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

Published : 29 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 9 Documents
Search
Journal : Statistika

Derajat Ketaksesuaian dan Derajat Kedivergenan Kalman filter Sutawanir Darwis; Aceng Komarudin Mutaqin; Yayat Karyana; Mohammand Sobri
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 14, No 2 (2014)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v14i2.1208

Abstract

Tulisan ini membahas beberapa ukuran derajat ketaksesuaian dan derajat kekonvergenan Kalmanfilter yang didefinisikan melalui matriks kovariansi. Pencilan mempengaruhi kinerja penaksiranstate. Dijabarkan pengaruh pencilan terhadap taksiran state dan kovariansi error.
Nonlinear Least Squares (NLS) Tracer Model Siti Rahmawati; Sutawanir Darwis
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 9, No 1 (2009)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v9i1.992

Abstract

Tracer test is used to study possible temperature decline and cooling of production well during longtermre-injection through studying connection between injection and production well. Tracer testinvolve injecting a chemical tracer into a hydrological system and monitoring it’s recovery throughtime. Concentration tracer in production well is nonlinear model. Nonlinear Least Squares Method isproposed to estimate the tracer parameter, i.e. average fluid velocity in the channel and dispersioncoefficient, by using simulation data tracer concentration from production wells AH-4bis, AH-19, andAH-22 in Ahuacapan, Elsavador. The results of estimation are used for predicting thermalbreakthrough and cooling production well during long term re-injection.
Updating Reservoir Models Using Ensemble Kalman Filter Sutawanir Darwis; AGUS YODI GUNAWAN; SRI WAHYUNINGSIH; NURTITI SUNUSI; ACENG KOMARUDIN MUTAQIN; NINA FITRIYATI
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 10, No 1 (2010)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v10i1.1007

Abstract

The Ensemble Kalman Filter (EnKF) has gain popularity as a methodology for real time updates ofreservoir models. A sample of models is updated whenever observation data available. Successfulapplication of EnKF to estimate reservoir properties has been reported. A flow modeling is missing inthis research area. This paper presents the applicability of EnKF in flow modeling for three cases:infinite reservoir, bounded reservoir and one dimensional composite reservoir. The solution of flowequation was derived and used as a modeling component of state space modeling of Kalman filterupdating formula. This three reservoir models shows that the EnKF methodology can be used forupdating the reservoir models.
ESTIMASI KERNEL FUNFSI INTENSITAS KEMUNCULAN GEMPA Ocke Kurniandi; Sutawanir Darwis
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 3, No 1 (2003)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v3i1.553

Abstract

Topik yang dibahas dalam tulisan ini adalah estimasi kernel untuk fungsi intensitas kemunculan gempa bumi dilingkungan sekitar pulau Jawa. Kemunculan gempa dapat dimodelkan sebagai proses Poisson spasial-waktu inhomogen dalamtulisan ini dibatasi pada masalah kemunculan gempa bumi dengan intensitas kekuatan gempa lebih dari 4 skala righter padadaerah domain suatu himpunan kompak di R2 (longitude dan Latitude). Estimasi kernel di R2 ini menghasilkan suatu peta konturintensitas kemunculan gempa yang berguna untuk prediksi daerah gempa.
Interpolasi Decline Rate Menggunakan Kriging Sekuensial Sutawanir Darwis; Gan Gan Saefullah
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 6, No 2 (2006)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v6i2.941

Abstract

Kriging sebagai metode interpolasi membutuhkan proses inversi matriks korelasi antar sample.Secara empiris, observasi yang berada jauh dari titik interpolasi cenderung memiliki bobot nol ataunegative (screen effect). Tulisan ini mengusulkan metode kriging sekuensial untuk interpolasi lajupenurunan (decline rate) laju alir (mass flow) sumur produksi disekitar sumur produksi.Krigingsekuensial merupakan proses updating penambahan data disekitar titik interpolasi. Laju penurunanproduksi ditaksir melalui regresi ln(mass flow) terhadap waktu menggunakan model laju alireksponensial. Korelasi spatial laju alir penurunan mass flow interpolasi kriging sekuensialmengikuti pola sperikal. Interpolasi kriging sekuensial menghasilkan kontur merepresentasikandistribusi laju penurunan disekitar sumur produksi. Hasil interpolasi dapat digunakan untukmemperkirakan cadangan (energi) lapangan panas bumi.
Statistical Process Control Vibrasi Bearing untuk Identifikasi Degradasi Riyani Desriawati; Sutawanir Darwis; Nusar Hajarisman; Suliadi Suliadi; Achmad Widodo
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 20, No 1 (2020)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v20i1.5298

Abstract

Statistical Process Control (SPC) is usually applied to  the production process of goods, with the aim of detecting the quality of a production item that is within or beyond the specified specifications. In this study, SPC was applied to the bearing vibration signal to detect the first observable defect on a machine that functions as part of a prognostic tool for maintenance decision making. The detection of damage and prognostic are two important aspects in machine maintenance based on current conditions or better known as Condition (data) Based Maintenance (CBM). This paper discusses the shewhart average level chart and adaptive shewhart average level chart to detect the first observable defect. The  shewhart chart is built with two assumptions, i.e. that the data must vary randomly around an established mean and follows a normal distribution. However, the adaptive Shewhart  chart there is no need for normal assumption. The exploration of our data shows that the assumption of normality is not fulfilled, so that the Shewhart average level chart is not implemented. The adaptive Shewhart  chart shows that the warning line for bearing 1 amounted to 5.547 and 3.631, for bearing 2 amounted to 5.491 and 3.635, for bearing 3 amounted to 5.762 and 3, 573, for bearing 4 of 5.604 and 33.615. The action line for bearing 1 is 6.026 and 3.152, for bearing 2 is 5.955 and 3.171, for bearing 3 is 6.309 and 3.026, for bearing 4 is 6.101 and 3.118. The first observable defect was t = 81 for bearing 1,  t = 146 for bearing 2,  t = 40 for bearing 3 and  t = 61 for bearing 4.  The adaptive Shewart chart can be used as a toll to estimate the initiation of transition state from normal to degenerate.
Run-Off Triangle Data dan Permasalahannya Aceng Komarudin Mutaqin; Dumaria R. Tampubolon; Sutawanir Darwis
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 8, No 1 (2008)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v8i1.975

Abstract

Run-off triangle data sering digunakan sebagai dasar untuk menaksir cadangan kerugian dari suatuperusahaan asuransi umum kelas bisnis long-tail. Run-off triangle data memuat gambaran klaimkeseluruhan (aggregate), dan merupakan ringkasan dari suatu data set klaim-klaim individu.Makalah ini mengupas run-off triangle data tersebut bersama-sama dengan permasalahan yang adadi dalamnya. Dalam makalah ini dikemukakan dua masalah, yaitu pertama untuk kasus-kasustertentu, tidak semua data dalam run-off triangle teramati. Kemudian masalah yang kedua adalahadanya nilai-nilai incremental data yang negatif dalam run-off triangle terutama dalam incurred claimsdata.
Second OrderMeasurement Model of Teaching Performance Sutawanir Darwis; A Zanbar Soleh; Roslan Lacutonsina
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 5, No 2 (2005)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v5i2.919

Abstract

This paper uses second order measurement model to analyse teaching performance. Theinstrument comprising lecturer competency, lecturer attitude, lecturer organization, and studentmotivation. A survey of first year calculus student was used to test the relationships between theconstructs in the model. Students were ask to evaluate the teaching activities using a 15 itemsquestionaire on a five point Likert scale ranging from strongly agree to strongly disagree. In general,the survey results supported the proposed second order measurement model. The data showed thatthe latent construct were strongly correlated and significantly affected the quality teaching strategy.
Analyzing Pattern of Mutation in mtDNA Using Markov Chain Lira Adiyani; Sutawanir Darwis; Achmad Saifuddin Noer
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 8, No 2 (2008)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v8i2.982

Abstract

Mutation in mtDNA becomes an interesting topic that needed to discuss. If someone has a mutationin his mtDNA, then it might be affect his health. Those effects could be some diseases or anothervariation that gives different characteristics. In study of mutation, there are two such things becomethe main problems: (1) does mutation occur dependently? and (2) what is the pattern? From theresearch (by 9 degrees of freedom χ2), DNA sequence shows a positional dependence. In addition, wecan also see a positional dependence in mtDNA sequence clearly (position i-1, i, i+1 are dependentwith i define as mutation) by sign test, which means, it is possibly that there is a pattern of mutation.This paper uses Markov chain to quantify the pattern and as results all bases will mutate if positioni+1 is C or cytosine (±40%). Moreover, A, C, and G will mutate (become T) if position i-1 is A oradenine (54.5%).