Sutawanir Darwis
Statistics Research Division, Faculty Of Mathematics And Natural Sciences, Institut Teknologi Bandung, Bandung,

Published : 28 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 28 Documents
Search

Deteksi Pemalsuan Minyak Zaitun Menggunakan Spektroskopi FTIR dengan Metode Kemometrika PCA-SVM Ghassany Fathiyah Kamal; Sutawanir Darwis
Bandung Conference Series: Statistics Vol. 4 No. 2 (2024): Bandung Conference Series: Statistics
Publisher : UNISBA Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/bcss.v4i2.12869

Abstract

Abstract. The issue of extra virgin olive oil adulteration in Europe in November 2023 has negatively impacted consumer safety and trust in the product. To address this problem, a study was conducted to detect adulteration using Fourier Transform Infrared (FTIR) spectroscopy combined with chemometrics Principal Component Analysis (PCA) and Support Vector Machine (SVM). FTIR is a spectroscopic technique that analyzes functional groups and molecular structures by examining the interaction of molecules with infrared radiation. PCA was employed to reduce the dimensionality and visualize the FTIR spectral data. SVM was used to classify the samples into their appropriate categories. The data used in this study was secondary, namely olive oil, pork oil, and their mixtures, which were tested using FTIR spectroscopy. The results indicated that the PCA-SVM multiclass one-against-one, using a polynomial kernel with a cost of 0.1, gamma of 0.01, and degree of 4, successfully detected olive oil adulteration with an average accuracy of 91.11%. The combination of FTIR spectroscopy with PCA-SVM chemometrics is effective in detecting olive oil adulteration. This research is expected to help combat olive oil adulteration, thereby protecting consumer safety and restoring trust in the product. Abstrak. Isu pemalsuan minyak zaitun ekstra virgin yang terjadi di Eropa pada November 2023 menimbulkan dampak negatif terhadap keamanan dan kepercayaan konsumen pada produk. Dalam upaya mengatasi masalah ini, dilakukan penelitian untuk mendeteksi pemalsuan tersebut menggunakan metode spektroskopi Fourier Transform Infra Red (FTIR) dengan kemometrika Principal Component Analysis (PCA) dan Support Vector Machine (SVM). FTIR adalah teknik spektroskopi yang memanfaatkan interaksi molekul dengan radiasi inframerah untuk menganalisis gugus fungsi dan struktur kimia molekul. Metode PCA digunakan untuk mengurangi dimensi dan visualisasi data spektra FTIR. Metode SVM digunakan untuk mengklasifikasikan sampel ke dalam kategori yang tepat. Data yang digunakan dalam penelitian ini adalah data sekunder, yaitu minyak zaitun, minyak babi, dan campurannya yang diuji menggunakan alat spektroskopi FTIR. Hasil penelitian menunjukkan bahwa metode PCA-SVM multiclass one agains one menggunakan kernel polinomial dengan parameter cost sebesar 0,1, gamma sebesar 0,01, dan degree sebesar 4 berhasil mendeteksi pemalsuan minyak zaitun dengan rata-rata akurasi 91,11%. Metode spektroskopi FTIR yang digabungkan dengan kemometrika PCA-SVM ini efektif untuk mendeteksi pemalsuan minyak zaitun. Dengan adanya penelitian ini, diharapkan dapat membantu mengatasi masalah pemalsuan produk minyak zaitun guna melindungi keamanan konsumen dan memulihkan kepercayaan terhadap produk.
Penerapan Model Autoregressive Fractionally Integrated Moving Average (ARFIMA) dalam Memprediksi Banyak Gempa Bumi di Barat Pulau Jawa Githa Aulia; Darwis, Sutawanir
Bandung Conference Series: Statistics Vol. 4 No. 2 (2024): Bandung Conference Series: Statistics
Publisher : UNISBA Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/bcss.v4i2.13908

Abstract

Abstract. Autoregressive Fractionally Integrated Moving Average (ARFIMA) is capable of describing both short and long-memory time series through the use of fractional differencing (d) values. This study aims to apply the ARFIMA (p,d,q) model to predict the frequency of earthquakes in west of Java, Indonesia, in upcoming periods. Utilizing secondary data from United States Geological Survey (USGS) spanning from 1971 to 2023, the parameters (p,q) were estimated using the maximum likelihood estimation method, while the differencing parameter (d) was estimated using the Rescaled Range Statistics (R/S) method, resulting in d = 0,273. The best fit model was ARFIMA (1;d;1) with the equation (1-〖〖∅_1 B)(1-B)〗^0,273 Z〗_t=θ_1 (B) e_t and with an AIC value of 110,883. The model predicts 7 future periods, indicating a general increase in earthquake activity in west of Java, although fluctuations in the predictions suggest a tendency towards decreasing volatility.Abstrak. Autoregressive Fractionally Integrated Moving Average (ARFIMA) mampu menjelaskan runtun waktu jangka pendek (short memory) maupun jangka panjang (long memory) dengan nilai differencing (d) bernilai pecahan. Tujuan utama penelitian ini adalah bagaimana penerapan model ARFIMA (p,d,q) dalam memprediksi banyak gempa bumi di barat Pulau Jawa pada periode selanjutnya. Menggunakan data sekunder USGS (United States Geological Survey) tahun 1971-2023, estimasi parameter (p,q) menggunakan metode maximum likelihood d dan estimasi parameter differencing (d) dengan metode analisis Rescaled Range Statistics (R/S) memberikan hasil d=0,273, dimana model terbaik terpilih adalah ARFIMA(1;d;1) dengan persamaan model (1-〖〖∅_1 B)(1-B)〗^0,273 Z〗_t=θ_1 (B) e_t dan nilai AIC sebesar 110,883 yang menghasilkan 7 periode prediksi dengan pergerakan kejadian gempa bumi di barat Pulau Jawa relatif meningkat meskipun fluktuasi prediksi cenderung menurun.
Penerapan Model Epidemic Type Aftershock Sequence (ETAS) pada Data Gempa Bumi Jawa Barat Sofi Sopiah; Sutawanir Darwis
Bandung Conference Series: Statistics Vol. 4 No. 2 (2024): Bandung Conference Series: Statistics
Publisher : UNISBA Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/bcss.v4i2.14956

Abstract

Abstract. Earthquake intensity mapping is an important step in mitigating the risk of natural disasters, especially earthquakes. Earthquake activity is still being studied from both seismological and stochastic aspects. One of the stochastic processes that can explain earthquake activity is the point process. The point process is a stochastic process that can explain natural phenomena that are random in both space and time. One of the models in the point process is the epidemic type aftershock sequence (ETAS) model. The ETAS model is expressed by a conditional intensity function that is useful for knowing the chances of earthquake occurrence. Therefore, the purpose of this study is to apply the ETAS model to earthquake data in West Java. Data obtained from the United State Geological Survey (USGS) was declustered with the Gardner Knopoff method and there were 49 clusters and took 1 cluster with the largest earthquake data, namely the earthquake on September 02, 2009 in Tasikmalaya. The maximum likelihood estimation method is used to obtain parameter estimates of the ETAS model. The null model and full model are used to show the magnitude of the exponential distribution, while the full model shows the magnitude of the gamma distribution. With the maximum likelihood method, the parameter estimates of the basic intensity function of the null model and full model were obtained. The earthquake in Tasikmalaya in 2009 obtained the best model, the full model. The results of the parameter estimation are the base seismicity rate of 0.04402, the aftershock productivity of 4.673611e-9, the efficiency of an earthquake with a certain magnitude to produce aftershocks of 59.4750, the time scale of aftershock decay rate of 0.0083, and the aftershock decay rate of 1.3825. Abstrak. Pemetaan intensitas gempa merupakan langkah penting dalam upaya mitigasi risiko bencana alam, khususnya gempa bumi. Aktivitas gempa bumi masih terus dikaji baik dari aspek seismologi maupun aspek stokastik. Salah satu proses stokastik yang dapat menjelaskan aktivitas gempa bumi adalah proses titik. Proses titik merupakan suatu proses stokastik yang dapat menjelaskan fenomena alam dimana sifatnya acak baik dalam ruang maupun waktu. Salah satu model dalam proses titik adalah model epidemic type aftershock sequence (ETAS). Model ETAS dinyatakan dengan fungsi intensitas bersyarat yang berguna untuk mengetahui peluang kemunculan terjadinya gempa bumi. Oleh karena itu tujuan penelitian ini adalah menerapkan model ETAS pada data gempa bumi di Jawa Barat. Data diperoleh dari United State Geological Survey (USGS) di decslustering dengan metode Gardner Knopoff terdapat 49 cluster dan mengambil 1 cluster data gempa bumi yang terbesar yaitu gempa pada tanggal 02 September 2009 yaitu di Tasikmalaya. Metode estimasi likelihood maksimum digunakan untuk memperoleh estimasi parameter model ETAS. Digunakan null model dan full model untuk menunjukkan magnitudo berdistribusi eksponensial, sedangkan full model menunjukkan magnitudo berdistribusi gamma. Dengan metode likelihood maksimum diperoleh estimasi parameter fungsi intensitas dasar dari null model dan full model. Gempa di Tasikmalaya pada tahun 2009 memperoleh model terbaik yaitu full model. Hasil estimasi parameter tersebut yaitu laju kegempaan dasar sebesar 0.04402, produktivitas gempa susulan sebesar 4.673611e-9, efisiensi gempa bumi dengan magnitudo tertentu menghasilkan gempa susulan sebesar 59.4750, skala waktu laju peluruhan gempa susulan sebesar 0.0083, dan laju peluruhan gempa susulan sebesar 1.3825.
Clustering Probabilistic Seismic Hazard Analysis Temporal Epidemic-Type Aftershock Sequence untuk Premi Asuransi Nurfauzan, Arsyi Fatiha; Darwis, Sutawanir
Jurnal Riset Statistika Volume 4, No. 1, Juli 2024, Jurnal Riset Statistika (JRS)
Publisher : UPT Publikasi Ilmiah Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jrs.v4i1.3864

Abstract

Abstract. An earthquake is an event that releases energy in the earth's crustal rocks, which can trigger aftershocks. The economic losses caused by earthquakes can be mitigated through earthquake insurance. However, conventional premium calculations only take into account the main earthquake (mainshock), while aftershocks are excluded through a declustering process. This article presents the development of Clustering Probabilistic Seismic Hazard Analysis (CPSHA) to evaluate earthquake risk by incorporating aftershock effects using a temporal Epidemic-Type Aftershock Sequence (ETAS) model. The approach is an improvement on traditional earthquake insurance premium calculations, which generally only consider the main earthquake. The ETAS model, with parameters θ=(μ,K,c,α,p), is estimated through Maximum Likelihood Estimation (MLE). This analysis combines seismic hazard ETAS with structural damage data to determine the pure premium of earthquake insurance with aftershock. The study uses West Java earthquake data from 21 November 2022 to 18 February 2023, sourced from the BMKG catalogue. The results indicate that customers in the Cianjur area must pay a total insurance premium (TP) of Rp 28.263.945,00 per year for clusters. This approach enhances the comprehension of earthquake risk and can aid in the calculation of more precise insurance premiums. Abstrak. Gempa bumi merupakan peristiwa pelepasan sejumlah energi pada batuan kerak bumi yang dapat memicu gempa susulan (aftershock). Dampak dari gempa bumi dapat menyebabkan kerugian ekonomi. Salah satu cara untuk menanggulangi hal tersebut dengan asuransi gempa bumi. Perhitungan premi konvensional hanya memperhitungkan gempa utama (mainshock) sementara aftershock disisihkan melalui proses declustering. Artikel ini menyajikan pengembangan Clustering Probabilistic Seismic Hazard Analysis (CPSHA) untuk mengevaluasi risiko gempa bumi dengan memasukkan efek aftershock menggunakan model Epidemic-Type Aftershock Sequence (ETAS) secara temporal. Perhitungan premi asuransi gempa bumi umumnya hanya mempertimbangkan gempa utama, namun pendekatan ini menyertakan aftershock dalam estimasi premi. Model ETAS dengan parameter θ=(μ,K,c,α,p) diestimasi melalui Maximum Likelihood Estimation (MLE). Analisis ini menggabungkan seismic hazard ETAS dengan data kerusakan struktur untuk menentukan premi murni asuransi gempa bumi dengan aftershock. Data yang digunakan adalah data gempa bumi Jawa Barat 21 November 2022 sampai 18 Februari 2023 melalui katalog BMKG. Hasil yang diperoleh menunjukkan total premi (TP) asuransi yang harus dibayarkan oleh nasabah untuk cluster di wilayah Cianjur sebesar Rp 28.263.946,00 per tahun. Pendekatan ini memberikan kontribusi pada pemahaman risiko gempa bumi dan dapat digunakan dalam perhitungan premi asuransi yang lebih akurat.
Penentuan Premi Asuransi Gempa Berdasarkan Declustering Katalog Jawa Barat Raisa Filmi Suryahadikusumah; Sutawanir Darwis
Jurnal Riset Statistika Volume 4, No. 2, Desember 2024, Jurnal Riset Statistika (JRS)
Publisher : UPT Publikasi Ilmiah Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jrs.v4i2.5029

Abstract

Abstract. Indonesia is considered as an area which is prone to natural disasters, especially earthquakes. The condition of being at risk of earthquake disaster causes the threat risk of loss due to earthquakes for the people of Indonesia. Earthquake insurance can be a preventive measure for the community in facing the threat of risk of loss due to earthquakes. This study aims to determine the earthquake insurance premium based on the declustering process of the earthquake catalog in the Cianjur Regency, West Java. Declustering is the process of separating the main earthquake data from the predecessor and aftershocks in the earthquake catalog where the data from the separation will then be used for the analysis of seismic hazard calculation. The seismic hazard in this study is calculated by two methods, namely Probabilistic Seismic Hazard Analysis (PSHA) which only considers the main earthquake, and Sequence Based-PSHA (SPSHA) which considers a combination of main earthquakes and aftershocks. The Probability of Exceedance (POE) results from these two methods are then used as the basis for seismic hazard calculations for earthquake insurance premiums. Through Gardner & Knopoff (1974) declustering, 11 clusters of earthquake sequences of were obtained, where each earthquake sequence has 1 main earthquake. By using 1 cluster of earthquake sequences consisting of 1 main earthquake and 74 aftershocks, it was found that the SPSHA method that considers a combination of main and aftershocks can produce a higher POE value than the PSHA method that only considers main earthquakes. Thus, the premium rate resulting from the SPSHA seismic hazard modeling is higher than the premium rate resulting from the PSHA seismic hazard modeling. Abstrak. Indonesia termasuk kedalam wilayah rawan terjadi bencana alam khususnya gempa bumi. Kondisi rawan bencana gempa menimbulkan adanya ancaman risiko kerugian akibat gempa bagi masyarakat Indonesia. Asuransi gempa bumi dapat menjadi langkah preventif masyarakat dalam menghadapi ancaman risiko kerugian akibat gempa. Penelitian ini bertujuan untuk menentukan premi asuransi gempa berdasarkan proses declustering katalog gempa wilayah Kabupaten Cianjur Jawa Barat. Declustering merupakan proses pemisahan data gempa utama dari gempa pendahulu dan gempa susulan pada katalog gempa dimana data hasil pemisahan tersebut selanjutnya akan digunakan untuk analisis perhitungan bahaya kegempaan. Bahaya kegempaan pada penelitian ini dihitung dengan dua metode yaitu Probabilistic Seismic Hazard Analysis (PSHA) yang hanya mempertimbangkan gempa utama, dan Sequence Based-PSHA (SPSHA) yang mempertimbangkan kombinasi gempa utama dan gempa susulan. Hasil Probability of Exceedance (POE) dari kedua metode ini kemudian digunakan sebagai dasar perhitungan bahaya kegempaan untuk premi asuransi gempa. Melalui declustering Gardner & Knopoff (1974) diperoleh hasil 11 cluster rangkaian gempa dimana setiap rangkaian gempa memiliki 1 gempa utama. Dengan menggunakan 1 cluster rangkaian gempa yang terdiri dari 1 gempa utama dan 74 gempa susulan diperoleh hasil bahwa metode SPSHA yang mempertimbangkan kombinasi gempa utama dan gempa susulan dapat menghasilkan nilai POE yang lebih tinggi daripada metode PSHA yang hanya mempertimbangkan gempa utama. Sehingga, tarif premi yang dihasilkan dari pemodelan bahaya kegempaan SPSHA lebih tinggi daripada tarif premi yang dihasilkan dari pemodelan bahaya kegempaan PSHA.
Penerapan Sequence Probabilistic Seismic Hazard Analysis dalam Pemodelan Seismogenic Hazard Novaida Zulfita; Sutawanir Darwis
Jurnal Riset Statistika Volume 4, No. 2, Desember 2024, Jurnal Riset Statistika (JRS)
Publisher : UPT Publikasi Ilmiah Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jrs.v4i2.5163

Abstract

Abstract. Seismogenic areas are areas in the earth's crust that have the potential to produce seismic waves due to tectonic activity. Identification of a seismogenic area measuring 30 x 100 km² with one main earthquake and eight aftershocks is the first step in earthquake risk analysis. Once identified, probabilistic modeling is used to estimate the frequency and magnitude of earthquakes. This modeling uses conditional probability and total probability to combine various earthquake sources, such as magnitude and distance, with data generated via Python software. The probability that the intensity of ground vibrations (Peak Ground Acceleration) will exceed the threshold within a certain period is calculated. This thesis integrates conditional and total probabilities from various earthquake scenarios to calculate the Sequence Probabilistic Seismic Hazard Analysis (SPSHA), which is a combination of main and aftershocks. This process begins with Probabilistic Seismic Hazard Analysis (PSHA) for the main earthquake, then Aftershock Probabilistic Seismic Hazard Analysis (APSHA) for aftershocks. The 1– m – a computational design includes configurations such as 1 – 1 – 0, 1 – 1 – 1, up to 1 – 1 – 8, where 1 represents the site, m the main earthquake, and a the aftershock. The discussion is limited to one and two dimensional spaces with a variety of geometric configurations. The analysis results show the important role of conditional probability in PSHA, APSHA, and SPSHA in seismogenic areas, where SPSHA provides a more realistic approach than PSHA by considering aftershocks that follow the main earthquake. This results in a higher annual rate of exceedance which is useful for earthquake risk analysis, disaster mitigation planning, and earthquake-resistant infrastructure design. Abstrak. Seismogenic area adalah wilayah dalam kerak bumi yang berpotensi menghasilkan gelombang seismik akibat aktivitas tektonik. Identifikasi seismogenic area berukuran 30 x 100 km² dengan satu gempa utama dan delapan gempa susulan adalah langkah awal dalam analisis risiko gempa. Setelah diidentifikasi, pemodelan probabilistik digunakan untuk memperkirakan frekuensi dan magnitudo gempa. Pemodelan ini menggunakan peluang bersyarat dan peluang total untuk menggabungkan berbagai sumber gempa, seperti magnitudo dan jarak, dengan data yang dihasilkan melalui software Python. Probabilitas bahwa intensitas getaran tanah (Peak Ground Acceleration) akan melebihi ambang batas dalam periode tertentu dihitung. Skripsi ini mengintegrasikan peluang bersyarat dan total dari berbagai skenario gempa untuk menghitung Sequence Probabilistic Seismic Hazard Analysis (SPSHA), yang merupakan gabungan gempa utama dan susulan. Proses ini dimulai dengan Probabilistic Seismic Hazard Analysis (PSHA) untuk gempa utama, kemudian Aftershock Probabilistic Seismic Hazard Analysis (APSHA) untuk gempa susulan. Desain komputasi 1 – m – a meliputi konfigurasi seperti 1 – 1 – 0, 1 – 1 – 1, hingga 1 – 1 – 8, dimana 1 menyatakan site, m gempa utama, dan a gempa susulan. Pembahasan dibatasi pada ruang dimensi satu dan dua dengan variasi konfigurasi geometri. Hasil analisis menunjukkan peran penting peluang bersyarat dalam PSHA, APSHA, dan SPSHA di seismogenic area, dimana SPSHA memberikan pendekatan lebih realistis dibandingkan PSHA dengan mempertimbangkan gempa susulan yang mengikuti gempa utama. Ini menghasilkan annual rate of exceedance lebih tinggi yang berguna untuk analisis risiko gempa, perencanaan mitigasi bencana, dan desain infrastruktur tahan gempa.
Law of Total Probability of Aftershocks in Earthquake Insurance Darwis, Sutawanir; Hajarisman, Nusar; Suliadi; Fatiha Nurfauzan, Arsyi; Aulia, Githa
Statistika Vol. 25 No. 1 (2025): Statistika
Publisher : Department of Statistics, Faculty of Mathematics and Natural Sciences, Universitas Islam Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/statistika.v25i1.3886

Abstract

Abstract. Seismic hazard expressed in annual rate of exceedance of a peak ground acceleration traditionally refers to mainshock. A similar seismic hazard, APSHA, has been adopted for aftershock probabilistic seismic hazard. Probabilistic seismic hazard assessment (PSHA) refers to a homogeneous Poisson process to describe mainshock while APSHA models aftershock occurrence as a nonhomogeneous Poisson process whose rate modeled as Omori law. It shown that the combination of PSHA and APSHA results seismic hazard for mainshock-aftershock seismic sequence/cluster (SPSHA/CPSHA). This study shows how to combine results of APSHA and PSHA and proposes a method for earthquake insurance. The study was carried out for West Java region with 206 occurrences consist of 11 clusters. One cluster with 74 aftershocks was chosen for further study. The parameters of SPSHA/CPSHA was estimated using maximum likelihood. The results of SPSHA/CPSHA combined with damage probability matrix (DPM) yields an expected annual damage ratio (EADR) as an indicator of earthquake insurance. The proposed method in this study can be used as a method for computing earthquake insurance premium. Due to limited data further study is needed to obtain accurate and reasonable results.
Monte Carlo Based Sampling Distribution of Annual Rate of Exceedance for Earthquake Insurance Darwis, Sutawanir; Hajarisman, Nusar; Suliadi; Widodo, Achmad; Diahsty Marasabessy, Munira; Arya Ramadhan, Iqbal
Statistika Vol. 24 No. 1 (2024): Statistika
Publisher : Department of Statistics, Faculty of Mathematics and Natural Sciences, Universitas Islam Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/statistika.v24i1.3173

Abstract

ABSTRACT Seismic hazard expressed in terms of annual rate of exceedance and is used to calculate the earthquake insurance premium. Annual rate of exceedance is a complicated function of magnitudes, distances from site to earthquake sources and attenuation. Due to its complexity, determination of exact sampling distribution of earthquake insurance premium is not an easy task. This research proposes Monte Carlo simulation approach to determine the sampling distribution of earthquake insurance premium. Annual rate of exceedance was simulated first and then the insurance premium calculated based on simulation of annual rate of exceedance. The simulation involves quantifying synthetic catalogue similar to historical catalogue. Its simulation is conducted in order to construct annual rate of exceedance as an indicator of earthquake risk used in earthquake insurance. The simulation of 25 iteration and sample size of 100 shows that the sampling distribution of insurance premium is skewed to the right. The idea of using Monte Carlo simulation to study the sampling distribution of annual rate of exceedance is the originality of the study and the study contributes to the methodology of earthquake insurance.