Melola Discovery

← Melola Discovery

Akses terbuka Jurnal Tanpa peringkat SINTA

Physically Consistent Peak Ground Acceleration Prediction in Bali Using Ensemble Machine Learning

JURNAL GEOCELEBES · Vol. 10 No. 2 (2026) · hlm. 306–321

DOI
10.70561/akefdb62
ISSN
2579-5546, 2579-5821
Terbit
2026-10-02
Penerbit
Departemen Geofisika, FMIPA - Universitas Hasanuddin, Makassar
Bahasa
en
Lisensi
creativecommons.org/licenses/by-nc-sa/4.0

Abstrak

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.

Kata kunci

Jaringan kolaborasi para penulis

11 penulis · 27 tautan kepenulisan bersama · 13 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Meteorological, Climatological, And Geophysical Agency
  • Bandung Institute of Technology
  • Indonesian National Institute of Aeronautics and Space
  • International Research Center for Space and Planetary Environmental Science, Kyushu University
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Martha Lisauli RajagukgukMeteorological, Climatological, And Geophysical Agency2
Robby WallanshaMeteorological, Climatological, And Geophysical Agency1
Fajri Syukur RahmatullahMeteorological, Climatological, And Geophysical Agency1
Sigit Eko PramonoMeteorological, Climatological, And Geophysical Agency1
Suaidi AhadiMeteorological, Climatological, And Geophysical Agency1
Nanang T. PuspitoBandung Institute of Technology1
Gunawan IbrahimBandung Institute of Technology1
Sarmoko SarosoIndonesian National Institute of Aeronautics and Space1
K. YumotoInternational Research Center for Space and Planetary Environmental Science, Kyushu University1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://journal.unhas.ac.id/index.php/geocelebes/oai, pertama di-harvest 3 Nov 2026, terakhir terlihat 15 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en">Physically Consistent Peak Ground Acceleration Prediction in Bali Using Ensemble Machine Learning</dc:title>
	<dc:creator xml:lang="en">Hielmy, Rayhan Irfan</dc:creator>
	<dc:creator xml:lang="en">Muzli, Muzli</dc:creator>
	<dc:subject xml:lang="en">Attenuation relationship</dc:subject>
	<dc:subject xml:lang="en">Ensemble learning</dc:subject>
	<dc:subject xml:lang="en">Hyperparameter tuning</dc:subject>
	<dc:subject xml:lang="en">Seismic hazard assessment</dc:subject>
	<dc:subject xml:lang="en">Strong-motion records</dc:subject>
	<dc:description xml:lang="en">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 &amp;gt; 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.</dc:description>
	<dc:publisher xml:lang="en">Departemen Geofisika, FMIPA - Universitas Hasanuddin, Makassar</dc:publisher>
	<dc:date>2026-10-02</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.unhas.ac.id/index.php/geocelebes/article/view/51982</dc:identifier>
	<dc:identifier>10.70561/akefdb62</dc:identifier>
	<dc:source xml:lang="en">JURNAL GEOCELEBES; Vol. 10 No. 2: October 2026; 306–321</dc:source>
	<dc:source>2579-5546</dc:source>
	<dc:source>2579-5821</dc:source>
	<dc:language>en</dc:language>
	<dc:relation>https://journal.unhas.ac.id/index.php/geocelebes/article/view/51982/15079</dc:relation>
	<dc:rights xml:lang="en">Copyright (c) 2026 JURNAL GEOCELEBES</dc:rights>
	<dc:rights xml:lang="en">https://creativecommons.org/licenses/by-nc-sa/4.0</dc:rights>
</oai_dc:dc>