Desain dan Evaluasi Diagram Kontrol Extended Exponentially Weighted Moving Variance (EEWMV) untuk Pemantauan Varians Proses

Angrenani, Arin Berliana (2022) Desain dan Evaluasi Diagram Kontrol Extended Exponentially Weighted Moving Variance (EEWMV) untuk Pemantauan Varians Proses. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6003201013-Master_Thesis.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Kualitas produk mempunyai hubungan yang erat dengan kepuasan pelanggan. Oleh karena itu, perusahaan selalu berupaya mempertahankan kualitas produk yang dimiliki dengan melakukan pemantauan proses produksi. Pemantauan proses secara statistik dapat menggunakan diagram kontrol. Diagram kontrol umumnya mengamati rata-rata atau varians proses. Dalam sebagian besar proses produksi, memantau perubahan dalam varians proses lebih penting dibanding rata-rata proses. Namun, penggunaan diagram kontrol untuk memantau varians proses belum sebanyak untuk memantau rata-rata proses. Contoh diagram kontrol untuk memantau varians proses adalah Exponentially Weighted Moving Variance (EWMV) yang menggunakan bobot yang sama pada pengamatan saat ini dan sebelumnya serta New Exponentially Weighted Moving Average (NEWMA) yang menggunakan transformasi logaritma varians sampel. Statistik diagram kontrol Extended EWMA (EEWMA) dibentuk dengan memberikan bobot positif pada pengamatan saat ini dan bobot negatif pada pengamatan sebelumnya, sehingga menghasilkan varians yang lebih kecil dalam memantau rata-rata proses. Namun, diagram kontrol EEWMA belum dapat digunakan untuk pemantauan varians proses, sehingga diusulkan diagram kontrol Extended EWMV (EEWMV). Pada penelitian ini, disajikan desain dari statistik EEWMV dengan pendekatan EEWMA (EEWMVM) dan pendekatan transformasi logaritma (EEWMVL) serta menetapkan batas kendali guna memantau varians proses. Kinerja diagram kontrol EEWMVM lebih baik dibandingkan kedua diagram kontrol yang telah ada. Hal ini dibuktikan oleh nilai ARL1 yang lebih kecil pada pergeseran kecil hingga besar dan plot menggunakan data karakteristik kualitas gula, Brix, yang mendeteksi out-of-control lebih cepat. Sementara kinerja diagram kontrol EEWMVL lebih baik dalam mendeteksi pergeseran besar. Nilai ... berada di antara 0,1 hingga 0,3 serta ... mendekati .... bekerja lebih baik dibandingkan .... dan ....2 lainnya pada kedua diagram kontrol usulan. Tidak sepeti diagram kontrol EEWMVL, diagram kontrol EEWMVM dapat mendeteksi pergeseran varians proses pada data simulasi.
====================================================================================================================================
Product quality has a close relationship with customer satisfaction, so the company always strives to maintain the quality of its products by monitoring the production process. Statistical process monitoring can use control charts. Control charts generally observe the mean or variance of the process. In most production processes, monitoring changes in process variance is more important than process mean. But, control charts for monitoring the process variance were not as much as control charts for monitoring the process mean. Examples of control charts for uniting process variances are the Exponentially Weighted Moving Variance (EWMV) which has the same weight on the current and previous observations, and the New Exponentially Weighted Moving Average (NEWMA), which uses the logarithmic transformation of the sample variance. Statistical control chart Extended EWMA (EEWMA) is formed by assigning a positive weight to the current observation and a negative weight to the previous observation, resulting in a smaller variance in monitoring the process average. Statistical control chart Extended EWMA (EEWMA) is formed by assigning a positive weight to the current observation and a negative weight to the previous observation, resulting in a smaller variance in monitoring the process average. However, the EEWMA control chart cannot be used for monitoring process variance, so the Extended EWMV (EEWMV) control chart is proposed. In this study, the design of EEWMV statistics with the EEWMA approach (EEWMVM) and the logarithmic transformation approach (EEWMVL) is presented as well as setting control limits to monitor process variance. The performance of the EEWMVM control chart is better than the two existing control charts, evidenced by the smaller ARL1 values on small to large shifts and plots using the sugar quality characteristic data, Brix, which detects out-of-control faster. In contrast, the performance of the EEWMVL control chart is better in detecting large shifts. The value of

Item Type:	Thesis (Masters)
Additional Information:	RTSt 519.86 Ang d-1 2022
Uncontrolled Keywords:	ARL, EWMV, Gula, Pengendalian Kualitas, Varians, Quality Control, Sugar, Variance.
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD9980.5 Service industries--Quality control.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	29 Apr 2026 07:55
Last Modified:	29 Apr 2026 08:22
URI:	http://repository.its.ac.id/id/eprint/132936

Actions (login required)

View Item