Physically Consistent Peak Ground Acceleration Prediction in Bali Using Ensemble Machine Learning
DOI:
https://doi.org/10.70561/akefdb62Kata Kunci:
Attenuation relationship, Ensemble learning, Hyperparameter tuning, Seismic hazard assessment, Strong-motion recordsAbstrak
Bali faced significant seismic risks from the Java Trench and back-arc thrust faults, making accurate Peak Ground Acceleration (PGA) prediction crucial. While traditional empirical models struggled with complex, non-linear site effects, machine learning offered a robust alternative. This study aimed to develop and compare Random Forest and XGBoost algorithms, utilizing moment magnitude (Mw), hypocentral distance (Rhypo), and local site conditions (Vs30) as primary inputs, to predict PGA in the Bali region. The research utilized 1,000 strong-motion records from 287 earthquakes (Mw > 4.0) recorded between 2018 and 2025. Following rigorous quality control, we applied logarithmic transformations and hyperparameter optimization via Randomized Search with 3-fold cross-validation. Expanding upon our preliminary model evaluations, this comprehensive analysis demonstrated that the Random Forest architecture outperformed XGBoost by maintaining superior predictive stability, yielding a Coefficient of Determination of 0.741, a Mean Absolute Error of 0.340, and a highly constrained residual standard deviation of 0.434. Furthermore, feature importance analysis revealed that Random Forest was physically consistent with seismological principles. It correctly identified hypocentral distance as the dominant ground-motion attenuation predictor, whereas XGBoost exhibited an overfitted bias toward local site conditions. Ultimately, the Random Forest algorithm provided a highly reliable model for PGA prediction, effectively supporting the development of Probabilistic Seismic Hazard Analysis maps and rapid real-time ShakeMaps in Indonesia.
Referensi
Allen, T. I., & Wald, D. J. (2009). On the Use of High-Resolution Topographic Data as a Proxy for Seismic Site Conditions (VS30). Bulletin of the Seismological Society of America, 99(2A), 935–943. https://doi.org/10.1785/0120080255
Apriliansyah, F., Nuha, M. U., Atmojo, A. T., Setiawan, K. T., & Syafnur, A. (2026). Comparative Accuracy of Satellite-Derived Bathymetry Using Random Forest, Multiple Linear Regression, and Van Hengel and Spitzer Algorithm. Jurnal Geocelebes, 48–68. https://doi.org/10.70561/geocelebes.v10i1.43582
Badan Standardisasi Nasional. (2019). SNI 1726:2019: Tata Cara Perencanaan Ketahanan Gempa untuk Struktur Bangunan Gedung dan Nongedung. Badan Standardisasi Nasional. https://sitaba.pu.go.id/publikasi/6977fb4f-d02f-4fd0-a857-65219c38f24e
Baumann, C., & Dalguer, L. A. (2014). Evaluating the Compatibility of Dynamic Rupture-Based Synthetic Ground Motion with Empirical Ground-Motion Prediction Equation. Bulletin of the Seismological Society of America, 104(2), 634–652. https://doi.org/10.1785/0120130077
Bojórquez, E., Payán-Serrano, O., Bojórquez, J., Rodríguez-Castellanos, A., Ruiz, S. E., Reyes-Salazar, A., & Velarde, F. (2025). Support Vector Machines to Propose a Ground Motion Prediction Equation for the Particular Case of the Bojorquez Intensity Measure INp. AI, 6(10), 254. https://doi.org/10.3390/ai6100254
Chen, T., & Guestrin, C. (2016). XGBoost : A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
Feng, Y., Li, Y., Wang, F., & Yang, L. (2025). Hybrid CNN-GRU-Attention-XGBoost Model for Rapid and Accurate Prediction of Peak Acceleration and Trigger Time in Dam Safety Monitoring. In 2025 5th International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA), 1245–1251. https://doi.org/10.1109/CAIBDA65784.2025.11183451
Heath, D. C., Wald, D. J., Worden, C. B., Thompson, E. M., & Smoczyk, G. M. (2020). A global hybrid VS30 map with a topographic slope–based default and regional map insets. Earthquake Spectra, 36(3), 1570–1584. https://doi.org/10.1177/8755293020911137
Hielmy, R. I. (2024). Perbandingan Metode Davenport, Milne, dan McGuirre dalam Penentuan Nilai Percepatan Tanah Maksimum di Pulau Bali. Buletin Meteorologi, Klimatologi Dan Geofisika, 4(5), 9–16. https://balai2bmkg.id/index.php/buletin_mkg/article/view/133
Hielmy, R. I., Pranata, B., Wijayanto, & Daryono. (2025a). Performance Evaluation of Automated and Manual Seismic Phase Picking for Rapid Earthquake Parameter. Jurnal Meteorologi Dan Geofisika, 26(2), 89–107. https://doi.org/10.31172/jmg.v26i2.1189
Hielmy, R. I., Aritonang, B. H., Sijabat, N. A., An-nizami, M. S., Setyowati, P. R., Wijaya, A., Adzan, M. A., Aji, T. W., Kristianto, A., Rachmawardani, A., & Sianipar, D. S. J. (2025b). Sosialisasi Mitigasi Bencana Gempa Bumi , Tsunami, dan Hidrometeorologi di SMAN 6. To Maega | Jurnal Pengabdian Masyarakat, 8(3), 607–616. https://doi.org/10.35914/hk0rn065
Hielmy, R. I., Kean, F. A. C., Gosi, M. F. Al, Syifa, F., Daraman, M. A., Lala, R. Y., Sianipar, D. S. J., Justicea, A. A., Suardi, I., Muzli, Karyono, Harahap, D., Adi, S. P., Wandono, & Peera, F. (2026). Evaluasi Efektivitas Simulasi Mitigasi Bencana Gempa Bumi di Sekolah An-Nisaa’ Izada, Tangerang Selatan. To Maega | Jurnal Pengabdian Masyarakat, 9(1), 1–12. https://doi.org/10.35914/4gqdgn87
Hielmy, R. I., & Muzli, M. (2026). Comparative Analysis of Random Forest and XGBoost for Peak Ground Acceleration Prediction in Bali, Indonesia. Proceedings of the 8th Asia Pacific Meeting on Near Surface Geoscience & Engineering, 2026(1), 1–4. European Association of Geoscientists & Engineers. https://www.earthdoc.org/content/papers/10.3997/2214-4609.202671070
Hielmy, R. I., & Rajagukguk, M. L. (2025). Analisis Probabilistik Bahaya Seismik di Denpasar dan Sekitarnya Berdasarkan Pendekatan PSHA. Journal Online of Physics, 11(1), 21–28. https://doi.org/10.22437/jop.v11i1.47825
Hielmy, R. I., Santoso, E., Muzli, M., Pramono, S., & Prayoedhie, S. (2026). Machine learning-based ground motion prediction models for Indonesia: A performance evaluation of XGBoost and Random Forest against conventional GMPE. Jurnal Meteorologi dan Geofisika, 27(1), 51–67. https://doi.org/10.31172/jmg.v27i1.1233
Irsyam, M., Sengara, W., Aldiamar, F., Widiyantoro, S., Triyoso, W., Hilman, D., Kertapati, E., Meilano, I., Suhardjono, Asrurifak, M., & Ridwan, M. (2010). Ringkasan Hasil Studi Tim Revisi Peta Gempa Indonesia 2010. Kementerian Pekerjaan Umum Republik Indonesia.
Jozinović, D., Lomax, A., Štajduhar, I., & Michelini, A. (2020). Rapid Prediction of Earthquake Ground Shaking Intensity Using Raw Waveform Data and a Convolutional Neural Network. Geophysical Journal International, 222(2), 1379–1389. https://doi.org/10.1093/gji/ggaa233
Khan, M. Y., Qayoom, A., Nizami, M. S., Siddiqui, M. S., Wasi, S., & Raazi, S. M. K.-u.-R. (2021). Automated Prediction of Good Dictionary EXamples (GDEX): A Comprehensive Experiment with Distant Supervision, Machine Learning, and Word Embedding-Based Deep Learning Techniques. Complexity, 2021(1), 2553199. https://doi.org/10.1155/2021/2553199
Khosravikia, F., & Clayton, P. (2021). Machine learning in ground motion prediction. Computers and Geosciences, 148, 104700. https://doi.org/10.1016/j.cageo.2021.104700
Kong, Q., Trugman, D. T., Ross, Z. E., Bianco, J., Meade, B. J., & Gerstoft, P. (2019). Machine Learning in Seismology : Turning Data into Insights. Seismological Research Letters, 90(1), 3–14. https://doi.org/10.1785/0220180259
Kramer, S. L., & Stewart, J. P. (2024). Geotechnical Earthquake Engineering. CRC Press. https://doi.org/10.1201/9781003512011
Kubo, H., Kunugi, T., Suzuki, W., Suzuki, S., & Aoi, S. (2020). Hybrid predictor for ground-motion intensity with machine learning and conventional ground motion prediction equation. Scientific Reports, 10(1), 1–12. https://doi.org/10.1038/s41598-020-68630-x
Leema, A., Balakrishnan, P., Kiruba, G. G., Rajarajan, G., Goel, S., & Aggarwal, P. (2025). SeismoQuakeGNN: A Hybrid Framework for Spatio-Temporal Earthquake Prediction with Transformer-Enhanced Models. Frontiers in Artificial Intelligence, 8, 1690476. https://doi.org/10.3389/frai.2025.1690476
Lisa, A., Refrizon, R., & Samdara, R. (2026). The Earthquake Prediction in the Southern Part of Sumatra Using Deep Learning (Long Short-Term Memory) Models. Jurnal Geocelebes, 1–11. https://doi.org/10.70561/geocelebes.v10i1.42958
Mohammadi, A., Karimzadeh, S., Banimahd, S. A., Ozsarac, V., & Lourenço, P. B. (2023). The potential of region-specific machine-learning-based ground motion models: Application to Turkey. Soil Dynamics and Earthquake Engineering, 172, 108008. https://doi.org/10.1016/j.soildyn.2023.108008
Monterrubio-Velasco, M., Callaghan, S., Modesto, D., Carrasco, J. C., Badia, R. M., Pallares, P., Vázquez-Novoa, F., Quintana-Ortí, E. S., Pienowska, M., & de la Puente, J. (2024). A machine learning estimator trained on synthetic data for real-time earthquake ground-shaking predictions in Southern California. Communications Earth & Environment, 5(1), 258. https://doi.org/10.1038/s43247-024-01436-1
Moon, Y., Lee, Y., Hwang, Y., & Jeong, J. (2024). Long Short-Term Memory Autoencoder and Extreme Gradient Boosting-Based Factory Energy Management Framework for Power Consumption Forecasting. Energies, 17(15), 3666. https://doi.org/10.3390/en17153666
Mousavi, S. M., & Beroza, G. C. (2022). Deep-learning Seismology. Science, 377(6607), eabm4470. https://doi.org/10.1126/science.abm4470
Muzli, M., Lythgoe, K. H., Hielmy, R. I., Triyono, R., & Wei, S. (2026). Aftershock Characteristics of the 2018 Palu Earthquake: Implications for Supershear Rupture Segment. Indonesian Physical Review, 9(2), 374–385. https://doi.org/10.29303/ipr.v9i2.666
Nemutlu, Ö. F., Özçelik, S. T. A., & Freeshah, M. (2025). A Machine Learning Framework for Regional Damage Assessment Using Multi-Station Seismic Parameters: Insights from the 2023 Kahramanmaraş Earthquakes. Buildings, 15(18), 3326. https://doi.org/10.3390/buildings15183326
Ningrum, R. W., Amelia, R. N., Taib, S., Achmad, R., & Aswan, M. (2022). Pemetaan Potensi Kerentanan Seismik sebagai Mitigasi Bencana Gempa di Morotai Selatan. Jurnal Geocelebes, 37–46. https://doi.org/10.20956/geocelebes.v6i1.19150
Parker, G. A., Stewart, J. P., Boore, D. M., Atkinson, G. M., & Hassani, B. (2022). NGA-subduction global ground motion models with regional adjustment factors. Earthquake Spectra, 38(1), 456-493. https://doi.org/10.1177/87552930211034889
Probst, P., Wright, M. N., & Boulesteix, A.-L. (2019). Hyperparameters and tuning strategies for random forest. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 9(3), e1301. https://doi.org/10.1002/widm.1301
PuSGeN. (2024). Buku Peta Sumber dan Bahaya Gempa Indonesia Tahun 2024. Kementerian Pekerjaan Umum Republik Indonesia. https://ciptakarya.pu.go.id/admin/assets/upload/galeri/gempa/2026/02/10/887359_0.%20Final_BUKU_PETA_SUMBER_DAN_BAHAYA_GEMPA_2024_Publish.pdf
Rachmadan, A., Koeshidayatullah, A., & Kaka, S. L. I. (2025). Developing ground motion prediction models for West Java: A machine learning approach to support Indonesia’s earthquake early warning system. Applied Computing and Geosciences, 25, 100212. https://doi.org/10.1016/j.acags.2024.100212
Ramadan, F., Nissen-Meyer, T., Koelemeijer, P., & Fry, B. (2026). PGVnet: a Machine Learning Framework for the Generation of Rapid, Physics-Consistent PGV Maps. Geophysical Journal International, 244(3), 1–14. https://doi.org/10.1093/gji/ggag008
Shi, Y., Zhang, Z., Xue, C., & Feng, Y. (2024). Machine Learning Prediction of Co-Seismic Landslide with Distance and Azimuth Instead of Peak Ground Acceleration. Sustainability, 16(19), 8332. https://doi.org/10.3390/su16198332
Shokrgozar-Yatimdar, E., & Chen, P. (2025). Explainable Deep Learning for Real-Time Prediction of Uniform Hazard Spectral Acceleration for On-Site Earthquake Early Warning. Geophysical Journal International, 243(2), 1–18. https://doi.org/10.1093/gji/ggaf345
Somala, S. N., Chanda, S., Alhamaydeh, M., & Mangalathu, S. (2024). Explainable XGBoost–SHAP Machine-Learning Model for Prediction of Ground Motion Duration in New Zealand. Natural Hazards Review, 25(2), 4024005. https://doi.org/10.1061/NHREFO.NHENG-1837
Sun, R., Qi, W., Zheng, T., & Qi, J. (2023). Explainable Machine-Learning Predictions for Peak Ground Acceleration. Applied Sciences, 13(7), 4530. https://doi.org/10.3390/app13074530
Taya, R., Mittal, H., Saini, A., & Kumar, R. (2026). Comparative Evaluation of Machine-Learning Models and Recalibrated GMPEs for Ground-Motion Prediction. EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17735, https://doi.org/10.5194/egusphere-egu26-17735
Thompson, E. M., Wald, D. J., & Worden, C. B. (2014). A VS30 Map for California with Geologic and Topographic Constraints. Bulletin of the Seismological Society of America, 104(5), 2313–2321. https://doi.org/10.1785/0120130312
Wald, D. J., & Allen, T. I. (2007). Topographic slope as a proxy for seismic site conditions and amplification. Bulletin of the Seismological Society of America, 97(5), 1379–1395. https://doi.org/10.1785/0120060267
Withers, K. B., Moschetti, M. P., & Thompson, E. M. (2020). A machine learning approach to developing ground motion models from simulated ground motions. Geophysical Research Letters, 47(6), e2019GL086690. https://doi.org/10.1029/2019GL086690
Zhang, H., Liu, Q., Jin, X., Cai, H., Chen, Z., & Wang, S. (2026). Hybrid Approach of Real-Time Ground Motion Prediction for Earthquake Early Warning. Seismica, 5(1), 1–18. https://doi.org/10.26443/seismica.v5i1.2194
Zhu, C., Cotton, F., Kawase, H., & Nakano, K. (2023). How Well Can We Predict Earthquake Site Response So Far? Machine Learning vs Physics-Based Modeling. Earthquake Spectra, 39(1), 478–504. https://doi.org/10.1177/87552930221116399
Ziar, A., & Basari, E. (2025). Accurate Prediction of Peak Ground Acceleration Using Random Forests: A Data-Driven Approach with PEER Updated NGA-WEST2 Ground Motion Records. Archives of Advanced Engineering Science, 0(0), 1–9. https://doi.org/10.47852/bonviewAAES52026930
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2026 JURNAL GEOCELEBES
Artikel ini berlisensiCreative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
