Pengendalian Kualitas Tetes Tebu di PG Tjoekir Jombang Menggunakan Peta Kendali Multivariate Time Series

Widiyantoro, Eka Rahmadina (2018) Pengendalian Kualitas Tetes Tebu di PG Tjoekir Jombang Menggunakan Peta Kendali Multivariate Time Series. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06211645000007-Undergraduate_Theses.pdf - Published Version
Restricted to Repository staff only

Download (5MB) | Request a copy

Abstract

PG Tjoekir Jombang memiliki 7 stasiun pengolahan dalam melakukan pengolahan tebu menjadi gula pasir. Produk sampingan yang dihasilkan adalah tetes tebu. Tetes tebu yang dihasilkan harus memiliki kandungan sukrosa yang rendah, sehingga gula yang dihasilkan tidak kehilangan sukrosa yang banyak. Tetes tebu diukur dengan dua karakteristik kualitas, yakni persentase Brix dan persentase HK. Dalam proses produksi gula di PG Tjoekir Jombang, tetes tebu yang dihasilkan teridentifikasi memiliki hubungan antar pengamatan sehingga dalam melakukan pengendalian kualitasnya tidak dapat digunakan peta kendali konvensional. Oleh karena itu, pengendalian kualitas dilakukan dengan menggunakan peta kendali multivariat time series pada residual dari model terbaik yang diperoleh pada model VARI(30,1). Hasil analisis menunjukkan pada peta kendali MEWMV diperoleh pembobot optimum omega=0,1 dan lambda=0,2 menunjukkan variabilitas proses yang belum terkendali. Serta, pada peta kendali MEWMA dengan pembobot optimum sebesar lambda=0,9 menunjukkan rata-rata proses yang tidak terkendali pula. Penyebab teridentifikasi adalah pada faktor mesin yang berhenti. ======================= PG Tjoekir Jombang has 7 processing stations in processing sugar cane into sugar. The secondary products are molasses. The molasses must have low sucrose level, so that the level of sucrose which has lost from sugar is not much. Molasses are measured in two quality characteristics, namely percentage of Brix and percentage of HK. In the process of sugar production in PG Tjoekir Jombang, Molasses that produced in time has the auto correlation in process data, so in order to control the quality, conventional control chart can not be used. Therefore, multivariate time series control chart is used based on the residual from the best model, VARI(30,1). The variability process from MEWMV control chart is out-of-control by using the optimum weight omega=0,1 and lambda=0,2 . The mean process from MEWMA control chart is out-of-control as well by using the optimum weight lambda=0,9. The cause of the out-of-control process due to the machine factor.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Autokorelasi, MEWMA, MEWMV, Tetes, VARI, Autocorrelation, Molasses, multivariate time series control chart, Quality control
Subjects: H Social Sciences > HA Statistics > HA30.3 Time-series analysis
Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis
T Technology > TS Manufactures
T Technology > TS Manufactures > TS156 Quality Control. QFD
Divisions: Faculty of Mathematics, Computation, and Data Science > Statistics > (S1) Undergraduate Theses
Depositing User: Eka Rahmadina Widiyantoro
Date Deposited: 16 Oct 2018 05:02
Last Modified: 16 Oct 2018 05:02
URI: http://repository.its.ac.id/id/eprint/52768

Actions (login required)

View Item View Item