokkan Daerah Rawan Bencana Kebakaran Menggunakan Model CRISP-DM. Jurnal Teknologi dan Informasi (JATI), 12(1), 64–77. https://doi.org/10.34010/jati.v12i1

Dinh, D. T., Fujinami, T., & Huynh, V. N. (2019). Estimating the Optimal Number of Clusters in Categorical Data Clustering by Silhouette Coefficient. Communications in Computer and Information Science, 1103 CCIS, 1–17. https://doi.org/10.1007/978-981-15-1209-4_1

Faturrahman, Hariani, S., & Kusumawati, R. (2023). Evaluasi Clustering K-Means dan K-Medoid pada Persebaran Covid-19 di Indonesia dengan Metode Davies-Bouldin Index (DBI). Jurnal MNEMONIC, 6(2), 117–128. https://doi.org/10.36040/mnemonic.v6i2.6642

Frege, I. A., Bliecke, V., Bradshaw, S., Dijkzeul, D., Funke, C., Kienzl, P., Küsters, K., Masuch, L., Nasreen, M., Radtke, K., Schneider, S., Thielbörger, P., Weller, D., Wiegers, O., Yaman, D. K., & Zennig, K. (2023). World Risk Report 2023. Bündnis Entwicklung Hilft.

Hafid, H., Meliyana, S. M. ., Muthahharah, I. ., & Mar’ah, Z. . (2025). Implementation K-Medoids Algorithm for Clustering Indonesian Provinces by Poverty and Economic Indicators. Quantitative Economics and Management Studies, 6(2), 219–225. https://doi.org/10.35877/454RI.qems3940

Ikhram, A., & Sani Mutia. (2025). Penerapan Algoritma K-Medoids untuk Pengelompokan Provinsi di Indonesia Berdasarkan Indikator Kemiskinan: Akses Perumahan dan Kesejahteraan. Emerging Statistics and Data Science Journal, 3(2), 583–596. https://doi.org/10.20885/esds.vol3.iss.2.art13

Karo, I. M. K., Dewi, S., Mardiana, Ramadhani, F., & Harliana, P. (2023). K-means and K-medoids Algorithm Comparison for Clustering Forest Fire Location in Indonesia. Jurnal ECTIPE: Electronic, Control, Telecommunication, Information, and Power Engineering, 10(1), 86–94. https://doi.org/10.33019/jurnalecotipe.v10i1.3896

Luchia, N. T., Handayani, H., Hamdi, F. S., Erlangga, D., & Octavia, S. F. (2022). Comparison of K-Means and K-Medoids on Poor Data Clustering in Indonesia. MALCOM: Indonesian Journal of Machine Learning and Computer Science, 2(2), 35–41. https://doi.org/10.57152/malcom.v2i2.422

Meliyana, S. M., Dunggio, A. S. A., Muhammad, S., & Rahman, A. (2025). Comparative Performance Analysis of Two Clustering Methods for Grouping Indonesian Provinces Based on Forest Area Size. Daengku: Journal of Humanities and Social Sciences Innovation, 5(4). https://doi.org/10.35877/454RI.daengku4439

Monalia, I., & Noorratri, E. D. (2024). Gambaran tingkat pengetahuan masyarakat tentang kesiapsiagaan dalam menghadapi bencana gunung meletus di Desa Jrakahselo Boyolali. IJOH: Indonesian Journal of Public Health, 2(4), 790–807. https://jurnal.academiacenter.org/index.php/IJOH

Muthahharah, I., Meliyana, S. M., Ahmar, A. S., & Rahman, A. (2024). K-Means Cluster for Grouping Regencies/Cities in South Sulawesi Province Based Human Development Index on the 2023. ARRUS Journal of Social Sciences and Humanities, 4(3), 357–364. https://doi.org/10.35877/soshum2587

Paembonan, S., & Abduh, H. (2021). Penerapan Metode Silhouette Coeficient Untuk Evaluasi Clutering Obat. PENA TEKNIK: Jurnal Ilmiah Ilmu-Ilmu Teknik, 6(2), 48–54. https://ojs.unanda.ac.id/index.php/jiit/article/view/659

Pratama, A. R., Maulana, B., Rianda, R. D., & Hasyim, S. El. (2023). Comparison of K-Means and K-Medoids Algorithms for Grouping Video Game Sales Data in North America. IJIRSE: Indonesian Journal of Informatic Research and Software Engineering, 3(2), 111–118. https://journal.irpi.or.id/index.php/ijirse

Shalih, O., Adi, A. W., Wiguna, S., Shabrina, F. Z., Putra, A. S., Karimah, R., Eveline, F., Alfian, A., Syauqi, Septian, R. T., Widiastomo, Y., Bagaskoro, Y., Dewi, A. N., Rahmawati, I., Seniarwan, Hafizh, A., Suryaningrum, H. A., Purnamasiwi, D. I., Puspasari, T. J., & Iftidah, A. (2023). RBI: Risiko Bencana Indonesia “Memahami Risiko Sistematik di Indonesia” (R. Yunus (ed.); Pertama). Pusat Data Informasi Komunikasi Bencana Badan Nasional Penanggulangan Bencana.

Turnip, Y. E., & Fitriana, R. (2023). Clustering Kabupaten Berdasarkan Luas Hutan Menggunakan Metode K-Means di Provinsi Jawa Tengah. Jurnal Teknologi Industri Pertanian, 33(1), 1–9. https://doi.org/https://doi.org/10.24961/j.tek.ind.pert.2023.33.1.1

Umagapi, I. T., Umaternate, B., Hazriani, & Yuyun. (2023). Uji Kinerja K-Means Clustering Menggunakan Davies-Bouldin Index Pada Pengelompokan Data Prestasi Siswa. Prosiding Seminar Nasional Sistem Informasi dan Teknologi (SISFOTEK), 303–308.

United Nations Office for Disaster Risk Reduction. (2022). Global Assesmet Report on Disaster Risk Reduction. United Nation. www.undrr.org/GAR2022

Utami, P. Y., Hudjimartsu, S. A., Viona, T. A., & Sharfina, H. (2023). Optimasi Parameter Algoritma DBSCAN untuk Mendeteksi Titik Panas Kebakaran Hutan dan Lahan. Jurnal Edukasi dan Penelitian Informatika (JEPIN), 9(3), 355. https://doi.org/10.26418/jp.v9i3.61714

Wicaksono, D. A., & Susetyo, Y. A. (2023). Clustering Zonasi Daerah Rawan Bencana Alam di Provinsi Sumatera Utara Menggunakan Algoritma K-Means dan Library Geopandas. Jurnal Indonesia?: Manajemen Informatika dan Komunikasi, 4(3), 989–998. https://doi.org/10.35870/jimik.v4i2.225