Khatulistiwa, Savero Janus (2025) Sistem Prediksi Nilai Overall Equipment Effectiveness Untuk Peringatan Breakdown Dengan Metode Recurrent Neural Network Pada Pengantongan Urea. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2040211001-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2040211001-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy |
Abstract
PT X merupakan salah satu perusahaan terkemuka dalam produksi pupuk dan bahan kimia yang menghadapi tantangan operasional pada Unit Pengantongan Urea I. Pengantongan Urea I merupakan unit pengantongan yang terdiri dari empat line pengantongan (A, B, C, dan D) yang befungsi untuk mengantongi pupuk Urea. Permasalahan utama yang dihadapi adalah kejadian breakdown yang sering kali menyebabkan tidak tercapainya target pengantongan harian. Kondisi ini berdampak pada penurunan nilai Overall Equipment Effectiveness (OEE) yang merupakan indikator penting efisiensi operasi manufaktur. OEE dihitung berdasarkan elemen Availability, Performance, dan Quality. Untuk mengatasi permasalahan, penelitian ini mengusulkan implementasi metode prediksi menggunakan Recurrent Neural Network (RNN) yang dirancang untuk menganalisis data time-series OEE dan elemen yang ada didalamnya guna memprediksi pola operasional dari Line D di Pengantongan Urea I A. Data OEE dari line D akan diintegrasikan ke dalam sistem untuk menampilkan dataset dan hasil prediksi dari RNN serta dashboard tampilan yang memberikan peringatan terkait waktu, hari, dan tanggal potensi terjadinya breakdown dimasa yang akan datang. Pendekatan ini diharapkan dapat meningkatkan strategi perencanaan pemeliharaan sehingga dapat mengurangi kejadian breakdown serta meningkatkan efektivitas operasional di line D. Hasil dari penelitian ini didapatkan performa terbaik menggunakan model LSTM dengan konfigurasi 2 Layer dan 64 unit untuk tiap-tiap layer dan diperoleh MAPE sebesar 1.68%. Pada prediksi 30 hari kedepan, hasil prediksi menunjukkan terdapat penurunan OEE dengan nilai 23.8% pada 02 April 2025 dan ditampilkan melalui GUI sebagai dashboard grafik serta label peringatan.
====================================================================================================================================
PT X is a leading company in the production of fertilizers and chemicals that faces operational challenges in the Urea Bagging Unit I. This unit comprises four bagging lines (A, B, C, and D) responsible for packaging urea fertilizer. A major issue encountered is the frequent occurrence of breakdowns, which often results in the failure to meet daily bagging targets. This condition negatively impacts the value of Overall Equipment Effectiveness (OEE), a key indicator of
manufacturing operational efficiency. OEE is calculated based on three elements: Availability, Performance, and Quality. To address this problem, this study proposes the implementation of a predictive method using a Recurrent Neural Network (RNN) designed to analyze the time series data of OEE and its components to forecast the operational patterns of Line D in the Urea Bagging Unit I A. The OEE data from Line D is integrated into a system that displays
both the dataset and the RNN prediction results, along with a dashboard that provides alerts regarding the potential breakdowns, including the time, date, and day. This approach is expected to enhance maintenance planning strategies, thereby reducing breakdown occurrences and improving operational effectiveness in Line D. The results of this study indicate that the best performance was achieved using an LSTM model configured with 2 layers and 64 units in each layer, yielding a MAPE of 1.68%. In a 30-day forecast, the prediction revealed a significant OEE drop to 23.8% on April 02 2025, which was presented via a GUI dashboard displaying graphs and warning labels.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | OEE, RNN, LSTM , OEE, RNN, LSTM |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.585 TCP/IP (Computer network protocol) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101 Telecommunication |
| Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
| Depositing User: | Savero Janus Khatulistiwa |
| Date Deposited: | 15 Aug 2025 09:21 |
| Last Modified: | 15 Aug 2025 09:21 |
| URI: | http://repository.its.ac.id/id/eprint/128129 |
Actions (login required)
 |
View Item |