Siana Halim
Faculty of Industrial Technology, Petra Christian University

Published : 6 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 6 Documents
Search

PENGGUNAAN BOOTSTRAP DATA DEPENDEN UNTUK MEMBANGUN SELANG KEPERCAYAAN PADA PARAMETER MODEL PERAMALAN DATA STASIONER Siana Halim; Herman Mallian
Jurnal Teknik Industri Vol. 8 No. 1 (2006): JUNE 2006
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (82.073 KB) | DOI: 10.9744/jti.8.1.pp. 54-60

Abstract

The Bootstrap is a lively research area. A lot Of ideas are around and have let to quiet different proposals. In this paper we sketch briefly some Bootstrap methods for independent and dependent data. Finally we give an Bootstrap example for constructing confidence interval in the forecasting for stationer data. Abstract in Bahasa Indonesia : Bootstrap merupakan area penelitian yang terus berkembang. Ada banyak ide dan proposal-proposal yang berbeda telah diberikan oleh para peneliti. Namun demikian, dalam makalah ini hanya akan diulas secara singkat beberapa metode Bootstrap untuk data independen maupun data dependen. Akhirnya akan diberikan sebuah contoh kasus penggunaan Bootstrap untuk membangun selang kepercayaan pada peramalan data stasioner. Kata kunci: Bootstrap, resampling, peramalan
REGRESSION ANALYSIS OF PRODUCTIVITY USING MIXED EFFECT MODEL Siana Halim; Indriati N Bisono
Jurnal Teknik Industri Vol. 9 No. 2 (2007): DECEMBER 2007
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (182.675 KB) | DOI: 10.9744/jti.9.2.pp. 125-130

Abstract

Production plants of a company are located in several areas that spread across Middle and East Java. As the production process employs mostly manpower, we suspected that each location has different characteristics affecting the productivity. Thus, the production data may have a spatial and hierarchical structure. For fitting a linear regression using the ordinary techniques, we are required to make some assumptions about the nature of the residuals i.e. independent, identically and normally distributed. However, these assumptions were rarely fulfilled especially for data that have a spatial and hierarchical structure. We worked out the problem using mixed effect model. This paper discusses the model construction of productivity and several characteristics in the production line by taking location as a random effect. The simple model with high utility that satisfies the necessary regression assumptions was built using a free statistic software R version 2.6.1.
APLIKASI MARKOV RANDOM FIELD PADA MASALAH INDUSTRI Siana Halim
Jurnal Teknik Industri Vol. 4 No. 1 (2002): JUNE 2002
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (44.669 KB) | DOI: 10.9744/jti.4.1.pp. 19-25

Abstract

Markov chain in the stochastic process is widely used in the industrial problems particularly in the problem of determining the market share of products. In this paper we are going to extend the one in the random field so called the Markov Random Field and applied also in the market share problem with restriction the market is considered as a discrete lattice and Pott's models are going to be used as the potential function. Metropolis sampler is going to be used to determine the stability condition. Abstract in Bahasa Indonesia : Rantai Markov dalam proses stokastik seringkali digunakan dalam penyelesaian masalah industri khususnya dalam masalah penentuan market share. Dalam artikel ini akan dibahas perluasan Rantai Markov tesebut ke dalam sebuah random field yang disebut sebagai Markov Random Field (MRF) yang juga akan diaplikasikan pada masalah market share dengan batasan daerah pemasarannya dianggap sebagai sebuah lattice diskrit dan fungsi potensial yang akan digunakan adalah Potts models. Akan digunakan Metropolis sampler untuk menentukan kondisi stabil. Kata kunci: proses stokastik, Markov Random Field, Gibbs Random Field, Potts model, Metropolis sampling.
PETA KENDALI X DENGAN UKURAN SAMPEL DAN INTERVAL PENGAMBILAN SAMPEL YANG BERVARIASI Pauline Astari Singgih; Siana Halim; Tanti Octavia
Jurnal Teknik Industri Vol. 2 No. 2 (2000): DESEMBER 2000
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (81.058 KB) | DOI: 10.9744/jti.2.2.pp. 72-83

Abstract

Shewhart X chart is widely used in statistical process control for monitoring variable data and has shown good performance in detecting large mean shift but less sensitive in detecting moderate to small process shift. X chart with variable sample size and sampling interval (VSSI X chart) is proposed to enhance the ability of detecting moderate to small process shift. The performance of VSSI X chart is compared with those of Shewhart X chart, VSS X chart (Variable Sample Size X chart) and VSI X chart (Variable Sampling Interval X chart). Performance of these control charts is presented in the form of ATS (Average Time to Signal) which is obtained from computer simulation and markov chain approach. The VSSI X chart shows better performance in detecting moderate mean shift. The simulation is then continued for VSSI X chart and VSS X chart with minimum sample size n 1=1 and n 1=2. Abstract in Bahasa Indonesia : Peta kendali X Shewhart telah umum digunakan dalam pengendalian proses statistis untuk data variabel dan terbukti berfungsi dengan baik untuk mendeteksi pergeseran rerata yang besar, namun kurang cepat dalam mendeteksi pergeseran rerata yang sedang hingga kecil. Untuk mengatasi kelemahan ini, diusulkan penggunaan peta kendali X dengan ukuran sampel dan interval pengambilan sampel yang bervariasi (peta kendali VSSI). Kinerja peta kendali X VSSI dibandingkan dengan kinerja peta kendali Shewhart, peta kendali X VSS (peta kendali X dengan ukuran sampel yang bervariasi), dan peta kendali X VSI (peta kendali X dengan interval waktu pengambilan sampel yang bervariasi). Kinerja peta kendali dinyatakan dalam nilai ATS (Average Time to Signal) yang didapatkan dari hasil simulasi program komputer maupun perhitungan Rantai Markov. Peta kendali X VSSI terbukti mempunyai kinerja yang lebih baik dalam mendeteksi pergeseran rerata yang sedang. Selain itu juga disimulasikan penggunaan peta kendali X VSSI dan peta kendali X VSS dengan ukuran sampel minimum n1=1 dan n1=2. Kata kunci: peta kendali X, variable sample size, variable sampling interval, ATS.
PENERAPAN JARINGAN SARAF TIRUAN UNTUK PERAMALAN Siana Halim; Adrian Michael Wibisono
Jurnal Teknik Industri Vol. 2 No. 2 (2000): DESEMBER 2000
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (139.371 KB) | DOI: 10.9744/jti.2.2.pp. 106-114

Abstract

Many methods have been developed to get the optimal result in forecasting. One of them that will be used in this paper is using Neural Network for forecasting. The result will be compared with GARCH(1,1) in the terms of Means Absolute Deviation (MAD) and Means Square Error (MSE). Besides that the accuracy and the power to damp the jump will be observed. The data is currency rate from 4 countries in Asia taken during the Asian Monetary Crisis from 1997 up to 1999 since the jump was happened in that series. Abstract in Bahasa Indonesia : Ada banyak metode yang telah dikembangkan untuk mencapai hasil yang optimal dari suatu peramalan. Salah satu yang akan diulas pada makalah ini adalah penggunaan Neural Network atau jaringan saraf untuk mendapatkan hasil peramalan yang diharapkan dapat meningkatkan optimasi dan akurasinya. Hasil dari metode ini akan dibandingkan dengan metode GARCH(1,1) dalam bentuk Means Absolute Deviation (MAD) dan Means Square Error (MSE). Selain itu dilakukan pula pengamatan terhadap peredaman jump (perubahan mendadak). Data yang digunakan adalah nilai tukar mata uang dari empat negara di Asia yang diambil selama krisis moneter di Asia. Kata Kunci: Backpropagation, MAD, MSE, GARCH(1,1), jump.
PENDEKATAN MODEL MATEMATIS UNTUK MENENTUKAN PERSENTASE MARKUP HARGA JUAL PRODUK Oviliani Yenty Yuliana; Yohan Wahyudi; Siana Halim
Jurnal Teknik Industri Vol. 4 No. 2 (2002): DECEMBER 2002
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1074.253 KB) | DOI: 10.9744/jti.4.2.pp. 58-72

Abstract

The purpose of this research is to design Mathematical models that can determine the selling volume as an alternative to improve the markup percentage. Mathematical models was designed with double regression statistic. Selling volume is a function of markup, market condition, and substitute condition variables. The designed Mathematical model has fulfilled by the test of: error upon assumption, accurate model, validation model, and multi collinear problem. The Mathematical model has applied in application program with expectation that the application program can give: (1) alternative to decide percentage markup for user, (2) Illustration of gross profit estimation that will be achieve for selected percentage markup, (3) Illustration of estimation percentage of the units sold that will be achieve for selected percentage markup, and (4) Illustration of total net income before tax will get for specific period. Abstract in Bahasa Indonesia : Penelitian ini bertujuan untuk merancang model Matematis guna menetapkan volume penjualan, sebagai alternatif untuk menentukan persentase markup harga jual produk. Model Matematis dirancang menggunakan Statistik Regresi Berganda. Volume penjualan merupakan fungsi dari variabel markup, kondisi pasar, dan kondisi pengganti. Model Matematis yang dirancang sudah memenuhi uji: asumsi atas error, akurasi model, validasi model, dan masalah multikolinearitas. Rancangan model Matematis tersebut diterapkan dalam program aplikasi dengan harapan dapat memberi: (1) alternatif bagi pengguna mengenai berapa besar markup yang sebaiknya ditetapkan, (2) gambaran perkiraan laba kotor yang akan diperoleh setiap pemilihan markup, (3) gambaran perkiraan persentase unit yang terjual setiap pemilihan markup, dan (4) gambaran total laba kotor sebelum pajak yang dapat diperoleh pada periode yang bersangkutan. Kata kunci: model Matematis, aplikasi program, volume penjualan, markup, laba kotor.