Wahyudi, Hana Safira (2024) Pengendalian Kualitas Statistika Pupuk ZK di PT Petrokimia Gresik Menggunakan Diagram Kontrol Multivariate Exponentially Weighted Moving Covariance Matrix (MEWMC) dan Multivariate Exponentially Weighted Moving Average (MEWMA) Berbasis Residual Model Time Series. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003201059-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003201059-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (4MB) | Request a copy |
Abstract
Sektor pertanian memiliki peran krusial dalam mendukung keberlanjutan pangan dan perekonomian suatu negara, salah satu elemen penting dalam peningkatan sektor tersebut adalah produksi pupuk. PT Petrokimia Gresik, sebagai anak perusahaan dari PT Pupuk Indonesia, memproduksi berbagai macam pupuk termasuk pupuk ZK (Kalium Sulfat). Pupuk ini memiliki keunggulan dalam sumber unsur kalium (K) dan sulfur (S) yang penting untuk pertumbuhan tanaman. Pupuk ZK memiliki kandungan unsur hara, antara lain H2O, K2O, SO3, FA (H2SO4), dan Cl- yang penting untuk dijaga karakteristiknya agar sesuai dengan spesifikasi sehingga manfaat dari pupuk tersebut dapat dicapai dengan maksimal. Dalam menjaga kualitas dari proses produksi pupuk tersebut, diperlukan pengendalian kualitas statistika menggunakan diagram kontrol multivariat berbasis residual model time series karena antar karakteristik kualitas saling berhubungan dan antar pengamatan terdapat autokorelasi. Adanya autokorelasi tersebut dapat menyebabkan false alarm yang berdampak pada pengambilan keputusan yang tidak tepat. Oleh karena itu, dilakukan pemodelan Vector Autoregressive (VAR) dengan model terbaik yang didapatkan yakni VARI(5) untuk mendapatkan residual yang akan digunakan pada analisis pengendalian kualitas dengan diagram kontrol Multivariate Exponentially Weighted Moving Covariance Matrix (MEWMC) dan Multivariate Exponentially Weighted Moving Average (MEWMA) yang robust terhadap non-normal distribution. Hasil pengendalian kualitas menggunakan kedua diagram kontrol tersebut menunjukkan bahwa variabilitas proses tidak terkendali secara statistik karena pada fase I maupun fase II terdapat pengamatan out of control pada diagram MEWMC dengan pembobot optimum λ = 0,1. Sedangkan, untuk pengendalian rata-rata proses menggunakan diagram MEWMA dengan pembobot optimum λ = 0,1 didapatkan hasil bahwa pada fase I tidak terkendali secara statistik tetapi proses telah in control pada fase II. Dari hasil pengendalian kualitas tersebut, dilakukan identifikasi penyebab out of control dengan kesimpulan bahwa karakteristik kualitas Cl- pada pupuk ZK menjadi penyebab proses produksi tidak terkendali yang dipicu oleh ketidakkonsistenan dalam bahan baku. Berdasarkan analisis kapabilitas proses pada data yang tidak berdistribusi normal, diperoleh hasil bahwa nilai MCNpk lebih dari 1 sehingga keseluruhan proses produksi pupuk ZK dapat dikatakan kapabel.
==============================================================================================================================
The agricultural sector has a crucial role in supporting the sustainability of food and the economy of a country, one of the important elements in improving the sector is fertilizer production. PT Petrokimia Gresik, as a subsidiary of PT Pupuk Indonesia, produces various kinds of fertilizers including ZK (Potassium Sulfate) fertilizer. This fertilizer has advantages in the source of potassium (K) and sulfur (S) elements which are important for plant growth. ZK fertilizer contains nutrients, including H2O, K2O, SO3, FA (H2SO4), and Cl- which are important to maintain the characteristics to match the specifications so that the benefits of the fertilizer can be achieved optimally. In maintaining the quality of the fertilizer production process, statistical quality control is required using multivariate control charts based on residual time series model because the quality characteristics are interconnected and there is autocorrelation between observations. The existence of autocorrelation can cause false alarms that have an impact on improper decision making. Therefore, Vector Autoregressive (VAR) modeling is carried out with the best model obtained namely VARI(5) to obtain residuals that will be used in quality control analysis with Multivariate Exponentially Weighted Moving Covariance Matrix (MEWMC) and Multivariate Exponentially Weighted Moving Average (MEWMA) control charts that are robust to non-normal distribution. The results of quality control using the two control diagrams show that the variability process is not statistically controlled because in phase I and phase II there are out of control observations on the MEWMC diagram with the optimum weight λ = 0,1. Meanwhile, for controlling the mean process using the MEWMA diagram with the optimum weight λ = 0,1, the results show that phase I is not statistically controlled but the process is in control in phase II. From the results of quality control, identification of the causes of out of control is carried out with the conclusion that the quality characteristics Cl- in ZK fertilizer is the cause of the uncontrolled production process triggered by inconsistencies in raw materials. Based on the process capability analysis on data that is not normally distributed, the results show that the value of MCNpk is more than 1 so that the entire ZK fertilizer production process can be said to be capable.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | MEWMC, MEWMA, quality control, VAR, ZK fertilizer, pengendalian kualitas, pupuk ZK |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HD Industries. Land use. Labor > HD9980.5 Service industries--Quality control. Q Science S Agriculture > S Agriculture (General) > S633.5 Fertilizers |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Hana Safira Wahyudi |
| Date Deposited: | 14 Aug 2024 02:00 |
| Last Modified: | 14 Aug 2024 02:00 |
| URI: | http://repository.its.ac.id/id/eprint/114691 |
Actions (login required)
 |
View Item |