Konstruksi Double Modified Exponentially Weighted Moving Average-Tukey Control Chart (DMEWMA-TCC)

Faizana, Falidazia Hasanah (2026) Konstruksi Double Modified Exponentially Weighted Moving Average-Tukey Control Chart (DMEWMA-TCC). Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6003242009-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Statistical Process Control (SPC) merupakan metode statistik yang digunakan untuk memantau dan mengendalikan kualitas proses agar kestabilan proses tetap terjaga. Salah satu alat utama dalam SPC adalah diagram pengendali Exponentially Weighted Moving Average (EWMA) dan pengembangannya seperti Double Modified EWMA (DMEWMA) yang memiliki sensitivitas tinggi dalam mendeteksi pergeseran kecil hingga sedang. Namun, sebagian besar diagram pengendali berbasis EWMA masih menggunakan batas pengendali parametrik yang dibangun berdasarkan asumsi distribusi normal sehingga performanya dapat menurun pada data nonnormal.Penelitian ini bertujuan mengkonstruksi diagram pengendali DMEWMA berbasis Tukey Control Chart (TCC) dengan mengintegrasikan ukuran robust berbasis kuartil dan interquartile range (IQR) pada pembentukan batas pengendali. Evaluasi kinerja dilakukan menggunakan ukuran Average Run Length (ARL) dan Standard Deviation of Run Length (SDRL) melalui simulasi Monte Carlo pada distribusi normal dan Gamma. Selain itu, diagram pengendali diterapkan pada data sintesis serta data kualitas semen berupa FCaO dan C4Af menggunakan indikator out of control (OOC), accuracy, false alarm, dan first signal. Hasil penelitian menunjukkan bahwa DMEWMA-TCC secara umum cenderung menghasilkan nilai ARL dan SDRL yang lebih kecil dibandingkan diagram pengendali DMEWMA sehingga mampu mendeteksi kondisi out of control lebih cepat dan lebih stabil pada distribusi normal maupun Gamma. Pada penerapan data sintesis dan data kualitas semen, DMEWMA-TCC juga menunjukkan performa monitoring yang lebih baik dengan kemampuan mendeteksi kondisi out of control yang lebih tinggi, tingkat akurasi yang lebih baik, nilai false alarm yang relatif rendah, serta first signal yang lebih cepat pada beberapa kondisi. Oleh karena itu, DMEWMA-TCC berpotensi digunakan sebagai alternatif diagram pengendali untuk pemantauan kualitas proses pada data nonnormal.
=====================================================================================================================================
Statistical Process Control (SPC) is a statistical method used to monitor and control process quality in order to maintain process stability. One of the primary tools in SPC is the Exponentially Weighted Moving Average (EWMA) control chart and its developments, such as the Double Modified EWMA (DMEWMA), which have high sensitivity in detecting small-to-moderate process shifts. However, most EWMA-based control charts still employ parametric control limits constructed under the assumption of normality, which may reduce their performance when applied to nonnormal data. This study aims to construct a DMEWMA control chart based on the Tukey Control Chart (TCC) by integrating robust measures based on quartiles and the interquartile range (IQR) into the control limit formulation. Performance evaluation was conducted using Average Run Length (ARL) and Standard Deviation of Run Length (SDRL) through Monte Carlo simulation under normal and Gamma distributions. In addition, the control charts were applied to synthetic data and cement quality data consisting of FCaO and C4Af using out-of-control (OOC), accuracy, false alarm, and first signal indicators. The results indicate that DMEWMA-TCC generally tends to produce smaller ARL and SDRL values than the DMEWMA control chart, enabling faster and more stable detection of out-of-control conditions under both normal and Gamma distributions. In the application to synthetic data and cement quality data, DMEWMA-TCC also demonstrated better monitoring performance with higher out-of-control detection capability, better accuracy, relatively low false alarm values, and faster first signals under several conditions. Therefore, DMEWMA-TCC has the potential to be used as an alternative control chart for process quality monitoring on nonnormal data.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	DMEWMA, DMEWMA-TCC, EWMA, kualitas semen (clinker), Statistical Process Control, cement quality (clinker), DMEWMA, DMEWMA-TCC, EWMA, Statistical Process Control.
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD9980.5 Service industries--Quality control. T Technology > TS Manufactures > TS156 Quality Control. QFD. Taguchi methods (Quality control)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	Falidazia Hasanah Faizana
Date Deposited:	03 Aug 2026 03:58
Last Modified:	03 Aug 2026 03:58
URI:	http://repository.its.ac.id/id/eprint/142208

Actions (login required)

View Item