Hakim, Deyyana Aulia (2024) Implementasi Diagram Kontrol Maximum MCUSUM (Max-MCUSUM) Berbasis Residual Multioutput Least Square SVR (MLS-SVR) untuk Pengendalian Kualitas Air di PDAM Surya Sembada Kota Surabaya. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003201052-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003201052-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (1MB) | Request a copy |
Abstract
PDAM Surya Sembada merupakan Badan Usaha Milik Daerah (BUMD) pemerintah Surabaya yang bergerak di bidang jasa penyediaan dan pelayanan air bersih bagi masyarakat Kota Surabaya. Pengolahan air di Perusahaan ini diproses melalui Instalasi Penjernihan Air Minum (IPAM) yang terletak di daerah Ngagel dan Karang Pilang. Terdapat 24 variabel dalam kualitas air yang selama ini telah dimonitor oleh PDAM Surya sembada Surabaya. Karakteristik air produksi dimonitor setiap jam sehingga mengindikasikan adanya autokorelasi antar pengamatan. Oleh karena itu, diperlukan metode alternatif untuk mengatasi permasalahan autokorelasi sebelum dilakukan pengendalian kualitas dengan diagram kontrol. Pada penelitian ini, dilakukan pengendalian kualitas air dengan menggunakan diagram Max-MCUSUM berbasis Residual Multioutput Least Square Support Vector Regression (MLS-SVR). Variabel yang digunakan meliputi kekeruhan, zat organik, dan sisa chlor dari IPAM Ngagel II. Ketiga variabel terbukti berautokorelasi antar waktu sehingga perlu adanya penanganan autokorelasi yaitu dengan MLS-SVR. Hasil dari pemodelan MLS-SVR dengan hyper-parameter optimal yaitu optimal yaitu y’ = 210, y’’ = 2−1, dan o = 2−15 menunjukkan bahwa residual telah memenuhi asumsi white noise sehingga dapat digunakan pada diagram kontrol Max-MCUSUM berbasis MLS-SVR. Hasil yang diperoleh pada analisis fase I dengan reference value k senilai 0,5 yaitu diagram telah terkendali secara statistik setelah mengganti 7 pengamatan sehingga dilanjutkan pada pengendalian fase dua. Sedangkan pada fase II, diagram belum terkendali secara statistic yang diduga disebabkan oleh variabel turbidity. Pada analisis kapabilitas proses, kinerja proses produksi air produksi di IPAM Ngagel II belum kapabel. Hasil penelitian ini diharapkan dapat memberikan informasi maupun rekomendasi bagi PDAM Surya Sembada Kota Surabaya dalam memonitor kualitas air produksi yang dihasilkan sehingga dapat meningkatkan kualitas air bersih dan meningkatkan kepuasan pelanggan.
==================================================================================================================================
PDAM Surya Sembada is a Regional-Owned Enterprise (BUMD) of the Surabaya government which is engaged in the provision and service of clean water for the people of Surabaya City. Water treatment in this company is processed through the Drinking Water Purification Plant (IPAM) located in Ngagel and Karang Pilang areas. There are 24 variables in water quality that have been monitored by PDAM Surya sembada Surabaya. The production water characteristics are monitored hourly, indicating autocorrelation between observations. Therefore, an alternative method is needed to overcome the autocorrelation problem before quality control is carried out with a control diagram. In this study, water quality control is carried out using Max-MCUSUM diagram based on Residual Multioutput Least Square Support Vector Regression (MLS-SVR). The variables used include turbidity, organic matter, and residual chlorine from Ngagel II Water Treatment Plant. The three variables are proven to be autocorrelated between time so it is necessary to handle autocorrelation, namely with MLS-SVR. The results of MLS-SVR modeling with optimal hyper-parameters = 210, y’’ = 2−1, and o = 2−15show that the residuals have met the white noise assumption so that they can be used on the MLS-SVR-based Max-MCUSUM control diagram. The results obtained in phase I analysis with a reference value k worth 0.5 are that the diagram has been statistically controlled after replacing 7 observations so that it continues in phase two control. While in phase II, the diagram has not been statistically controlled which is thought to be caused by the turbidity variable. In the process capability analysis, the performance of the production water production process at IPAM Ngagel II is not yet capable. The results of this study are expected to provide information and recommendations for PDAM Surya Sembada Surabaya City in monitoring the quality of production water produced to improve the quality of clean water and increase customer satisfaction.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Air, Analisis Kapabilitas, Autokorelasi, MAX-MCUSUM, MLS-SVR, Pengendalian Kualitas, Autocorrelation, Capability Analysis, Quality Control, Water |
| Subjects: | 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: | Deyyana Aulia Hakim |
| Date Deposited: | 27 Aug 2024 01:56 |
| Last Modified: | 27 Aug 2024 01:56 |
| URI: | http://repository.its.ac.id/id/eprint/115243 |
Actions (login required)
 |
View Item |