Evaluasi dan Optimasi Ground Motion Model untuk Gempa Subduksi Interface Pulau Jawa Berbasis Algoritma Metaheuristik

Atha, Ilhan Mantiz (2026) Evaluasi dan Optimasi Ground Motion Model untuk Gempa Subduksi Interface Pulau Jawa Berbasis Algoritma Metaheuristik. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5001221036-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Pulau Jawa dapat diklasifikasikan kedalam wilayah dengan aktivitas seismik tinggi akibat penunjaman Lempeng Indo-Australia ke bawah Lempeng Eurasia melalui sistem Sunda Megathrust yang terjadi pada zona subduksi. Pada penelitian ini, empat Ground Motion Model (GMM) yakni model Atkinson & Boore (2003) (AB03), Lin & Lee (2008) (LL08), Zhao et al. (2006) (Zhao06), dan Ashadi & Kaka (2019) (AK19) dievaluasi dan dioptimasi menggunakan 3.384 rekaman dari 73 kejadian gempa subduksi interface di Pulau Jawa periode 2020–2024 dengan M_w 4,1–5,7 dan jarak hiposenter 22–500 km. Optimasi dilakukan melalui pendekatan one-stage mixed-effects estimation untuk minimalkan fungsi objektif yabg berupa Negative Log-Likelihood dengan menggunakan empat algoritma metaheuristik yakni, Particle Swarm Optimization (PSO), Teaching Learning Based Optimization (TLBO), Grey Wolf Optimizer (GWO), dan Gazelle Optimization Algorithm (GOA). Performa hasil optimasi dievaluasi menggunakan analisis korelasi, metrik evaluasi berdasarkan metrik akurasi dan pemilihan model, serta analisis residual terhadap magnitudo, jarak, dan kondisi tanah. Setelah optimasi, seluruh GMM mengalami peningkatan performa berdasarkan parameter-parameter evaluasi. Selain itu, uji konvergensi dan Wilcoxon Signed-Rank menunjukkan bahwa performa PSO dan TLBO secara konsisten mengungguli GWO dan GOA pada semua GMM. Kombinasi model LL08 dengan PSO menghasilkan prediksi gempa subduksi interface di Pulau Jawa yang terbaik, dengan Negative Log-likelihood terbaik serta konsistensi statistik terbaik berdasarkan uji Wilcoxon Signed-Rank pada 30 run independen.
========================================================================================================================================
Java Island can be classified as a region of high seismic activity due to the subduction of the Indo-Australian Plate beneath the Eurasian Plate through the Sunda Megathrust system in the subduction zone. In this study, four Ground Motion Models (GMMs) Atkinson & Boore (2003) (AB03), Lin & Lee (2008) (LL08), Zhao et al. (2006) (Zhao06), and Ashadi & Kaka (2019) (AK19) were evaluated and optimized using 3,384 records from 73 interface subduction earthquake events in Java for the 2020–2024 period, with Mw 4.1–5.7 and hypocentral distances of 22–500 km. Optimization was performed using a one-stage mixed-effects estimation approach to minimize the Negative Log-Likelihood objective function, employing four metaheuristic algorithms, Particle Swarm Optimization (PSO), Teaching-Learning-Based Optimization (TLBO), Grey Wolf Optimizer (GWO), and Gazelle Optimization Algorithm (GOA). The performance of the optimization results evaluated using correlation analysis, evaluation metrics based on accuracy metrics and model selection, as well as residual analysis against magnitude, distance, and soil conditions. After optimization, all GMMs showed performance improvements across these evaluation parameters. Furthermore, convergence testing and the Wilcoxon Signed-Rank test showed that PSO and TLBO consistently outperformed GWO and GOA across all GMMs. The combination of the LL08 model with PSO produced the best ground-motion prediction for interface subduction earthquakes in Java, with the best Negative Log-likelihood and the best statistical consistency based on the Wilcoxon Signed-Rank test over 30 independent runs.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Algoritma Metaheuristik, Gempa Subduksi Interface, GMM, Mixed-effect, Pulau Jawa., Metaheuristic Algorithm, Interface Subduction Earthquake, GMM, Mixed-effects, Java Island.
Subjects:	G Geography. Anthropology. Recreation > GB Physical geography Q Science > QA Mathematics > QA9.58 Algorithms Q Science > QC Physics Q Science > QE Geology > QE538.8 Earthquakes. Seismology Q Science > QE Geology > QE539.2.S4 Seismic models
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Ilhan Mantiz Atha
Date Deposited:	04 Aug 2026 01:27
Last Modified:	04 Aug 2026 01:27
URI:	http://repository.its.ac.id/id/eprint/142677

Actions (login required)

View Item