Pengendalian Kualitas Proses Produksi Hasil Gula Kristal Putih di Pabrik Gula Djatiroto PT Perkebunan Nusantara XI Menggunakan Diagram Kontrol Multivariate Maximum Cumulative Sum (Max-MCUSUM) Berbasis Residual Model Multioutput Least Square Support Vector Regression (MLS-SVR)

Yunika, Annisa (2022) Pengendalian Kualitas Proses Produksi Hasil Gula Kristal Putih di Pabrik Gula Djatiroto PT Perkebunan Nusantara XI Menggunakan Diagram Kontrol Multivariate Maximum Cumulative Sum (Max-MCUSUM) Berbasis Residual Model Multioutput Least Square Support Vector Regression (MLS-SVR). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211840000007-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Gula yang merupakan bahan dasar pokok digunakan masyarakat sebagai pemanis dan juga pengawet makanan. Salah satu perusahaan agribisnis yang bergerak fokus pada produksi gula yaitu PT Perkebunan Nusantara XI (PTPN XI). Gula kristal putih adalah gula tebu atau bit yang melalui proses kristalisasi yang digunakan untuk konsumsi rumah tangga yang telah dijelaskan peraturan SNI 3140.3:2010. Pabrik Gula Djatiroto merupakan salah satu dari enam belas pabrik gula yang berada dibawah PTPN XI. Pada penelitian ini karakteristik kualitas yang digunakan yaitu warna larutan, besar jenis butir, dan kadar air yang masing-masing karakteristik kualitas memiliki hubungan. Hubungan pada ketiga variabel menunjukkan adanya kasus autokorelasi. Adanya autokorelasi menyebabkan metode diagram kontrol konvensional tidak lagi optimal untuk digunakan karena asumsi independensi data tidak terpenuhi. Oleh karena itu, penelitian ini dilakukan pengendalian kualitas secara statistik pada produksi gula kristal putih menggunakan diagram kontrol multivariat dengan pendekatan residual. Hal ini disebabkan karena residu yang didapatkan bersifat independen sehingga dapat memonitor proses dengan lebih akurat. Penelitian ini menggunakan metode Residual Model Multioutput Least Square Support Vector Regression (MLS-SVR) yang dikombinasikan dengan diagram kontrol multivariat Maximum Cumulative Sum (Max-MCUSUM). Berdasarkan hasil analisis, didapatkan karakteristik kualitas yang berpengaruh secara signifikan adalah besar jenis butir, warna larutan, dan kadar air. Hasil penelitian menunjukkan bahwa proses produksi gula kristal putih di Pabrik Gula Djatiroto sudah terkendali secara statistik dan kapabel dengan nilai kapabilitas proses secara multivariat MCp sebesar 2,5575 dan MCpk sebesar 2,2605.
==============================================================================================================================
Sugar is a staple used by the community as a sweetener and also a food preservative. One of the agribusiness companies that focuses on sugar production is PT Perkebunan Nusantara XI (PTPN XI). White crystal sugar is cane sugar or beet that goes through a crystallization process used for household consumption which has been explained by SNI regulation 3140.3: 2010. Djatiroto Sugar Factory is one of the sixteen sugar factories under PTPN XI. In this study, the quality characteristics used were the color of the solution, the size of the grain type, and the water content, each of which had a relationship. The relationships on all three variables indicate the presence of cases of autocorrelation. The existence of autocorrelation causes false alarms and gets incorrect decision making. The analysis used was with the Multi-output Least Square Support Vector Regression (MLS-SVR) algorithm model to overcome autocorrelated data and perform control using Max-MCUSUM control diagrams. MLS-SVR model inputs were determined based on significant lag of the PACF plot, where in this study on the color quality characteristics of the solution were significant at lag-1, 2, 11, and 23, the quality characteristics of large grain types were significant at lag-1, 2, 3, and 28, as well as the quality characteristics of significant moisture content at lag-27. The results of the MLS-SVR model input and the optimal combination of hyper-parameters produce residual values that can reduce autocorrelation. Residual values are used in the creation of Max-MCUSUM control diagrams. In phase I with k = 0.5 a statistically controlled diagram was obtained after it was overcome, and in phase II using the same k showed no points that went out of the control boundary. The results of identifying variables that cause out of control obtained variables in the color of the solution and the size of the type of grain that has factors that cause out of control. The results of the process capability have been capable and the process performance has been good in a multivariate manner.

Item Type:	Thesis (Other)
Additional Information:	RSSt 519.86 Yun p-1 2022
Uncontrolled Keywords:	Autocorrelation. Max-MCUSUM. MLS-SVR. Sugar. Quality Control. Autocorrelation. Max-MCUSUM. MLS-SVR. Sugar. Quality Control.
Subjects:	H Social Sciences > HA Statistics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	10 Jun 2026 02:38
Last Modified:	10 Jun 2026 03:26
URI:	http://repository.its.ac.id/id/eprint/133675

Actions (login required)

View Item