Haromain, Akhmad Imam (2025) Pengendalian Kualitas Statistika Produk Portland Composite Cement (PCC) Di PT Semen Tonasa Menggunakan Maximum Half-Normal Multivariate Control Chart (Max-Half-Mchart) Berbasis Regression Adjusted Control Chart. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003211146-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003211146-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy |
Abstract
Penelitian ini bertujuan untuk melakukan pengendalian kualitas produk semen PCC di PT Semen Tonasa dengan pendekatan diagram kendali multivariat Max-Half-Mchart berbasis Regression Adjusted Control Chart. Metode ini digunakan karena karakteristik kualitas utama, yaitu kuat tekan semen (3 hari, 7 hari, dan 28 hari) memiliki hubungan yang erat satu sama lain dan dipengaruhi oleh beberapa variabel prediktor yaitu faktor fisika serta faktor kimia, seperti LOI, MgO, SO3, BTL, FL, Blaine, dan R45. Analisis dilakukan dengan menggunakan model regresi nonparametrik spline truncated untuk mendapatkan residual dari masing-masing kuat tekan semen. Penggunaan model tersebut dilakukan berdasarkan hasil uji Terasvirta yang menunjukkan adanya hubungan nonlinear antara variabel prediktor dan respon. Setelah mendapatkan residual yang memenuhi ε_i~i.i.d.N_p (0,Σ), yaitu asumsi identik, independen, dan distribusi normal, residual tersebut kemudian dimonitor menggunakan diagram kendali Max-Half-Mchart. Diagram kendali Max-Half-Mchart ini menggunakan pendekatan distribusi Half-Normal. Pengembangan ini bertujuan mendeteksi pergeseran kecil dan besar pada proses produksi serta mengurangi risiko kesalahan false alarm yang sering terjadi pada diagram kendali berbasis distribusi normal standar. Pada diagram kendali Max-Half-Mchart fase I belum terkendali secara statistik karena masih terdapat pengamatan out of control. Setelah dilakukan penanganan terhadap pengamatan out of control, maka didapatkan data fase I yang telah in control. Pada fase II, diketahui terdapat 1 pengamatan out of control yang teridentifikasi sebagai pergeseran rata-rata proses sehingga pada fase II belum terkendali secara statistik. Pada analisis kapabilitas proses baik secara univariat maupun multivariat dari ketiga kuat tekan semen telah memenuhi batas spesifikasi perusahaan sehingga proses produksi semen PCC telah kapabel.
==========================================================================================================================================
This study aims to conduct quality control of PCC cement products at PT Semen Tonasa using the multivariate Max-Half-Mchart approach based on the Regression Adjusted Control Chart. This method is employed because the main quality characteristics, namely cement compressive strength (at 3 days, 7 days, and 28 days), are closely related to one another and are influenced by several predictor variables, including both physical and chemical factors such as LOI, MgO, SO3, BTL, FL, Blaine, and R45. The analysis is performed using a nonparametric truncated spline regression model to obtain the residuals of each cement compressive strength. The use of this model is based on the results of the Terasvirta test, which indicate a nonlinear relationship between the predictor variables and the responses. After obtaining residuals that meet the ε_i~i.i.d.N_p (0,Σ) assumption, identically distributed, independent, and normally distributed, the residuals are then monitored using the Max-Half-Mchart control chart. This Max-Half-Mchart utilizes the Half-Normal distribution approach. The development of this chart aims to detect both small and large shifts in the production process while reducing the risk of false alarms that frequently occur in control charts based on the standard normal distribution. In Phase I, the Max-Half-Mchart was not statistically in control, as there were still out of control observations. After addressing the out of control observations, Phase I data became in control. In Phase II, one out of control observation was detected, which was identified as a shift in the process mean, indicating that the process in Phase II was not yet statistically in control. However, process capability analysis, both univariate and multivariate, for the three compressive strength measures has met the company’s specification limits, indicating that the PCC cement production process is capable.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Kuat Tekan Semen, Max-Half-Mchart, PT Semen Tonasa, Regression Adjusted Control Chart, Spline Truncated, Cement Compressive Strength, Max-Half-Mchart, PT Semen Tonasa, Regression Adjusted Control Chart, Spline Truncated |
| Subjects: | H Social Sciences > HD Industries. Land use. Labor H Social Sciences > HD Industries. Land use. Labor > HD9980.5 Service industries--Quality control. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Akhmad Imam Haromain |
| Date Deposited: | 31 Jul 2025 06:29 |
| Last Modified: | 31 Jul 2025 06:29 |
| URI: | http://repository.its.ac.id/id/eprint/124682 |
Actions (login required)
 |
View Item |