Maulana, R. Irfan Rizky (2024) Pengendalian Kualitas Statistika Pupuk Urea PT Petrokimia Gresik Menggunakan Metode Diagram Kendali MEWMV dan MSEWMA Berbasis Multioutput Least Square Support Vector Regression (MLS-SVR). Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
5003201141-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Pupuk merupakan bahan penting untuk meningkatkan kualitas pertanian di Indonesia. PT Petrokimia Gresik merupakan perusahaan yang memiliki pabrik pupuk terlengkap di Indonesia. Salah satu jenis pupuk bersubsidi yang diproduksi oleh PT Petrokimia Gresik yaitu pupuk urea. Terdapat tiga kandungan pupuk urea yaitu Nitrogen (N), Biuret, dan Air (H_2 O). Semua kandungan yang terdapat di pupuk urea harus terkandung sesuai dengan kadar yang telah ditetapkan, agar manfaat dari pupuk tersebut didapatkan secara maksimal. Oleh karena itu, akan dilakukan pengendalian kualitas statistika pada pupuk urea menggunakan diagram kendali Multivariate Exponentially Weighted Moving Variance (MEWMV) dan Multivariate Sign Exponentially Weighted Moving Average (MSEWMA) berbasis residual Multioutput Least Square Support Vector Regression (MLS-SVR). Pemodelan dilakukan untuk mengatasi adanya autokolerasi pada data. Pada penelitian ini hyperparameter yang terpilih yaitu γ^'=2^(-5), γ^''=2^(-7), dan σ=2^0 dengan nilai MSE sebesar 0,0013. Nilai residual dari pemodelan MLS-SVR akan digunakan dalam pembuatan diagram kendali MEWMV dan MEWMA untuk fase I dan MEWMV, MEWMA, dan MSEWMA untuk fase II. Pada fase I, didapatkan pengamtan yang out of control baik dalam variabilitas maupun rata-rata proses sehingga belum terkendali secara statistik. Pada fase II, mendapatkan hasil bahwa masih terdapat pengamtan yang out of control dalam variabilitas sebanyak 6 pengamatan dan dalam rata-rata proses sebanyak 5 dan 4 pengamatan sehingga belum terkendali secara statistik yang diduga disebabkan oleh variabel nitrogen dan air. Hasil kapabilitas proses menujukkan bahwa proses produksi pupuk belum kapabel secara multivariat.
============================================================
Fertilizer is an important ingredient for improving the quality of agriculture in Indonesia. PT Petrokimia Gresik is a company that has the most complete fertilizer factory in Indonesia. One type of subsidized fertilizer produced by PT Petrokimia Gresik is urea fertilizer. There are three ingredients in urea fertilizer, namely Nitrogen (N), Biuret, and Water (H_2 O). All ingredients contained in urea fertilizer must be contained in accordance with the levels that have been determined, so that the maximum benefits from the fertilizer are obtained. Therefore, statistical quality control will be carried out on urea fertilizer using Multivariate Exponentially Weighted Moving Variance (MEWMV) control charts and Multivariate Sign Exponentially Weighted Moving Average (MSEWMA) based on MLS-SVR residuals. Modeling is carried out to overcome the presence of autocorrelation in the data. In this study, the hyperparameters chosen were γ^'=2^(-5), γ^''=2^(-7), and σ=2^0 with an MSE value of 0.0013. The residual values from MLS-SVR modeling will be used in creating MEWMV and MEWMA control diagrams for phase I and MEWMV, MEWMA and MSEWMA for phase II. In phase I, observations were obtained that were out of control in both variability and process averages so that they were not yet statistically controlled. In phase II, the results showed that there were still observations that were out of control in the variability of 6 observations and in the mean process of 5 and 4 observations so that it was not yet statistically controlled, which was thought to be caused by the variables nitrogen and water. The process capability results show that the fertilizer production process is not yet multivariate capable.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Diagram Kendali, Pupuk Urea, MEWMV, MSEWMA, MLS-SVR, Control Chart, Urea Fertilizer, MEWMV, MSEWMA, MLS-SVR. |
Subjects: | H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics H Social Sciences > HD Industries. Land use. Labor H Social Sciences > HD Industries. Land use. Labor > HD9980.5 Service industries--Quality control. Q Science > Q Science (General) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
Depositing User: | R. Irfan Rizky Maulana |
Date Deposited: | 08 Aug 2024 12:18 |
Last Modified: | 08 Aug 2024 12:18 |
URI: | http://repository.its.ac.id/id/eprint/114940 |
Actions (login required)
![]() |
View Item |